SOIL CHANGE MATTERS 24–27 MARCH 2014 BENDIGO, AUSTRALIA
PAPERS AND ABSTRACTS SOIL CHANGE MATTERS
INTERNATIONAL WORKSHOP 24-27 MARCH 2014, BENDIGO, VICTORIA, AUSTRALIA
This workshop is being hosted by the Victorian Government’s Department of Environment and Primary Industries (DEPI). The organising committee welcomes delegates to this workshop and thanks all those who have responded to the invitation to contribute papers. We have been able to develop what promises to be an interesting and exciting program which we anticipate will generate plenty of good discussion. The theme ‘Soil Change Matters’ reflects the need to monitor and understand the critical processes occurring in soils as a result of agricultural and land management in an era of expanding population and increased pressure on land and water resources.
Local organising committee (DEPI unless otherwise noted): Richard MacEwan (Convenor) Jennifer Alexander Helaina Black (James Hutton Institute, UK) Doug Crawford Phil Dyson (North Central CMA) Jane Fisher Gemma Heemskerk Jonathan Hopley Pauline Mele Rebecca Mitchell David Rees Dugal Wallace Dale Webster
Scientific committee: Mr Richard MacEwan (DEPI, Victoria) Dr Dominique Arrouays (National Institute of Agronomic Research, France) Helaina Black (James Hutton Institute, UK) Mr Doug Crawford (DEPI, Victoria) Dr Ben Marchant (Geoscience, UK) Dr Pauline Mele (DEPI, Victoria) Dr Budiman Minasny (University of Sydney, NSW) Professor Dan Richter (Duke University, USA) Mr Nathan Robinson (DEPI, Victoria)
Acknowledgments The conference organising committee would like to thank DEPI for project support; the organisation of the workshop has been funded under the DEPI project ‘Understanding Soil and Farming Systems’, Cathie Boschert for assisting with the organisation of catering, Michael Adelana for reviewing papers, Grant Boyle for assistance with the field trip and Mark Imhof and Angela Avery for project support. PAGE iii SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
INDEX
Tuesday 25 March 2014 – Soil Matters Symposium 1-49
Soil Change Matters Workshop: Technical Session 1 50-62 Modelling changes in soil carbon
Soil Change Matters Workshop: Technical Session 2 63-92 Monitoring soil condition and land use
Wednesday 26 March 2014 – Soil Change Matters 93-113 Workshop: Policy and Science of Soil Change
Soil Change Matters Workshop: Technical Session 3 114-132 Soil properties and processes
Soil Change Matters Workshop: Technical Session 4 133-143 Land use and soil change
Soil Change Matters Workshop: Technical Session 5 143-153 Modelling soil change
Soil Change Matters Workshop: Technical Session 6 154-160 Mapping soil change
Thursday 27 March 2014 – Soil Change Matters Workshop Plenary: Moving forward - Data and information 161-178 challenges for understanding soil change
Soil Change Matters Workshop Posters – static 178-193
Sponsors page 194 PAGE 1 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
PAPERS AND ABSTRACTS
TUESDAY 25 MARCH 2014 - SOIL MATTERS SYMPOSIUM
Frameworks for soil matters Richard MacEwan1
1 Agriculture Research Division, Department of Environment and Primary Industries, Box 3100, Bendigo Delivery Centre, VIC 3554, [email protected] Introduction
Soil is important to all of us – a silent servant, uncomplainingly delivering services 24-7 – although those services can rapidly diminish or even disappear if we are, from our side, poor servants. How can we provide good service to soil so that, in return, the soil continues to be sustained and provide the services necessary for our survival and environmental health? Knowledge is the key to good management and scientific inquiry is one important path to knowledge (experience is another). Soils are known to be complex and diverse. Understanding how they form and perform requires expertise from all branches of science – chemistry, physics, mathematics, biology – and, within and between each of these disciplines, there are concatenations of detailed specialities, any of which could consume a lifetime of research. So, having frameworks in which to propose and test hypotheses and to order that inquiry is essential. In soil science two frameworks have passed the test of time and served to support education of new soil scientists and to underpin new conceptual models and investigations into soil properties and processes (Jenny 1941, Simonson 1959). A third framework complements these two by providing a system for decision making (Steinitz 1990, 2012).
The frameworks Hans Jenny (1941) 1. Soil (formation) = f {(Climate, Organisms, Relief) x Parent Material} x time Now popularly known as the ‘Clorpt’ equation, Hans Jenny’s formulation of environmental factors (climate, organisms, and relief or landscape position) acting on a parent material over time persists as the ‘101’ model for explaining soil formation and hence soil diversity. Its most recent manifestation has been the adaptation of the principles into ‘SCORPAN’ in which these factors are used to predict spatial relationships in Digital Soil Mapping (McBratney et al 2003). 2. Roy Simonson (1959) Whilst the formulation of Jenny defines the factors of soil formation, Simonson’s paper describes the processes of soil formation in principle as the expression of additions, removals, transfers and transformations. Carl Steinitz (1990, 2012) Coming from the more generalised world of geography and landscape planning, Carl Steinitz proposed a decision making framework for landscape design but which has universal applications. Applied to landscape, Steinitz describes the decision making process as requiring answers to six iteratively linked questions concerning: PAGE 2 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
primary data to represent the landscape, understanding of processes, evaluation of how well things are working, description of scenarios for change, impact analysis of potential change scenarios, and a decision (or management) question.
A synthesis of frameworks How do these three conceptual frameworks encompass the matters of soil and soil change and help us understand these matters and formulate inquiry or develop solutions? The first two frameworks represent a neat analogue for the soil that we see now – the subject of our measurements and management - as a scene in a play (Figure 1). It may be a bit of a ‘whodunnit?’ in which case some forensics are in order, and Jenny and Simonson have set down some principles for us in that investigation of cause and effect. As in watching theatre, we also try to anticipate the plot and where it might be heading. Measurement and monitoring of soil is the key to becoming particularly cluey about where things may be heading with respect to soil change.
PEDON or FACTORS + PROCESSES 3D PROFILE
� Cl imate � Additions � Attributes � Organisms � Removals � Soil type � R elief + � Transfers (classification) � Parent Material � Transformations � Qualities � T ime � Suitability
The Theatre : the The Play : the action A stilled action scene: setting for the play taking place in the theatre a snapshot in time
Figure 1 Soil forming factors and processes interact to leave their imprint on the soil rather like a stilled scene in a play, where the stage is set by the factors and the script is written in the processes. Adding the dimensions of Carl Steinitz’s questions to these two frameworks brings the theory of pedogenesis (soil formation) out from academia and into the world of land management, decisions and policy making. To follow the process advocated by Steinitz we have to assemble data and information, and apply knowledge and understandings to represent the soil and landscape (the stilled action scene in Figure 1). We also have to identify and model the processes that are occurring within the current scene and interpret these for their relative importance. Evaluation of whether things are working well may be an economic question (is the soil supporting a healthy crop without excessive management inputs?), an environmental one (is the soil eroding? Does water readily infiltrate?), or even a social question (what is the best use for this land and its soil? Horticulture? Housing?). Steinitz’s fourth question ‘what might change’ involves some of the big questions of our time around climate change, population growth, economic conditions and technological innovation. Each of these can have a profound impact on soil, the pressures on the landscape and the use of the land – this, the fifth question, is the most challenging level in Steinitz’s framework as we are required to look into the future and make reasonable judgements about the likely impacts if, in the final analysis, we are to answer the sixth question and make sound policies and good decisions about the future use of the soil. PAGE 3 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Conclusion In simplistic terms the broad territory of soil science has been outlined in order to encompass the dimensions of data, information and knowledge that are needed in order to understand soil matters and appreciate and respond to soil change. In this workshop we will hear from many speakers and will explore our knowledge within the dimensions of these three frameworks, particularly in the context of what we know about soil change, how it affects our future and what decisions need to be made in order to improve our knowledge and hence our management of soil. During the sessions over the next three days I hope that we can have our minds on the future with respect to what our science offers, and bring our knowledge to the table as a contribution for better understandings and planning of land use and land management.
References Jenny H. 1941 Factors of soil formation. McGraw-Hill, New York, NY. McBratney AB, Mendonça Santos ML, Minasny B. 2003 On digital soil mapping. Geoderma 117(1–2), 3-52. Simonson RW. 1959 Outline of a generalized theory of soil formation. Soil Science Society of America Proceedings 23:152-156 Steinitz C. 2012 A framework for design, changing geography by design. Esri Press Steinitz C. 1990 A Framework for Theory Applicable to the Education of Landscape Architects (and other Environmental Design Professionals). Landscape Journal, October 1990.
Advances in Soil Biology: What does this mean for assessing soil change? Helaina Black1 and Pauline Mele2
1 The James Hutton Institute, Craigiebuckler, Aberdeen, UK 2 DEPI-Vic AgriBio, 5 Ring Rd Bundoora,Victoria, Australia
Introduction Our interests in soil change are moving away from soil properties and increasingly towards changes in the processes and functioning of soils. Soil organisms are fundamental to dynamics and change in soils through their fundamental role in soil processes (Bardgett et al., 2005). However it is only with recent technical and theoretical advances that we have started to establish quantitative relationships between soil biology and soil change (c.f. Wall and Neilson, 2012). It is this predictive understanding that will enable us to fully integrate soil biology into the effective monitoring and sustainable management of soils. This paper outlines some of the recent advances in soil biology and discusses their relevance to monitoring and management.
Revealing the diversity below our feet We can now ‘see’ the world below our feet in high definition thanks to progressive advances in genetic, imaging and biochemical techniques that are coupled with use of ever more powerful computational techniques. This combination is shedding new light on our understanding of the complexities of the soil ecosystem and its contribution to wealth creation, recreation and environmental sustainability (Ritz et al., 2009). With PAGE 4 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
the development of genetic tools, we can extract DNA and RNA directly from soil and have far less reliance on isolating or culturing soil organisms to identify or count them. The application of genetic tools to soils around the world has demonstrated that soil ecosystems are amongst the most diverse and biologically active in the entire world, as illustrated in Table 1. Estimates of the number of individual bacterial cells in a gram of soil are upwards of 10 billion, more than the number of people on earth (Nature Reviews Editorial, 2011). The field of soil biology is being revolutionised with the rate of discovery of new species at weekly for bacteria and monthly for fungi (NCBI 2014). This rate of discovery is likely to continue as DNA approaches are applied to less well-characterised groups (e.g. protozoa, acari, collembola) and to the multitude of diverse and rare soils and habitats around the world. There are now maps of the soil bacterial taxa of many soils of the world and we know which taxa represent the most abundant and which typically represent the rarer forms (e.g. www.earthmicrobiome.com). This type of information is required for the other soil biota groups and is being undertaken through the Global Soil Biodiversity Initiative (http://www.globalsoilbiodiversity.org). Table 1. Summary of the biological diversity associated with the major soil-dwelling organisms.
Completely Species Species Group/taxon (size) sequenced described estimates genomes
Viruses (nm) 1,832 39541 >>4,000,000
Bacteria (µm) 11,082 45201 >4,000,000
Archaea (µm) 453 2721 >>10,000
Fungi (µm-mm) 24,645 5871 1,500,000
Protozoa (µm-mm) 114,109 12 200,000
Nematodes (µm-mm) 15,000 9633 20,000
Collembola (µm-mm) 6,500 04 15,000
Acari (mm) 20,000 54 80,000
Isoptera (mm) 2,600 04 10,000
Oligochaeta (mm) 3,650 ? 8,000
1 as described in NCBI 2 https://www.sanger.ac.uk/resources/downloads/protozoa 3as described in www.nematodes.org, 4as described in http://www.arthropodgenomes.org/ wiki Our understanding of the diversity of soil biology is rapidly extending beyond this classical taxonomic approach and is increasingly focussed on the characterisation and quantification of the functional diversity of soil organisms. This is not straightforward since one species can have multiple roles across different soil processes (e.g. both the production and oxidation of methane can be carried out by the same soil microbes) and many species or organisms can carry out the same process (e.g. degradation of complex organic compounds). This principle is known as ‘functional redundancy’ and it is this PAGE 5 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
feature that gives soils an enormous capacity to resist, recover and adapt to change. The ability to define and ultimately predict the implications of a change in soil biodiversity for a loss or the improvement in soil processes still remains the “holy grail” in soil biological research. We need to unravel the idiosyncratic relationships between soil biodiversity and function as recently illustrated by Neilson and co-authors (2011). This may require revision of the more classical approach to functional soil biodiversity, which has been heavily reliant on the allocation of conventional knowledge of taxa to the assignment of function. It will require a revision of the role of soil organisms in (bio) geochemical and (bio) physical processes, of the models that represent these processes (e.g. Bradford et al. 2007) and how environmental change alters these dynamics (e.g. Heath et al 2005). The UK NERC Soil Biodiversity and Function Programme, implemented on a Scottish upland farm, developed several methodological approaches to explore these relationships (Bardgett et al., 2005). DNA-based methods continue to develop at a rapid pace to identify and estimate the role of soil organisms in multiple soil processes including the mineralisation of nutrients, the control of pathogens and the remediation of pollutants. DNA cloning techniques have enabled detailed studies on the distribution of important functional groups of soil microbes. In Scotland, these have been used to explore the occurrence and distribution of ammonia oxidising bacteria across different land uses and soils (Yao, et al 2013). More intensive use of DNA approaches takes us into the territory of quantitative PCR (qPCR) (Sharma et al 2007) and microarray methods (He et al 2007) which can show the presence, and estimate the abundance, of a whole range of genes associated with different processes of interest. The application of qPCR methods have been used in Australia to survey the distribution and abundance of genes associated with N-mineralisation and fixation across three major Victorian soil types demonstrates the utility of this approach (Hayden et al 2010). Further technological advances in four areas - metagenomics (Daniel 2005, Delmont et al, 2010, Wooley et al 2010), metatranscriptomics (Carvalhais et al 2012), metabolomics (Simpson and McKelvie 2009) and proteomics (Giagnoni et al 2011) - have opened up exciting new ways to assess the function of soil biology. These methods enable the simultaneous investigation of multiple processes performed by a particular soil biological population at fixed point in time and space. And now we have the ability to link genetic methods with isotopic tools to directly link organism to function (e.g. NanoSIP, Pett-Ridge & Weber, 2011). The increasing challenge is synthesising and interpreting the immense amount of data that these techniques generate.
Soil biology – what’s where and why? The structure and activity of soil biological communities is reflected in their biogeography. A key challenge is to be able to predict why particular organisms and functional groups of organisms occur, or do not occur, in particular locations. This is the foundation of being able to utilise soil biology in monitoring or management. In the case of bacteria, it has been proposed that this biogeography is controlled primarily by the edaphic features of an ecosystem (Fierer and Jackson, 2006). This means that everything is not everywhere but instead is defined by the physico-chemical features across a range of scales from the soil pore/aggregate (µm of) the landscape (km). In Australia, there is significant investment in measuring the soil microbial diversity associated with regional grain production systems (e.g. www.soilquality.org.au) and more broadly across the major terrestrial biomes (e.g. www.bioplatforms.com.au/special-initiatives/environment/ soil-biodiversity). Preliminary data from Australian soil biodiversity surveys suggests strong correlations between soil type, regional rainfall distribution patterns and land-use. A PAGE 6 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
regulator of soil community diversity is soil pH. For example, we see higher abundance of the bacterial phylum Acidobacteria in the low pH Ferrosols compared to the high pH Calcarosols of the Mallee (Mele et al 2008). This agrees with trends observed in soil biodiversity surveys elsewhere (Fierer and Jackson, 2006). In the UK, there has been a similar investment in exploring the biogeography of soil biology. The Defra funded SQID project investigated the structure and function of soil organisms in typical land uses across the British mainland (Black et al., 2011). This study demonstrated the overwhelming influence of land use on the structure and activity of soil biota (other than soil bacteria) with an underlying influence of other factors including soil chemistry and plant community structure. This research has moved on to establishing characteristic soil biological communities for different land uses, to establish a monitoring framework, and in quantifying the impact of drivers of change such as nitrogen deposition from point sources (e.g. Mitchell et al., 2013) and management systems (Vink et al., 2014) as a precursor to defining indicators of change. Thus, with new techniques available for assessing soil biodiversity, questions related to the impacts of land use, management, pollution and climate change can be assessed. Given that biodiversity expresses a biogeography, it is becoming apparent that change in soil biodiversity needs to be measured and interpreted within a regional context to ensure that pro-active management, remedial or protective actions are relevant and hence effective. What these studies have demonstrated is that soil biology cannot be investigated in isolation of other soil properties, soil processes and wider environmental factors.
Why bother with soil biology? With increasing emphasis on recognising and quantifying the value of soils to society (Costanza et al 1997), the “value” of soil biology has been considered in various ways. Ultimately, it is recognised that soil biology, with its vital role in key soil processes, is valuable to food, water and energy security. All are reliant upon the role of soil biology in nutrient release, pest and disease regulation and retention and in the stability / erodibility of soils, amongst other key ecosystem services. Soil biology also has a vital role in regulating our climate, as it is the key in the turnover of soil organic matter and release of greenhouse gases (Mele, 2011). Our habitats that are valued for their conservation status often rely upon soil organisms (e.g. symbionts) for their characteristic flora and fauna while many important pharmaceutical products have been developed from soil derived biological compounds. However can we really put an economic value on soil biology? In 1997, Pimentel and co-authors estimated that the global value of soil biodiversity exceeded 1.5 trillion dollars. Although this figure demonstrates the importance of soil biology to a wide range of ecosystem services, the economic values are difficult to relate to management within specific regions or for particular land uses. At a national level, Glenk and colleagues (2010) explored the value of soil biology to the Scottish economy. Ultimately, this study demonstrated that the ability to value the contribution of soil biology to food, water, energy, conservation and other services is severely limited by local scale data and defined relationships between soil biology and these services. There are a few examples globally where we can truly appreciate and realise the value of soil biology, which can be illustrated by examples from Australia. Here, soil is viewed as a resource for sustaining and improving agricultural production and profitability. There are two major drivers for investment in Soil Biology. The first relates to the mounting concern regarding the continued degradation of the soil resource (State of the Environment and National Soils RDE Strategy) and the second the escalating input costs of production and the perception that soil biology can offset these. Investment in soil biology is largely industry based with $45M invested by the Grains Research and PAGE 7 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Development Corporation (GRDC) since 1992 (Mele, 2009). This has served to focus research activity on applied outcomes related to demonstrating and quantifying the agronomic significance and developing management strategies (and products) that promote and enhance beneficial processes for productivity gains. Increasing energy costs that are driving up input (fertiliser N/P, lime, biocides) costs, and a general concern that soils are degrading further with increasing levels of salinity, acidity and disease and declining or low levels of carbon are considered by a growing stakeholder group to represent the major constraints to sustainable profitability (Campbell, 2008, Sanderman et al., 2010). The Grains Research and Development Corporation (GRDC) is the major funder of soil biology research in Australia investing more than $16M in two initiatives from 2002-2007 and 2009-2014. An economic evaluation based on judgements of the possible levels of adoption and benefits in areas relating to nutrient balance, disease control (including inoculants) and inoculants (yield increasing) indicate a net present value of $32 million and a benefit/cost ratio of the order of 4. The time lag from 2002 to the break-even point is of the order of 20 years. Sensitivity analyses aiming to capture the possible range of adoption and benefits indicated a range from break-even to a net present (2009) value of $120 million (Agtrans Research, 2009). The time lapse between the investment and the return, coupled with the considerable effort in integrating information into a regional context represent the two major challenges for ongoing investment in this area.
Where are we at now and what are some of the remaining challenges? There has been a revolution in knowledge regarding soil biodiversity with the development of molecular and imaging techniques. An analysis of the history of technology shows that technological change is exponential. So, at the present rate, in the context of soil biology, we are on the upslope and will experience something like 100’s of years of progress in 10 year (http://www.kurzweilai.net, 2001). Keeping the speed of advances in mind, the critical features of studying and, ultimately, utilising soil biology must be that they have a focus (a question), a scale (a time frame and spatial context), a multi-disciplinary approach and a pathway to adoption that is meaningful to multiple end-users. It is crucial that soil biology is integrated fully with soil assessments to gain a holistic appreciation of soil change in both space and time. Examples of some big (and difficult) questions are those posed by Handelsman and co- workers (Handelsman, 2007) for microbial communities which can be expanded to soil biology in general: ›› How resilient are soil biological communities in the face of rapid global change? ›› Can soil biological communities help to buffer and mediate key elemental cycles which are now undergoing rapid shifts? ›› Can changes in soil biological communities serve as sensors and early-alarm systems of environmental change? ›› To what extent can we manage soil biological communities to modulate the effects of human activities on natural elemental cycles sensibly and deliberately? We are advancing slowly towards some answers with massive genetic datasets being generated and analysed that will fill some of the knowledge gaps in defining species and function of soil biology across the soils of the world. There are also many outstanding challenges. A significant technical challenge is requirement for sophisticated informatics approaches that can process enormous amounts of data and statistical techniques PAGE 8 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
to explore the non-linearity in relationships between soil biology and its environment. Similarly a significant social challenge is transforming the way we think about and investigate soil biology. This requires a greater attention to systems and non-linear thinking. In practical terms, this means increasing use of metadata, often described as the ‘halo of data’, which includes soil chemical, physical and historical agronomic/ land-use data that enables an accurate description of the soil as a habitat for biological functions. This, in turn, requires cooperation across a range of skill-sets from the soil biologist, soil scientists to experiential skills of land-holders. A noteworthy model of such a coordinated systems-based approach is provided by the human microbiome project (Turnbaugh et al 2007) which shares a central tenet with soil biodiversity; the absolute reliance on a diverse microbial community to sustain a healthy system, whether a human body or a soil. No doubt these are exciting times for mapping the biogeography of soil biology and the changes in functionality as a result of human interventions. This will not only challenge some long held ecological concepts borrowed from above-ground systems but will generate indisputable evidence of soil biology as an asset worth protecting.
References Agtrans Research (2009). Impact Assessment: An Economic Analysis of Investment in the GRDC Soil Biology Program (Themes 1-3) pp 1-35 Bardgett, RD, Usher, MB, Hopkins DW (2005) Biological diversity and function in soils. Cambridge University Press. Black HIJ, Ritz, K, Harris, JA, Cameron, CM, Campbell, CD, Chamberlain, PM Creamer, R Pawlett, M Wood, C Singh, BK (2011). Scoping Biological Indicators of Soil Quality Phase II. Defra Final Contract Report SP0534. (http://randd.defra.gov.uk) Bradford, MA, Tordoff, GM, Black, HIJ, Cook, R, Eggers, T, Garnett, MH, Grayston, SJ, Hutcheson, KA, Ineson, P, Newington, JE, Ostle, N, Sleep, D, Stott, A and Jones, TH. (2007). Carbon dynamics in model grassland with functionally different soil communities. Functional Ecology 21, 690-697 Campbell A (2008). Managing Australia’s Soils: A Policy Discussion paper. Canberra 83. Carvalhais, LC, Dennis, PG, Tyson, GW, Schenk, PM (2012) Application of metatranscriptomics to soil environments. Journal Of Microbiological Methods 9 (2), 246-251 Costanza, R, d’Arge, R, de Groot, R, Farber, S, Grasso, M, Hannon, B, Limburg, K, Naeem, S, O’Neill, RV, Pareulo, J, Raskin, RG, Sutton, P and van den Belt, M (1997). The value of the world’s ecosystem services and natural capital. Nature 387, 253-26 Daniel, R (2005). The metagenomics of soil. Nature Reviews 3, 470-478 Delmont, TO, Robe, P, Cecillon, S, Clark, IM, Constancias, F, Simonet, P, Hirsch, PR, Vogel, TM (2011). Accessing the Soil Metagenome for Studies of Soil Diversity. Applied and Environmental Microbiology 77 (4), 1315-1324 Fierer, N, and Jackson, RB. (2006). The diversity and biogeography of soil bacterial communities. PNAS 103 (3), 626-631 Giagnoni,L, Magherini, F, Landi, L, Taghavi, S, Modesti, A, Bini, L, Nannipieri, P, Van der Lelie, D. and Renella, G (2011). Extraction of microbial proteome from soil: potential and limitations assessed through a model study. European Journal of Soil Science 62, 74-81 PAGE 9 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Glenk, K, McVittie, A, Towers, W, Watson, C and Black HIJ (2010) Socio-economic data on Scottish soils – collection and development. Scottish Environment Protection Agency Commissioned Report. [online] www.sepa.org.uk/land/idoc.ashx?docid=ebebd86e-4af1- 4a48-b2e4-952c534fce8e&version=-1 Handelsman, J , Tiedje, J, Alvarez-Cohen, L, Ashburner, M, Cann, IK, DeLong, EF, Doolittle, WF, Fraser-Liggett CM, Godzik,A, Gordon, JL, Riley, M and Schmid, MB (2007) The New Science of Metagenomics: Revealing the Secrets of our microbial planet. The National Academies Press Washington, DC. Hayden,HL, Drake, J, Imhof, M, Oxley, AP, Norng, S, and Mele, PM .(2010). The abundance of nitrogen cycle genes (amoA and nifH) depends on land-use and soil type in South-Eastern Australia. Soil Biology and Biochemistry 42, 1774-1783 He, Z, Gentry, TJ, Schadt, CW, Wu, L, Liebich, J, Chong, SC, Huang, Z, Wu, W, Gu, B, Jardine, P, Criddle C, and Zhou, J (2007). GeoChip: a comprehensive microarray for investigating biogeochemical, ecological and environmental processes. The ISME Journal 1, 67–77; doi:10.1038/ismej.2007.2 Heath, J, Ayres, E, Possell, M, Bardgett, RD, Black, HIJ, Grant, H, Ineson, P, Kerstiens, G (2005) Rising atmospheric CO2 reduces sequestration of root-derived soil carbon. Science 309 (5741), 1711-1713 Mele, PM (2009). A New Charter for Investing in Soil Biology RDE in Australia. Occasional Publication. Land and Water Australia. Canberra. Mele, PM (2011). Soil health, soil biota and climate change. Chapter 8. In ‘Soil Health and Climate Change’. Editors: Bhupinder P Singh, Annette L Cowie & K Yin Chan. Soil Biology Series (Springer, Amsterdam). Mele, P, Hayden, H, Methe, B, Stockwell, T, Pfannkoch, C, Lewis, M, Tanenbaum, D, Rusch, D, Heidelberg, K (2008). Assessing soil microbial communities using a metagenomic approach. 12th International Symposium on Microbial Ecology, Cairns Australia. Mitchell, RJ, Beesley, L, Briggs, R, Cuthbert, A, Dawson, JJC, Helliwell, RC, Hewison, RL, Kerr, C, Leith, ID, Owen, IJ, Potts, JM, Newman, G, Smith, D, Shephard, LJ, Sturgeon, F, Taylor, AFS, White, D, Williams, A, Williams, E, Black, HIJ (2013). Testing soil quality indicators. Scottish Environment Protection Agency Commissioned Report No.HP1108. Nature Reviews Microbiology, Editorial 9, 628 (September 2011) doi:10.1038/ nrmicro2644 NCBI (2014) data from www.ncbi.nlm.nih.gov Neilson, UN, Ayers, E, Wall, DH, Bardgett, RD (2011) Soil biodiversity and carbon cycling: a review and synthesis of studies examining diversity-function relationships. European Journal of Soil Science 62, 105-116 Pett-Ridge, J and PK Weber (2011) NanoSIP: NanoSIMS applications for microbial biology. Ed Navid A. Microb Microbial Systems Biology: Methods and Protocols. Methods in Molecular Biology 881. Springer Ritz K, Black HIJ, Campbell CD, Harris JA, Wood C (2009) Selecting biological indicators for monitoring soils: a framework for balancing scientific and technical opinion to assist policy development. Ecol Indic 9, 1212–1221 PAGE 10 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Sanderman J, Farquharson R, Baldock J (2010). Soil Carbon Sequestration Potential: A review for Australian agriculture CSIRO Land and Water Canberra. Schloss, PD and Handelsman, J (2003). Biotechnological prospects from metagenomics. Curr Opin Biotech.14: 303-310 Sharma, S, Radl, V, Hai, B, Kloos, K, Fuka, MM, Engel, M, Schauss, K, Schloter, M (2007) Quantification of functional genes from procaryotes in soil by PCR. Journal of Microbiological Methods 68, 445–452 Simpson, MJ and McKelvie, JR (2009) Anal Bioanal Chem. 394, 137-149 Turnbaugh, PJ, Ley, RE and Gordon, JI (2007). The human microbiome project: exploring the microbial part of ourselves in a changing world. Nature 18: 804-810. Vaccari, D (2009). Phosphorus Famine: The Threat to Our Food Supply. Scientific American Magazine. Vink, SN, Neilson R., Robinson D and Daniell TJ (2014) Temporal and land use effects on soil bacterial community structure of the machair, an EU Habitats Directive Annex I low- input agricultural system. Applied Soil Ecology 73, 116-123 Wall, DH and Nielsen, UN (2012) Biodiversity and Ecosystem Services: Is it the Same Below Ground? Nature Education Knowledge 3(12), 8 Wooley, JC, Godzik, A, Friedberg, I (2010). A Primer on Metagenomics. PLoS Computational Biology 6 (2), 1-13 Yao, H, Campbell, CD, Chapman, SJ, Freitag, TE, Nicol, GW, Singh, BK (2013) Multi- factorial drivers of ammonia oxidizer communities: evidence from a national soil survey. Environ Microbiol 15 (9), 1462-2920 PAGE 11 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The Soil and its Chemistry- Critical Futures Mike J. McLaughlin1
1 CSIRO Land and Water/University of Adelaide, Waite Campus, PMB 2, Glen Osmond, SA 5064, Australia
Abstract Unlike changes over time to the earth’s atmosphere or oceans, changes to soils occur at a very local scale, with much less interconnectedness between the points of change. A further difference between soils and other media is the much greater buffer capacity of the system to change, which means that changes are more gradual and difficult to measure. Further difficulties of assessing changes in soil “condition” are due to the natural heterogeneity of soils across landscapes, which poses sampling, analysis and statistical interpretation challenges of a much greater magnitude than those for the atmosphere or for surface or ground waters. Superimpose localised anthropogenic disturbances and enhancements to soil on this (at a farm or paddock level) and it is evident that knowledge and assessment of soil “condition” is needed at a relatively small scale compared to assessments of atmospheric or aquatic system “condition”. At the same time, soils contribute significantly at larger scales in modifying (in both good and bad ways) to the quality of water systems, to emissions of gases (greenhouse and others) to the atmosphere, and to feeding a steadily growing human population around the globe – so changes to soil really do matter. This juxtaposition of scale issues means that to understand the global impact of changes to soil “condition”, we need to measure changes at small scales and integrate the mosaic of effects up to larger scales i.e. think globally, measure locally.
Rates of soil change Changes to the chemistry of our soils have accelerated rapidly in the last 250 years due to the “industrialisation” of agriculture where inputs of chemicals and fossil fuel energy (in the form of mechanised equipment) has allowed deforestation, cultivation, fertilisation and farm chemical inputs on a much larger scale than observed in previous millennia – a good example for changes in soil organic carbon content over time was provided recently by Richter and Yaalon (2012) (Figure 1).
Figure 1. PAGE 12 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 1. Rates of change in soil organic carbon in response to cultivation, manure amendments, reforestation, and other practices contrast with rates of change in soils that are products of natural soil formation. Plotted are average rates of change over the indicated age of soil or soil-management regime (from Richter and Yaalon, 2012). Similarly here in Australia, we have seen dramatic changes over time in the chemistry of our soils due to the introduction of phosphatic fertilisers and legume-based pastures (Williams and Donald, 1957) (Figure 2) – and note these changes in soil properties have been predominantly positive in contrast to some negative yet beliefs about fertiliser use and soil health not backed up by good scientific data. Nonetheless, changes to soil chemical properties are generally slow gradual changes, due to the inherent inertia of the soil to additions or removal of elements, or to changes in composition in terms of minerals and organic matter – what soil scientists usually term the “buffer capacity”.
Figure 2. Relationship between amounts of single superphosphate applied to six properties and soil concentrations of nitrogen (from Williams and Donald, 1957). Because of the slow nature of change, it is imperative that measurement technologies are both accurate and precise (discussed further in the presentation), and that we have “sentinel” sites that alert us to adverse changes – in soil science the sentinels are often long-term field experiments (e.g. Rothamsted, Morrow plots, Waite Long Term Rotation Experiment) that provide invaluable information on how management practices affect crop production, soil properties, water quality and emissions of gases to the atmosphere (Korschens, 2006).
Scale of changes in soil properties As noted in Figure 1, use of soils in agricultural systems has in many cases increased the heterogeneity of soil properties across the landscape, and inference of soil condition from the pedogenic background state is becoming less relevant with time. In addition, pollution of soils from industrial or urban activities is also extremely localised, so that assessment of soil contamination (and remediation) is often confounded by variability of properties and analytes across quite small scales. PAGE 13 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Traditional assessments of soil chemical “condition” have usually been by field sampling (at various intensities dependent more often on factors such as time, cost and enthusiasm rather than actual heterogeneity) and laboratory analysis (with the suite of analytes again more often dependent on time, cost and enthusiasm than anything else). Cost is usually the critical factor which drives the adoption and use of soil analysis, so that multi-analyte tests are now much preferred in many countries e.g. the Mehlich soil test widely used in USA (Mehlich, 1984). Unfortunately the dominance of the cost driver in soil analysis means that the adoption of multi-analyte tests is often driven more by laboratory expediency than by predictive power. New technologies have emerged in the last 20 years which have revolutionised assessment of soil condition, which allow much faster analysis of soils at much lower cost, and which can be placed in the hands of land managers – direct spectroscopic methods which are field portable. Good current examples of these are near- and mid- infrared soil spectrometers, as well as laser induced breakdown spectrometers, portable x-ray fluorescence spectrometers and laser-induced fluorescence instrumentation. It is important these powerful multi-analyte techniques are well calibrated and validated before adoption but I believe rapid, direct and inexpensive spectroscopic analysis of soils is key to assessing and managing soils at the local scale, where all the “action” is.
References Korschens, M. (2006) The importance of long-term field experiments for soil science and environmental research - a review, Plant, Soil and Environment, 52, 1-8. Mehlich, A. (1984) Mehlich-3 soil test extractant - a modification of Mehlich-2 extractant, Comm. Soil Sci. Plant Anal., 15, 1409-1416. Richter, D.d. and Yaalon, D.H. (2012) “The Changing Model of Soil” revisited, Soil Sci Soc Am J, 76, 766-778. Williams, C.H. and Donald, C.M. (1957) Changes in organic matter and pH in a podzolic soil as influenced by subterranean clover and superphosphate, Aust. J. Agric. Res., 8, 179- 189.
The Soil habitat: microbiology in a structured world Iain M Young1
1 Iain Young
This talk aims to illustrate how the physics of soil controls the spread and activity of the soil microbial population. The geometry of the soil pores (how connected and how tortuous they are) controls resource flows to and from bacteria; so, structure controls how quickly oxygen moves to an aerobic microbe, and how quickly CO2 moves away. The structure also controls the distribution and connectivity of the water in soil through the moisture release curve. Also, whilst it is clear that billions of microbes exist in a handful of fertile soil, most of the surface of the inner space of soil is like a desert - devoid of life. This is due to the large surface areas that exist in most soils – as clay content increases soil does surface area. Isolation (patchiness) of microbial populations is a fact in soil. Finally, the talk will show how that same microbial population can influence its own environment by acting as interior designers and plumbers, controlling pores pathways and water. PAGE 14 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil Carbon Past and Future Jeff Baldock1
1 CSIRO Land and Water/Sustainable Agriculture Flagship, PMB 2, Glen Osmond, SA 5064, Email: [email protected] Abstract The quantity of organic carbon found in Australian soils and how this can be altered by land management practices have been the topic of much discussion and research. The reason for such interest relates to the positive contributions to soil productivity and resilience that organic carbon and its associated elements (N,P, S, O and H) can make, and to the greenhouse gas mitigation potential associated with increasing and maintaining stocks of soil organic carbon. However, broadly applicable quantitative relationships defining the contributions of organic carbon to soil productivity and resilience remain elusive and debate continues over the quantity of additional organic carbon can be retained in Australian soils. Soil organic carbon is composed of a variety of materials differing in extent of decomposition, chemical composition, particle size, and level of interaction with soil minerals. As a result, the various components of organic carbon will each contribute differently to soil productivity and resilience and carbon sequestration. Gaining an understanding of this diversity and developing a capability to measure and model it across the range of environments and soils that exist across Australia, is required to progress our understanding and develop robust predictive models. The recent Soil Carbon Research Program (SCaRP) conducted a baseline assessment of the stocks of soil organic carbon and its allocation to component fractions across many of Australia’s agricultural regions. SCaRP also established the potential of mid- infrared (MIR) spectroscopy to serve as a rapid and cost effective analytical tool for quantifying soil carbon content and composition. Although stocks of soil organic carbon content were often correlated with a variety of environmental and soil properties thought to be important to defining the amount of carbon present in a soil, significant levels of variation remained unexplained. The high variability of soil carbon stocks within defined management regimes, while indicating that a range of soil carbon sequestration outcomes where possible, made it difficult to detect statistically significant management effects. Moving forward, a nationally consistent application of MIR to soils is required and would result in great gains in the efficiency of calibration and prediction. Further work on the allocation of carbon to biologically significant components that differentially impact on soil productivity and carbon sequestration potential remains an ongoing priority. Development of a program for monitoring changes in soil carbon stocks through time will enhance the value associated with the SCaRP and other soil collection activities and start to provide the temporal data required for model calibration and validation. Additionally, a reassessment of the idea that broad statements can be made on carbon sequestration potential based on general classes of agricultural management (e.g. reduced tillage or pasture grazing management) is required. Given that soil carbon stocks are controlled by the balance between inputs and losses of organic carbon and the wide variations in production outcomes within defined management regimes, emphasis may be better placed on quantifying the amount of carbon added to soil irrespective of the management regime. PAGE 15 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The Cost Effectiveness of a Policy to Store Carbon in Australian Agricultural Soils to Abate Greenhouse Gas Emissions R. E. White1 and B. Davidson1
1 Department of Agriculture and Food Systems, Melbourne School of Land and Environment, The University of Melbourne, Parkville, Victoria 3010. Email [email protected]
Abstract Data for cropping and pastoral enterprises in southeastern Australia were used in a cost- effectiveness analysis to assess the feasibility of abating greenhouse gas (GHG) emissions through storing soil carbon (C). We used a C credit value of $24.15 per tonne (t) of CO2- equivalent and a C storage rate of 0.3 t C/ha/year. Given that a change of enterprise is driven primarily by farmer returns, we found that none of the changes were feasible, with the exception of wheat to cattle or sheep. However, assuming a modest adoption rate by wheat farmers of 10 percent, only 0.3 percent of current national GHG emissions would be abated.
Introduction The Australian Government has a stated objective of decreasing Australia’s greenhouse gas (GHG) emissions by 5 percent by the year 2020. Through a Direct Action Policy, the details of which are yet to be formulated, the government’s objective will be achieved by “abatement .... purchased via a market mechanism to achieve the lowest price....The lowest cost abatement may be a mix of energy efficiency, cleaning up waste coal mine gas, cleaning up power stations and landfill gas. It may be reafforestation of marginal lands or revegetation or improvement of soil carbon” (Hunt 2013). As yet, the Department of the Environment cannot say what the ‘mix’ of practices to reduce GHG emissions will be, and how much reliance will be placed on soil carbon (C) storage. However, a consortium of companies called BIO CCS claims Australia’s annual GHG emissions could be reduced by 150 million (M) tonnes of CO2-equivalent (CO2-e) by 2020, of which 45 Mt would be achieved through soil C storage, primarily by adopting ‘biological farming systems’ (BFS) (Anon. 2009). The BFS rely on a combination of increased plant production and enhanced activity of the soil biota to produce stable soil organic carbon (SOC). Elsewhere on this website it is claimed that 25 percent of Australia’s GHG per annum (approximately 138 Mt) could be abated by BFS practices applied to only 12.5 Mha of Australia’s intensively managed cropland. These ambitious claims are contrary to reviews of Australian and overseas literature showing that there were few examples of sustainable, significant increases in SOC being achieved under dryland farming (White 2012, Lam et al. 2013). For example, White (2012) concluded that the potential GHG abatement from soil C storage under average management across 470 Mha of intensively and extensively managed land was about 12 percent (68 Mt CO2-e) of 2011-12 national emissions of 552 Mt CO2-e. However, at the time the Department of Climate Change and Energy Efficiency’s (DCCEE) own projection for abatement from soil C sequestration in 2020 was 0.5-1.5 Mt CO2-e, amounting to only 0.1-0.3 percent of current emissions (White 2012). Clearly, the DCEE did not expect a high adoption rate by landholders of soil C storage for GHG abatement, which could be attributed to several factors, such as the difficulty of measuring and verifying changes in SOC; the cost of compliance with the rigorous PAGE 16 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
conditions of the Carbon Farming Initiative; the reluctance of farmers to change a profitable farming enterprise to one that was less profitable, or required new skills, simply to earn C credits; the relatively low value of Australian C credits, and the even lower value of C credits in European Union Emissions Trading Scheme, to which the Australian scheme was to be linked. Given the wide disparity between estimates of the physical potential for soil C storage and the expected contribution this might make to abating national GHG emissions, the aim of this paper is to analyze the economic consequences for farmers who change their current farming systems to other systems for the purpose of storing soil C under a government scheme. A range of common farming systems and their C storage potentials is considered, as are the changes to farmers’ gross margins, the costs of compliance with government policy, and the current C credit value available. Two measures are used to assess the viability of the program - the cost effectiveness per ha of making a change (a measure of the incentives farmers might need to partake), and the cost effectiveness per tonne of C (a measure of what the government might expect from the program). These measures are calculated for the short and long term (1 and 25 years, respectively).
Methods Cost-effectiveness analysis is a technique that assists in ensuring the efficient use of investment resources in sectors where benefits may be difficult to value. The technique can be used to select amongst alternative projects or practices which have the same objectives (quantified in physical terms). The approach can be used to identify the alternative that minimizes the actual value of costs to provide a given level of output; or conversely, the value of costs that maximizes the output level. The result of the analysis is a cost-effectiveness ratio of outputs to costs that can be used to measure the impact of a policy. Given that Sanderman et al. (2010) concluded the best means of increasing SOC by some 0.3-0.6 t C/ha/yr was by converting arable land to permanent pasture, this study is confined to the cost effectiveness of changing from a cropping to livestock enterprise. The ‘effect’ measured is the amount of C stored per ha. For simplicity, the options available to farmers to increase SOC in southern New South Wales are assessed and assumed to be representative of other farming regions in southeastern Australia. The costs are those associated with making a change from one enterprise to another and those of complying with the policy. It is assumed that the effect, measured over time, is the difference between the increased SOC under the new enterprise compared with the SOC under ‘business as usual’. This effect will often be larger than the difference calculated between SOC under the new enterprise and SOC at time zero (the baseline). The costs of changing enterprises are simplified by assuming they are confined to a change in the gross margin a farmer receives from his enterprise. This assumes that the capital costs associated with the change are non-existent. Table 1 gives data on gross margins for various cropping and livestock enterprises. The costs of compliance depend to a large extent on measuring and verifying the change in SOC. One commercial soil testing laboratory in Sydney put the measurement cost at $140.00 per sample. For a 100 ha paddock, a minimum set of 10 soil samples would be required, each to be taken at the beginning and end of an accrediting period (say 7 years). So the total measurement cost would be $28 per ha, discounted over 7 years. This is a conservative estimate and does not include any costs associated with auditing and reporting. PAGE 17 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
These costs and their effects need to be measured against the benefits to a farmer for undertaking the change. The value of a C credit was taken as the Australian value of $24.15 per t CO2-e for 2013-14, which was used in two ways. First, it is incorporated into the farm-level analysis as a benefit to farmers from undertaking the change, and so is subtracted from the costs of change and compliance. Although it is assumed the C credit is paid for every year of the analysis, it must be prorated up as the difference in SOC accumulated increases over time. Since the government intends to introduce a “25- year option for land based sequestration” (Hunt 2013), this is an appropriate time frame for the analysis. Second, the C credit value can be used to assess the whole policy, when it becomes the basic price the government has to pay to implement soil C sequestration as part of the policy. In this way, the value of a C credit necessary to achieve a desired level of abatement can be determined. This then determines the national cost of the policy, which can be compared with the sum of money the government is prepared to allocate to the policy.
Results and Discussion The results of the analysis presented in Table 2 have three main components. First, there is the distinction between changing enterprises from one of four main crops to either sheep or cattle. Second, there is the difference between the cost of saving a tonne of C to society and the cost to the farmer of saving C per ha. Third, there is a distinction between the short-term annual costs and the long-term costs (over 25 years discounted using net present value (NPV) techniques at 7% per yr). Table 1. Gross margins for wheat, soybeans, lucerne, sheep and cattle enterprises in southern NSW
Enterprise Units Soybeans Maize Wheat Lucerne Sheep Cattle Farm price $/t 625 290 200 310 - - Yield t/ha 4 11 3.5 15 10 2.6 Gross revenue $/ha 2500 3190 700 4650 1116 1497 Gross Costs $/ha 622 1396 442 2249 609 594 Gross Margin $/ha 1878 1794 257 2401 507 903
Notes: The sheep enterprise is assumed to be a 1000 ewe flock run at 10 dse/ha, whereas the cattle herd is assumed to be a 100 head mob, run at 2.6 lsu/ha. Source: NSW Department of Primary Industries (2012). PAGE 18 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 2. Cost effectiveness (per t C and per ha) of changing from a cropping to a livestock enterprise to store soil C (assuming an average soil C storage of 0.3 t C/ha/year)
Item Units Soybeans Maize Wheat Lucerne To society Annual – to cattle $/t C 309 284 -177 466 – to sheep $/t C 428 402 -58 584 Long term (25 years) – to cattle $/t C 3559 3266 -2104 5388 – to sheep $/t C 4944 4651 -719 6773
To the farmer Annual – to cattle $/ha 976 892 -644 1499 – to sheep $/ha 1372 1288 -247 1896 Long term (25 years) – to cattle $/ha 11511 10532 -7367 17606 – to sheep $/ha 16128 15149 -2751 22223
Note that the adjustment costs of making a change between enterprises are assumed to be zero. The main points to note are (a) with the exception of wheat, the costs per t of C stored are between $284 and $584 per year (includes the C credit). (b) Again, with the exception of wheat, in the long run these costs accumulate and in NPV terms amount to between $3266 and $6773 per t C stored. (c) Farmers need an incentive to change enterprises and, with the exception of wheat, the costs of change are between $892 and $1896 per ha per year. In the long run, these costs mount substantially to between $10532 and $22223 per ha. (d) The negative results for wheat can be interpreted as the opposite to a cost; in other words these are the benefits of changing. A change from wheat to cattle is estimated to yield a benefit of $177 per t C stored in one year and $2104 per t in the long run. On a per ha basis, farmers would improve earnings by an estimated $644 in the short run and by $7367 over 25 years. The benefits are less appealing in a move to sheep. (e) The question arises why farmers do not abandon wheat and take up cattle? Regardless of the benefits of C storage, this would appear to be a profitable move. However, a shortcoming of this analysis is that the costs of making the change are not incorporated and these would need to be less than $644 per ha in the first year and less than $7367 per ha over 25 years. (f) Although changing from a cropping enterprise to sheep appears less cost effective than a change to cattle, the adjustment cost of making this change may be less. However, grazing cattle or sheep will add to GHG emissions because of methane release. PAGE 19 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
A sensitivity test was conducted and it was concluded that changes of ±50 percent in prices for the products would have a significant impact on the final results, but conversely changing the rate of C storage and the C credit value have little impact on the final results.
Conclusions These results emphasize that enterprise choice is driven primarily by returns to the farmer. Thus, unless an incentive to change enterprises (such as a C credit) is very large, change will not occur. This analysis shows the best possible outcome and all other real situations present a worse-case scenario. With the exception of wheat production, changing to cattle or sheep to store soil C will not be a cost effective way of abating GHG emissions. With a current wheat price of $200 per t, a change to cattle or sheep could be profitable provided the initial adjustment cost of changing is less than $644 per ha. Assuming, unrealistically, that change occurred over the whole 13.9 Mha of Australian wheat land, the potential for soil C storage would be 15.3 Mt CO2-e per year. However, a more likely adoption rate of 10 percent offers potential storage of 1.53 Mt CO2-e per year, costing the government $37 million at a C price of $24.15 per t CO2-e. This storage figure, amounting to only 0.28 percent of current emissions, is very close to the former Department of Climate Change and Energy Efficiency’s projection of 0.1-0.3 percent abatement by soil C storage in 2020. Overall, C storage in agricultural soils is not a cost effective option for abating Australia’s GHG emissions.
References Anon. (2009) Building soil organic carbon using biological farming systems in Australia’s more intensive agricultural regions. www.sba.asn.au/sba/bioccs.asp (accessed 25 October 2013) Hunt G. (2013) Choosing the right market mechanisms for addressing environmental problems. www.greghunt.com.au (accessed 25 October 2013) Lam SK, Chen D. Mosier AR, Roush R (2013) The potential for carbon sequestration in Australian agricultural soils is technically and economically limited. Scientific Reports 3, 2179, 1-5. NSW Department of Primary Industries (2012) Farm budgets and costs. www.dpi.nsw. gov.au/agriculture/farm-business/budgets (accessed 1 October 2013) Sanderman J, Farquharson R, Baldock J (2010) Soil carbon sequestration potential: A review for Australian agriculture. A report prepared for Department of Climate Change and Energy Efficiency, Australian Government, Canberra. White RE (2012) ‘Tis an ill wind that blows nobody any good. In: ‘Soil solutions for diverse landscapes’. 5th Joint SSA and NZSSS National Conference, Hobart, 2-7 December. PAGE 20 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Remote sensing, productivity and soil condition Elizabeth Morse-McNabb1
1 Agriculture Research Division, Department of Environment and Primary Industries, Bendigo, Victoria, Australia. [email protected]
Abstract Remote sensing, productivity and soil condition are all current areas of focus in DEPI Agriculture Research. There are a plethora of sources of remotely sensed information available from free broad scale satellite imagery to expensive small scale unmanned aerial vehicle (UAV) imagery. Can each of these sources of imagery provide an understanding of agricultural productivity and soil condition? What kind of analysis and tools are required to extract information and where do we think this research is heading?
Application of remote sensing in precision agriculture Andrew Whitlock Precision Agriculture Consultant, PrecisionAgriculture.com.au, 7 Learmonth Rd, Wendouree, Victoria, 3355, [email protected], 0458 312 589
Abstract Measuring spatial variability across farms underpins the management techniques of modern farming systems. Site-specific crop and pasture management has the ability to deliver financial and environmental benefits for the agriculture community. Satellites are flying over our farms on a daily basis offering an accurate spatial snapshot of plant health/vigour. Farmers are able to use these maps to assist a wide range of actions such as variable rate fertiliser, targeted weed & insect control and revised farm layouts/drainage plans. Normalised Difference Vegetation Index (NDVI) offers a measure of crop and pasture health and vigour. Near infrared (NIR) and red reflectance values from a passive light source (sun light) are measured spatially in order to create an NDVI map of a paddock/ farm/landscape. NDVI = (NIR – Red) / (NIR + Red) Healthy vegetation with a higher density will absorb a large portion of the red light and reflect the majority of the NIR light equating to a higher NDVI value. Unhealthy sparse vegetation absorbs less red light and reflects less NIR light equating to a lower NDVI value. Farms can use near real-time NDVI data to manage current season crops and pastures but can also access historical data to identify trends in production over time. PAGE 21 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Engaging in SoilCare in the Goulburn Broken Catchment Rhiannon Apted1, Mark Cotter2, Greg Bekker3
1 Goulburn Broken Catchment Management Authority, 5/10 High St YEA, Vic, 3717, [email protected] 2 Formerly Goulburn Broken Catchment Management Authority, 5/10 High St YEA, Vic, 3717 3 Department of Environment and Primary Industries, PO Box 124 BENALLA, Vic, 3672
Engaging landholders in soil management has long been a key outcome of government programs in the Goulburn Broken Catchment, starting in 1912 when the Victorian Department of Agriculture launched a major campaign to lime soils. So, 100 years later, how are we engaging landholders, and is it working? Here we describe the approach taken by the Goulburn Broken CMA through its SoilCare project, funded through a competitive grant from Caring for Our Country. SoilCare aimed to improve the quality of ecosystem services produced by catchment soils. However, with an 18 month project timeframe, we knew that our best approach was to engage landholders in learning about their soil, with the aim of influencing value placed on soil and practices to manage it. We set about building our relationship through engaging landholders in activities of importance to them. Initially this was largely about managing soil fertility through understanding soil capability and soil acidity. The presentation of high quality consistent information created landholder trust that the program would meet their needs. Involvement of industry strengthened the key messages at workshops, and provided a useful one-stop-shop source of information. More than 650 participants came through the program; approximately 55% were landholders attending follow-up or multiple events. Many groups went on to conduct their own soil related activities, and five farmers have established demonstration trials of sustainable farm practices. The success of the program was well recognised by the Australian Government, who have reinvested in the project allowing it to expand catchment-wide. PAGE 22 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Monitoring for change - creating effective soil health monitoring for the Victorian Mallee Michael Moodie1, Narelle Beattie2, Ben Jones3
1 Mallee Sustainable Farming, 2/152 Pine Avenue, Mildura Vic 3500, [email protected], 2 Mallee Catchment Management Authority, P.O. Box 5017, Mildura, Vic, 3500, [email protected] 3 Mallee Focus, Melbourne, Vic, 3000, [email protected]
Abstract Within the low rainfall Victorian Mallee region, dryland soils support 2.4 million hectares of agriculture (cereal cropping and livestock), providing over $600 million production each year. The Mallee Dryland Agricultural Soil Health Monitoring Program was created to better understand the condition of the region’s soils. The program assessed key soil health indicators at 155 focus and three benchmark sites between 2010-2012. At each site, a composite topsoil (0-10cm) sample was collected and bulk density was measured in a permanent one hectare monitoring area. Organic carbon, total nitrogen, Colwell phosphorus, electrical conductivity (1:5) and pH were also measured. A standard soil was included in all laboratory analysis. The monitoring confirmed high soil variability in Mallee topsoils. Across all sites, organic carbon varied from 0.2-2.2%, total nitrogen from 0.02-0.22%, Colwell phosphorus from 4.23-57.17mgkg 1, electrical conductivity from 0.03-2.79dSm-1, pH(CaCl2) from 5.31-8.35 and bulk density 0.92 1.69gcm-3. Data collected by the program was reviewed in 2013 to assess if monitoring could adequately detect change in condition and to recommend improvements. The review found that although sampling variability was controlled, the methodology did not adequately deal with inherent variability (between-, within-year and rotational variability). Recommended improvements included switching from measuring groups of focus sites within land systems every third year to random averaging of site measurements over the monitoring interval. Standard soil should be sub-sampled once, and not repeatedly mixed. Screening out unrealistically low (and erroneous) measurements could improve bulk density measures. This project was supported by the Mallee Catchment Management Authority through funding from the Victorian Government. PAGE 23 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
DustWatch – tracking soil condition change and its causes across southern Australia John Leys1, Stephan Heidenriech2
1 Office of Environment and Heritage, PO Box 20, Gunnedah, NSW 2380 [email protected] 2 Office of Environment and Heritage, PO Box 20, Gunnedah, NSW 2380 [email protected]
Abstract Wind erosion is an excellent indicator of soil condition as it integrates the effects of climate and land management. If soil is eroding, the soil condition is declining; soil carbon is being lost, ecosystem services are declining and the potential soil productivity is being eroded. The Community DustWatch program has established 43 real time monitoring nodes (DWN) across southern Australia (Fig 1). The data are available on line via the Community DustWatch information interface (https://codii.environment.nsw.gov.au/) and monthly reports are published on the web (http://www.environment.nsw.gov.au/dustwatch/). Figure 1. Location of DustWatch nodes
DustWatch reports monthly where erosion is occurring and its possible causes; that is, is the cause of the dust the climate, land use or land management practices. For example in south-western NSW there are nine DWN covering the two major land uses; rangeland and cropping. Over the eight year monitoring period in seven years the cropping sites had nearly double the hours of dust than the rangelands sites. As the seasons became drier from 2005 to 2010, the rangelands became increasingly the source of dust until they produced nearly double the dust level of the cropping sites in 2009/10. In selected areas ground based observational surveys of, erosion type, groundcover land use and land management practice are undertaken. For example the decrease in wind erosion levels between 2003 and 2013 in south-western NSW can be attributed to the increase use of chemically prepared fallow, from 3% to 30% of sites, instead of the conventionally used tillage. PAGE 24 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Report card on sustainable natural resource use in the agricultural regions of Western Australia Noel Schoknecht1
1 Department of Agriculture and Food, Western Australia, Baron-Hay Court, Kensington, 6151, WA [email protected]
Abstract A report card which summarises the current knowledge of the status and trend in land condition in the agricultural areas of the south-west of Western Australia (WA) was published in September 2013 by the Department of Agriculture and Food, Western Australia. The report card draws on best available evidence from government and industry on the current condition and trend of ten land- and water-related natural resource themes relevant to agriculture, and discusses the implications of these results for the agricultural industries. The report also discusses the three main factors driving the performance of the land ‑ namely climate, land characteristics and land management. The first two factors are largely out of the control of land managers, and in a drying and variable climate of the agricultural areas of WA, land management practices need to be able to respond quickly to changing conditions. Although this report deals with several natural resource themes individually, it is important to note that the processes within these themes are often linked, and any land management response needs to consider the system as a whole, and how this integrated system may respond to a given management action. In summary the situation and outlook for our natural resources is mixed. Although there has been progress in some areas, such as managing wind and water erosion, the status and trend in many indicators of resource condition is adverse.
Introduction The report card presents the best available information on the current condition (or risk to condition) and trend in condition of the natural resources that support agriculture. In particular, it: ›› provides a transparent process explaining how this condition, risk and trend was determined ›› highlights any issues which impinge on the sustainable use of this resource ›› discusses the implications of these findings on the agricultural industries ›› provides recommended actions where appropriate. The study area covers the agricultural areas of the south-west of WA, other than native vegetation and reserves (Figure 1). PAGE 25 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 2 The three primary factors Figure 1 Study area that influence the environmental performance of the land.
Sustainable natural resource use for agriculture means maintaining (and where possible improving) the productive capability of the land which underpins agriculture, while mitigating off-site impacts. The environmental performance of our land is a complex interaction of numerous processes. In simple terms, however, the performance of the land is driven by three primary factors – climate, land characterisitcs and land management (Figure 2). Together, these factors will determine the current condition of the land and how the land is performing. An understanding of the trends in land use/land management and climate change will also provide evidence for determining and monitoring trends in land condition. The climate and land characteristics factors are mostly outside of the influence of land managers, although land characteristics can be modified to a limited degree by management options, such as claying and delving. Land managers must therefore work within the given climate and land conditions and adopt land management practices that lead to profitable and sustainable outcomes. Land management will need to respond where one of the other factors (e.g. climate) is changing. There are critical situations where the current land management, even under current conditions, is unsustainable and leading to an unacceptable decline in land condition (DAFWA 2013). Land management will need to change in response to these conditions, otherwise the land use will become unviable. In discussing specific natural resource degradation issues, the influences of these three factors are considered and commentary is provided on what management is appropriate to ensure a sustainable and profitable agricultural future. PAGE 26 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Methods The report card reports on 10 soil- and water-related themes: 1. Soil acidity 2. Wind erosion 3. Water erosion 4. Soil organic carbon 5. Soil compaction 6. Water repellence 7. Dryland salinity 8. Nutrient status (phosphorus) 9. Nutrient export (phosphorus) 10. Acidification of inland waterways Each theme is divided into two parts: 1) a summary, including resource condition and trend (map and table) and key messages, and 2) the detail of how the assessment was made for that theme, including an overview, assessment method, current condition and trend (results), discussion and implications and recommendation/s. The method used to assess the theme varied between themes depending on the information available for the theme, and the nature of the theme.
Results The results for the 10 themes are summarised in Table 1. More detailed spatial analysis, based on broad spatial units (soil and hydrological zones) are provided in each theme chapter. In summary the situation and outlook for our natural resources is mixed. Although there has been progress in some areas, such as managing wind and water erosion, the status and trend in many indicators of resource condition is adverse. PAGE 27 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS In trend In trend Confidence Confidence In condition In hazard Very good Very low Well in Well excess Good Low Excess Fair Optimal Moderate Poor Deficient High Condition and trend Very poor Very Very deficient Hazard/risk and trend High Very Summary due to and a major risk to production and widespread Severe of condition areas, of agricultural lime. In most insufficient use is declining. the soil profile on sandy soils and can be a severe and often Widespread land management under current to production major limitation systems. to optimise phosphorus (P) than is required more In most areas, in many agricultural soils. is stored production seasons, the risk is Despite several below average growing is Vigilance cover. maintaining ground managed through largely this dry year, an exceptionally because after however, required issue may be significant. land management, current managed through The risk is largely mostly unknown. although actual levels are is unknown. issue but exact severity and trend Widespread Theme Soil acidity Water repellence Nutrient status (P) Wind erosion erosion Water Soil compaction Table 1 Resource status and trend summary and trend WA status for the south-west of 1 Resource Table PAGE 28 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS Summary Soil organic with a drying and warming climate. carbon associated soil organic levels of risk to current Possible Limited data. is unknown. highly variable and the trend carbon levels are of off-site impact the agricultural activities; however, significant input from indicates catchments several coastal from Data nutrients applied to agricultural lands is unknown. the in deep drainage. However, due to a reduction static is largely acid groundwater of expression variable. Surface Situation trend is unknown. condition is highly variable and the impact is significant. This theme has localised impacts, the off-site Widespread risk with variable spatial and temporal impact. risk with variable spatial Widespread and resource agricultural land, water extent threatens Future biodiversity assets. Containing and adapting to salinity is feasible, areas. is viable in only a few though recovery Dryland salinity trend summary and status resource data for Themes with insufficient Theme Soil organic carbon Nutrient export (P) of Acidification inland waterways PAGE 29 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Discussion With increasing global demand for food and fibre there are many opportunities and challenges for the agri-food sector. The report card provides a snap-shot of some of the biophysical challenges the sector is facing. Our challenge is to balance our need to achieve agricultural productivity growth while ensuring our natural resources are healthy and resilient. The report card provides an evidence base to help meet this challenge. Key principles to be considered in this challenge are:
Stewardship of natural resources The maintenance or enhancement of this vital resource base for the long term – is of prime importance. Those who directly manage the land need to be provided with the information, resources and support to carry out this critical role.
Changing Climate – variability and trends Ensuring sustainable natural resource use with extreme events, such as long-term below average rainfall, short-term extreme events such as flooding or drought is difficult. Climate-smart agriculture is now being used to bring together actions that achieve a more resilient and climate aware agri-food sector to meet both trends in climate and extreme events.
Relevant resource information is important Knowledge and information systems are the basis for sound adaptive management. That is, we need to understand the state and trends in our natural resources, the impacts of the pressures on our environment and the impacts of our management strategies, so that we can progressively adapt and improve those strategies. Long-term collections of data in many aspects of our natural resources is currently limited, which severely constrains our ability to develop and enact evidence-based responses – from policy to on-ground action.
Understanding of process essential Many soil and water processes are linked, and efficient solutions to problems must consider the system as a whole, rather than the issue in isolation. An understanding of how the system operates, and how the three primary factors – climate, land characteristics and land management – interact is required when changes to land management are considered to address individual issues.
Economics driving sustainability To ensure the sustainable use of our natural resources for agriculture requires a viable rural economy. It is difficult to look at the long-term viability of soil and water resources if it is hard to maintain a viable farm business.
Sustainability Indicators Sustainability indicators could focus on natural resource condition and off-site environmental impacts, or could incorporate long-term farm income, managerial skills, and social and economic aspects of agriculture.
Innovation to support sustainable agriculture Innovation has been, and will be, a key contributor to solving the problems faced by the agri-food sector. For example, we already know the great benefits realised by precision PAGE 30 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
irrigation, minimum tillage, and improved seed varieties. The challenge is to ensure that innovation keeps flowing. That means getting the conditions right so that both private and public research and development can provide the solutions needed to achieve sustainable use of our natural resources.
Our natural resources are essential and require action at all levels Achieving sustainable agriculture is the responsibility of all participants in the system, including land managers, farm businesses, policymakers, researchers, retailers, and consumers. Each group has its own part to play and its own unique contribution to make to strengthen the sustainability of our agriculture.
Reference DAFWA (2013) Report card on the sustainable natural resource use in agriculture. Department of Agriculture and Food, Western Australia. Available online at www.agric. wa.gov.au
Visual assessment of soil structure in the field to improve land quality David McKenzie1
1 McKenzie Soil Management Pty Ltd, PO Box 2171, Orange NSW 2800 Email: david.mckenzie@ soilmgt.com.au
Abstract The critical importance of soil structure for successful land management is understood in general terms by most Australian farmers. Unfortunately, very few have a system in place to accurately assess and monitor the structural condition of their topsoil and subsoil. Many are “flying blind” when attempting to incorporate soil improvement programs into their farm financial plans. Objectives that can be achieved through successful management of soil structure include: substantial water entry and storage without development of severe waterlogging, avoidance of excessive soil hardness for root extension and function, and development of suitable habitat for soil biota. The local cotton industry has shown what is possible with systematic assessment and management of soil structure. Their SOILpak decision support system (McKenzie 1998, 2013a) recognises the need to efficiently measure the three aspects of soil structure recognised by Kay (1990) in the topsoil and subsoil; structural form (assessed via the visual-tactile ‘SOILpak score’), structural stability in water (determined through use of the ASWAT test and associated laboratory data) and structural resilience. The Cotton SOILpak manual has been modified for use by other rural industries, eg. Anderson et al. (1999). The rapid ‘SOILpak score’ procedure (McKenzie 2001a) – applied either through the use of trimmed soil pit faces or tiling spade profiles – has been successfully correlated with soil shear strength and air-filled porosity (McKenzie 2001b) and is linked to soil degradation thresholds (McKenzie and McBratney 2001). The results are numerical and can be mapped easily with red-amber-green colour coding; conventional pedality PAGE 31 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
descriptions (National Committee on Soil and Terrain 2009), in contrast, are difficult to map clearly. Measurements of soil structural form are an important component of estimates of soil water holding capacity (McKenzie et al. 2008). Alternatives to the ‘SOILpak score’ procedure used in other parts of the world are described by Ball et al. (2007) and Shepherd (2000). The ASWAT score (Field et al. 1997) is an abbreviated version of the Loveday and Pyle (1973) test. It also can be mapped easily with colour coding – unlike Emerson dispersion results (Emerson 2002) that include character data. ASWAT and ‘SOILpak score’ maps can be related to maps of ‘soil amelioration requirements’ (eg. loosening of compacted layers, gypsum application) and ‘cost of repair of soil constraints’, which can then be linked to crop yield maps to provide farm profitability maps. The development of new guidelines for assessment of prime agricultural land in the vicinity of proposed mining projects in eastern Australia is allowing soil consultants to show regulators and farmers what is possible with application of visual-tactile soil assessment procedures at a scale broader than individual paddocks. An example of the new requirements is the NSW Biophysical Strategic Agricultural Land (BSAL) Protocol (NSW Government 2013). The Spur Hill Agricultural Resource Assessment (McKenzie 2013b) is an example of a consultant report that was prepared in response to these new guidelines; it is part of a submission to the NSW Government ‘Mining and Petroleum Gateway Panel’. The increasing demand for soil assessment services in NSW will provide great opportunities for soil consultants, but skill levels need to be improved. Competence in visual soil examination and evaluation must be given greater priority in all soil teaching programs. Research is required to determine an appropriate balance between visual soil assessment and remote sensing / landscape modelling for contrasting settings across Australia.
References Anderson AN, McKenzie DC, Friend J (Eds.) (1999) ‘SOILpak for dryland farmers on the red soil of Central Western NSW.’ (NSW Agriculture: Orange); http://www.dpi.nsw.gov.au/ agriculture/resources/soils/guides/soilpak/central-west Ball BC, Batey T, Munkholm LJ (2007) Field assessment of soil structural quality – a development of the Peerlkamp test. Soil Use and Management 23, 329-337. Emerson WW (2002) Emerson dispersion test. In: ‘Soil physical measurement and interpretation for land evaluation’. (Eds N McKenzie, K Coughlan, H Cresswell) pp. 190- 199. (CSIRO Publishing: Collingwood). Field DJ, McKenzie DC, Koppi AJ (1997) Development of an improved Vertisol stability test for SOILpak. Australian Journal of Soil Research 35, 843–852. Kay BD (1990) Rates of change of soil structure under different cropping systems. Advances in Soil Science 12, 1-52. Loveday J, Pyle J (1973) The Emerson dispersion test and its relation to hydraulic conductivity. CSIRO Division of Soils, Technical Paper No. 15, Canberra. McKenzie DC (ed.) (1998) ‘SOILpak for cotton growers, third edition.’ (NSW Agriculture: PAGE 32 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Orange); http://www.dpi.nsw.gov.au/agriculture/resources/soils/guides/soilpak/cotton McKenzie DC (2001a) Rapid assessment of soil compaction damage. I. The SOILpak score, a semi-quantitative measure of soil structural form. Australian Journal of Soil Research 39, 117–125. McKenzie DC (2001b) Rapid assessment of soil compaction damage. II. Relationships between the SOILpak score, strength and aeration measurements, clod shrinkage parameters and image analysis data on a Vertisol. Australian Journal of Soil Research 39, 127–141. McKenzie DC (2013a) Visual soil examination techniques as part of a soil appraisal framework for farm evaluation in Australia. Soil & Tillage Research 127, 26-33. McKenzie DC (2013b) ‘Agricultural Resource Assessment to support a Gateway Application for the Spur Hill Underground Coking Coal Project.’ A report prepared for Spur Hill Management Pty Ltd by McKenzie Soil Management Pty Ltd, Orange NSW, November 2013; https://majorprojects.affinitylive.com/public/9c940f4f7830bb51dc891ac0440edaaa/ Appendix%20A_%20Agricultural%20Resource%20Assessment.pdf McKenzie DC, McBratney AB (2001) Cotton root growth in a compacted Vertisol (Grey Vertosol). I Prediction using strength measuring devices and ‘limiting water ranges’. Australian Journal of Soil Research 39, 1157–1168. McKenzie DC, Rasic J, Hulme PJ (2008) Intensive survey for agricultural management. In: ‘Guidelines for surveying soil and land resources: Second edition’ (Eds NJ McKenzie, MJ Grundy, R Webster, AJ Ringrose-Voase) pp. 469-490. (CSIRO Publishing: Collingwood). National Committee on Soil and Terrain (2009) ‘Australian Soil and Land Survey Field Handbook, Third Edition’ (CSIRO Publishing: Collingwood). NSW Government (2013) ‘Interim Protocol for site verification and mapping of biophysical strategic agricultural land’. http://www.planning.nsw.gov.au/Portals/0/ StrategicPlanning/interim_bsal_protocol.pdf Shepherd TG (2000) ‘Visual soil assessment. Field guide for cropping and pastoral grazing on flat to rolling country’. (Horizons.mw & Landcare Research: Palmerston North, New Zealand). PAGE 33 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Strategic Tillage in Conservation farming systems; its impact on soil health and productivity Mark Crawford1, Yash Dang1, Anna Balzer1, Vivian Rincon-Florez2, Lilia Carvalhais2
1 Department of Science, Information Technology, Innovation and the Arts (DSITIA), Qld, Australia, [email protected] 2 Queensland Alliance for Agriculture and Food Innovation, The University of Queensland, Australia
Abstract The adoption of conservation farming in Queensland (Qld) has greatly reduced energy and machinery inputs while significantly improving overall soil health and productivity. However, the control of crop weeds and diseases in No-Till (NT) systems has become increasingly difficult for landholders in the Northern Grain Belt, with strategic tillage (ST) being considered as a potential management option. This study investigated the effects of tillage on the physical, chemical and biological properties of soil, in soil profiles from five tillage trials located in the Northern Grain Belt of Australia. The study area extended from Biloela (Vertosol, 666mm annual rainfall), Condamine (Sodosol, 624mm annual rainfall), Moonie (Dermosol, 636mm annual rainfall) and Warwick (Vertosol, 675mm annual rainfall) in Qld to Wee Waa (Vertosol, 582mm annual rainfall) in New South Wales (NSW). Tillage implements/treatments included chisel plough, offset disc and prickle/ disc chain based on timing and frequency. Soil samples from 0-0.30m were analysed for total and particulate organic carbon (TOC, POC), available phosphorus, bulk density, soil moisture, and microbial activity. In-crop weed density was also recorded. The initial impacts of strategic tillage on soil moisture were largely restricted to the 0-0.10m range with slight non-significant decreases occurring. Available phosphorus, TOC, POC and total microbial activity were not significantly impacted by either frequency or implement type. The implications of ST for the purpose of weed/disease control in NT management systems in the Northern Grain Belt of Qld on soil health are discussed.
Introduction Within agricultural systems, the aim of conservation farming is to enhance soil health so that the land can maintain or increase farm profitability, as well as to conserve soil resources for future farming generations (Gregorich, 2002). Conservation farming has led to many benefits such as increased soil flora/fauna biodiversity, increased organic content, improved soil structure and fertility. A land management survey conducted during the period 2007-08 in Qld stated that of the 2.7 million hectares of land prepared for crops and pastures, 47% was prepared using no-till compared with 53% prepared using one or more cultivation passes. Nationally, 65% of the 26.9 million hectares used for crops and pastures was prepared using no-till (ABS, 2009). Zero-tilled cropping systems have created an environment that favours weed species that germinate on or near the soil surface, such as common sowthistle (Sonchus oleraceus), flaxleaf fleabane (Conyza bonariensis) and feathertop rhodes grass (Chloris virgata) (Walker, 2012). Argent et al. (2013) conducted a survey of landholders and industry advisors within the Northern Grain Belt of Australia and observed that there were greater reported issues with weed and disease control in NT farming systems than those implementing one or more cultivation passes. An integrated weed management (IWM) approach utilising strategic tillage within conservation farming systems could be used to combat the issues stated above. Conservation farming systems need to be flexible and responsive and able to work within the constraints and opportunities of the environment rather than against it (O’Gara, 2010). PAGE 34 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The impact of strategic or occasional tillage on soil health and productivity in conservation systems has not been well established by any previous research. Studies on silty loam and silty clay loam soils such as Kettler et al. (2000) and Wortmann et al. (2010) have reported that soil properties including organic matter, aggregate stability, pH, water content and microbial activity were not significantly affected by occasional or strategic tillage, and that little or no impact on crop production was found after 3-5 years. This is in contrast to studies on coarse and fine loamy soils such as Grandy et al. (2006) and Stavi et al. (2011) that observed negative impacts due to tillage on key soil properties, such as penetration resistance, bulk density, aggregation, water infiltration, field moisture capacity, and organic carbon concentration. This research aims to determine the impacts on soil health and productivity by introducing ST in otherwise NT systems for the purpose of controlling hard-to-kill weeds such as flaxleaf fleabane, feathertop rhodes and windmill grass (Chloris truncata) and glyphosate resistant barnyard grass (Echinochloa crus-galli). We selected five fields on long-term NT-managed soils (7-44 yrs) under semi-arid subtropical climatic conditions to represent typical CA farming systems across north-eastern Australia. The soil health indicators investigated were bulk density (BD), organic carbon (TOC), particulate organic carbon (POC), available phosphorus (P), volumetric moisture, weed density, total microbial activity and erosion/sediment loss. Productivity was assessed by crop grain yields for the 2012 winter cropping season.
Materials and Methods The study sites were selected to be representative of the Northern Grain Belt and included sites in Northern and Central Qld and Northern New South Wales. A total of five sites were chosen on three different soil types including Vertosols (three), Sodosol (one) and Dermosol (one) with <1% slope. Tillage treatments included chisel tine, offset disc and prickle/disc chain to depths of 0-0.20m; with four replicates per treatment of 100m in length and between 12 – 18m wide (sowing implement width). A randomised complete block experimental design was used for all sites except Warwick, where the design of the long term tillage trial (Marley and Litter, 1989) had the NT plots split in two with the application of chisel tine in 2012. For the sites at Moonie, Warwick and Wee Waa one tillage treatment was applied in March 2012. Condamine and Biloela received two tillage treatments in March and April 2012, with the site at Biloela receiving the same treatment the following year. Two soil samples from each replicate were taken 3 and 12 months after the initial tillage to depths of 0-0.30m and sectioned into 0-0.10, 0.10-0.20, 0.20-0.30m depth intervals. The first sample from each replicate was oven dried at 105ºC while the second sample was oven dried at 40ºC and ground to pass through a 2 mm-sieve. BD and soil moisture was calculated from the first sample. The second sample was used to determine available P by the Colwell procedure (Method 9B2, Rayment and Lyons, 2011); particulate organic carbon (POC) was determined only on the 0-0.10m layer and quantified by physical separation as <2 mm and >0.053 mm sizes (Cambardella and Elliot, 1992); and total soil organic carbon (TOC) was determined on a sub-sample from the second replicate and ground to pass through <0.5 mm-sieve (Method 6B3, Rayment and Lyons, 2011). Equivalent soil mass was used to compare TOC stocks (Wendt and Hauser, 2013). The procedure from Adam and Duncan (2001) was used to determine soil microbial activity by hydrolysis of fluorescein diacetate (FDA) in composited (0-0.20m) soil samples. Weed populations were determined at the tillering growth stage of the crop using a 1m x 1m quadrant, randomly placed within the trial plot area. Water runoff and sediment loss during the weather experienced over the period of 1 January2000 – 1 January2013 was simulated using the water balance model, Howleaky PAGE 35 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
(McClymont et al. 2013) for the potential impact of ST every three years. The soil model parameters were kept uniform and differences in leaf area index simulated due to changes in timing, type and frequency of tillage (Freebairn et al. 2009).
Results and Discussion Weed populations after three months in NT systems on all the soils examined were significantly lowered (Table 1) after the imposition of a single tillage operation. Weed populations were also significantly reduced after the imposition of a second tillage in 2012 at the Biloela and Condamine sites. In the following year a lower weed population over all treatments in Biloela and Moonie was observed. However, there were no significant differences in weed populations recorded between treatments for 2013. The Biloela site had the tillage treatments applied again for the purpose of assessing how many tillage operations could be used before the soil health deteriorated, and if the weed populations differed between wheat (2012) and chickpea (2013) crops. The weed population was significantly lower when compared between crops but was not significantly different between treatments. At Condamine, there was an increase in the weed population 12 months post tillage and in particular African turnip weed (Sysibrium thellungii). These results were significantly different between Chisel 2 passes and NT treatments with a high variance value also recorded. A comparison between the two sampling dates demonstrated a significant difference between the different crops, chickpea (2012) and wheat (2013) in NT. Seedling emergence can be impacted both positively and negatively by tillage, with problem/dominant weed species needing to be identified before tillage occurs to limit potential problems. Chauhan et al. (2006) states that seedling recruitment of wild radish was higher under minimum tillage than under no-till, while annual sowthistle and turnip weed recruitment was higher under the no- till system. Further research is required into the impacts of tillage on the Condamine site regarding the increase in African turnip weed population and other weeds that will potentially increase with soil disturbance. Table 1. Average in-crop weed populations (number/m2) 3 and 12 months after tillage in long-term NT. Within sites, means followed by the same letter do not differ significantly at P < 0.05. Lower case letters depict comparisons between treatments in the same sampling season and capital letters depict comparisons between sampling seasons in the same treatment.
Biloela Condamine Moonie Wee Waa Months 3 12 3 12 3 12 3 12 No-till 10.5aA 3aB 14.5aA 4.5aB 9.2aA 0.75aB 4.3a - Chisel 1 pass 1.25bA 0.3bB 2.25bA 15.6aA 1.0bA 0.125aA 0.7b - Chisel 2 passes 4.25cA 0.2bA 6.5cA 23.4bA - - - - Offset disc - - - 1.2bA 0.25aA - - Disc chain ------1.5b - LSD 5% 1.7 2.39 2.8 17.4 3.2 NS 1.3 -
Due to the nature of the tillage treatments BD results were highly variable after three months and to a lesser extent 12 months (results not shown). At Condamine a significant decrease was recorded after 12months. Soil moisture was not significantly impacted PAGE 36 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
by tillage after three months except for Warwick where there was a significant negative impact in the 0-0.10m layer. Twelve months after tillage there were no significant differences at any site in the 0-0.30m depth. At Biloela, where the site had a repeat tillage treatment the following year, no significant differences were recorded between tillage implements. A significant (P<0.05) increase in soil moisture was recorded at Biloela in the 0.10-0.20m depth between the 3 and 12 month period for the chisel treatments (Chisel 1 pass 2012-37.5mm, 2013-43.0mm) (Chisel 2 pass 2012-36.4mm, 2013-43.3mm). However, due to dry weather conditions the site at Wee Waa was not sampled at 12 months. The effect on TOC stock in the 0-0.30m depth 3 and 12 months after tillage was not significant for any of the tillage treatments at Biloela, Moonie, Warwick and Condamine. The impacts of any treatment on POC were not significant at any site in the 0-0.10m layer after three months. Available (Colwell) P in the 0-0.30m soil depth 3 and 12 months after a single tillage operation did not change significantly in NT and ST at any site except Biloela. There was a significant decrease at Biloela in the 0-0.10m layer between NT and tillage treatments three months after tillage; however, no significant difference was found after 12 months. The second application of tillage at Biloela did not significantly impact available phosphorus at any depth. Microbial activity between tillage treatments and no-till did not differ at any site. This may be due to hydrolysis of FDA being a “broad-scale” measurement for soil enzyme activity which may not be sensitive enough to detect changes in specific activities due to functional redundancy of soil microbial communities (Chaer et al., 2009). This suggests that major microbial functions were maintained in all treatments, possibly due to a high long term functional stability of soil microbial communities which has been suggested to occur under wheat systems (Drijber et al., 2000). Nevertheless, there was a difference in microbial activity between years within each treatment at Biloela. There were two main seasonal differences between years in Biloela: temperature and rainfall. In 2013, minimum average temperatures were five degrees higher than in the previous year. From harvest to sampling, the total rainfall was 343 and 601.4 mm for 2011-2012 and 2012- 2013, respectively. It has been suggested that temperature limits microbial utilisation of substrate carbon (Zogg et al., 1997). Heat shocks were also shown to decrease soil microbial biomass and respiration. This may result in changes in microbial community composition (i.e., different dominant populations at higher temperatures). A decline in active microbial biomass with temperature has also been previously reported, possibly because of greater metabolic stress (Zogg et al., 1997). There were slight positive impacts on productivity at Biloela (Wheat), Moonie (Barley), Warwick (Wheat) and Wee Waa (Wheat) after one pass of the chisel tine, however these were not statistically significant. The site at Condamine recorded a slightly (p=0.08) significant increase in chickpea yield (1.07 – 1.16t/ha) after a single chisel treatment, but the 2nd tillage did not further improve or reduce productivity. The results from the simulated sediment loss and runoff model suggested that for the climatic conditions recorded over the last 10 years, the introduction of a tillage treatment every 3 years would be manageable ( ≤0.3t sediment loss p.a) at all sites except Condamine. Due to the dispersive nature of Sodosols at this site the introduction of a tillage treatment could cause up to 1.2t/ha sediment delivery off-site as well as resulting in 57mm of annual runoff. PAGE 37 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
In summary, there were no overall significant impacts on soil health caused by ST 12 months after implementation, with the exception of BD at Condamine. There were significant positive impacts of ST for weed control after 3 months at all sites and a positive result 12 months at Biloela and Moonie. At Condamine a large number of African turnip weeds were recorded suggesting that ST had a negative impact on weed control 12 months after application at this site.
Acknowledgements We are indebted to our collaborative growers, Darren Jensen, Nev and Ron Boland, Rod and Sam Hamilton and Ken and John Stump for providing field sites, managing the trials and providing their generous support. We thank the Grains Research and Development Corporation for partial funding of this study.
References Adam G, Duncan H (2001) Development of a sensitive and rapid method for the measurement of total microbial activity using fluorescein diacetate (FDA) in a range of soils. Soil Biology and Biochemistry 33, 943-951. Argent S, Wixon A, Dang Y (2013) Farmers thoughts about CTF in Australia’s northern grain growing region. First International Controlled Traffic Farming Conference, Toowoomba, 25-27 February 2013. http://www.actfa.net/conferences/ctf2013/ CTF2013%20papers%20pdfs/Argent,%20Suzette.pdf. Australian Bureau of Statistics (2009) Land management and farming in Queensland, 2007-08, cat. No. 1318.3 - Qld Stats. Cambardella CA, Elliot ET (1992) Particulate soil organic-matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56, 777-783. Chaer G, Fernandes M, Myrold D, Bottomley P (2009) Comparative resistance and resilience of soil microbial communities and enzyme activities in adjacent native forest and agricultural soils. Microbial Ecology 58(2), 414-424. Chauhan BS, Gill G, Preston C (2006) Seedling recruitment pattern and depth of recruitment of 10 weed species in minimum tillage and no-till seeding systems. Weed Science 54(4), 658-668. Drijber RA, Doran JW, Parkhurst AM, Lyon DJ (2000) Changes in soil microbial community structure with tillage under long-term wheat-fallow management. Soil Biology & Biochemistry 32(10), 1419-1430. Freebairn DM, Wockner GH, Hamilton NA, Rowland P (2009) Impact of soil conditions on hydrology and water quality for a brown clay in the north-eastern cereal zone of Australia. Australian Journal of Soil Research 47, 389–402. Grandy AS, Robertson GP, Thelen KD (2006) Do productivity and environmental trade- offs justify periodically cultivating no-till cropping systems? Agronomy Journal 98, 1377- 1383. Gregorich EG (2002) Quality. In ‘Encyclopedia of soil science.’ (Ed. R Lal) pp. 1058–1061. (Marcel-Dekker: New York) Kettler TA, Lyon DJ, Doran JW, Powers WL, Stroup WW (2000) Soil quality assessment after weed-control tillage in a no-till wheat-fallow cropping system. Soil Science Society of America Journal 64, 339-346. PAGE 38 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Marley JM, Litter JW (1989) Winter cereal production on the Darling-Downs-an 11 year study of fallow practices. Australian Journal of Experimental Agriculture 29, 455-481. McClymont D, Freebairn DM, Rattray DJ, Robinson JB, Silburn DM, Owens J (2013) Howleaky (Version 5.44.07)-a tool for exploring the impact of management, soil and vegetation on water balance and water quality. http://www.howleaky.net/index.php/ download O’Gara FP (2010) Striking the balance – conservation farming and grazing systems for the semi-arid tropics of the Northern Territory. Second Edition. Northern Territory Government, Australia. Rayment GE, Lyons DJ (2011) Soil chemical methods-Australasia. CSIRO Publishing, Collingwood, Victoria, Australia. Stavi I, Lal R, Owens LB (2011) On-farm effects of no-till versus occasional tillage on soil quality and crop yields in eastern Ohio. Agronomy for Sustainable Development 31, 475- 482. Walker S (2012) “Capturing opportunities and overcoming obstacles in Australian agronomy”. Edited by I. Yunusa. Proceedings of 16th Australian Agronomy Conference 2012, 14-18 October 2012, Armidale, NSW. Wendt JW, Hauser S (2013) An equivalent soil mass procedure for monitoring soil organic carbon in multiple soil layers. European Journal of Soil Science 64, 58-65. Wortmann CS, Drijber RA, Franti TG (2010) One-time tillage of no-till crop land five years post-tillage. Agronomy Journal 102, 1302-1307. Zogg GP, Zak DR, Ringelberg DB, MacDonald NW, Pregizer KS, White DC (1997) Compositional and functional shifts in microbial communities due to soil warm. Soil Science Society of America Journal 61, 475-481. PAGE 39 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Space-time monitoring of soil moisture to improve farm management A. Horta, T. F. A. Bishop
Faculty of Agriculture and Environment, The University of Sydney, Sydney, NSW 2006, Australia Corresponding author – [email protected]
The expected increase in food demand will lead to increased water use in agricultural production in order to improve crop yields, maintain long-season dual-purpose crops and extend areas of perennial pastures. This combined with possible future climate shifts may mean that we will be farming drier and drier soils. One impediment to water-use efficiency and improved crop production is reliable estimates of soil moisture. Accurate estimates of soil moisture can help farmers make better decisions about what and when to sow in their paddocks or when and how much to fertilise pre- and mid-season. The need to monitor soil moisture has lead to the implementation of monitoring networks such as OzNet (Murrumbidgee catchment) or the ones maintained by DEPI Victoria and FarmLink Research (southern NSW). Although space-time datasets are available the question remains about how transferable the information is to field and farms without probes. Also, remote sensing products available for soil moisture only refer to the top few cm of soil and are provided at a coarse resolution. To tackle this issue we propose to predict soil moisture for the entire profile at a resolution of 1km and 8 days using a combination of in-situ measurements and remote sensing products which parameterize components of the soil water balance equation. Our methodologies will be tested for the Muttama catchment in southern NSW where soil moisture probes are being installed. Additional, historical data from the FarmLink Research network will be used to evaluate moisture changes in space in time and validate our approach. PAGE 40 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil moisture monitoring for crop management Dale Boyd1
1 Department of Environment and Primary Industries, PO Box 441,Victoria Echuca 3564, Dale.Boyd@ depi.vic.gov.au
Abstract The ‘Risk management through soil moisture monitoring’ project has demonstrated the capability of current technology to remotely monitor and communicate real time soil moisture data. The project investigated whether capacitance probes would assist making informed pre- and in-crop decisions. Crop potential and cropping inputs are increasingly being subject to greater instability and uncertainty due to seasonal variability.
Introduction There are limited examples of the use of soil moisture probes in a dryland cropping system in Victoria but increasing interest in the farming sector from both service providers and farmer groups. Growers current cropping systems may not be maximising water use efficiency, if they are using subjective assessments. The practice of using moisture probes has had limited use in the dryland cropping industry and as such many farmers are not aware of the way to utilise this technology. Moisture sensors and telemetry devices that use the mobile network to send the data to be securely stored on a server that is accessed with the internet. Farmer focus groups that have connection to local sites have assisted in the validation of examining deep soil moisture and determining usefulness, usability, and availability. The groups have been educated in interpretation of data, and during the project have received regular email updates explaining recent soil water changes. Once competency of data interpretation is satisfactory, login details were supplied that allow access to live data.
Results A scoping study on nine focus groups was conducted of farmers involved in the projects monitoring sites and found that:
1. Soil moisture levels are estimated/measured by over 90% of the group and conducted by a number of methods, generally subjective. Most popular were estimations from recent rainfall events, crop condition, and drive by observations. Some were using self-calibrated tools such as push probes and soil sampling and determining the wetness of soil. The general feeling was that estimations could be improved on current methods and enough cases existed where they had got it wrong to explore alternatives. 2. Agronomists generally had a greater interest in soil moisture monitoring technology as a need to measure, when advising on critical business decisions. Benefits were identified with time efficient monitoring and the ability to compare previous year’s data. There was a very small percentage of farmers who were using volumetric sampling, where millimetres of water could be calculated, with some farmers taking these measurements and entering them into a modelling program. PAGE 41 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
1.3. Being a state-wide project across many rainfall districts and soil types, there are some differences identified with use of probes by the groups of farmers. In low rainfall areas, using soil moisture probes may aid the crop choice decision, being guided by pre sowing plant available water and also the time of sowing where good soil moisture reserves give confidence to sow by the calendar. In higher rainfall zones with reliable winter rainfall, farmers will follow rotations as guided by Best Management Practices and not by soil moisture reserves, but knowing that soil moisture levels will allow strategic inputs through the growing season to target potential yields.
Access to this data enabled project participants - ›› To observe real time deep soil moisture (30-100cm) at one representative point on a district farm with a common soil type. ›› To observe crop upper and lower limits under different soil types and crop types using the absolute soil moisture content plotted on the Intelligraph software. Over the three seasons of monitoring the project has built up knowledge of estimated crop upper and lower limits under different soil and crop types. The seasons so far have been quite remarkable with at least one wet summer (some sites with two and/or floods). The majority of springs at the sites have been low rainfall (decile 2-4), which has seen a huge depletion in soil moisture through the months of late August to October. The value of sub-soil moisture has been clearly evident and participants have been amazed of the ability of crops to use moisture from 60cm and beyond and how quickly a large biomass crop will use moisture. The following are examples of where this information could be used to assist in decision making in dryland cropping.
Figure 1. Accumulation of soil moisture through summer and Autumn from a dry soil profile PAGE 42 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Site at Youanmite (North East Victoria), dotted red oval indicates the depletion of soil moisture reserves through spring 2011 and the dashed blue oval represents the flooding rains at the site in early March 2012 that refilled the profile. The star indicates the sowing time where the crop was planted with a good profile of deep moisture and farmer confidence in the season was high. The green shaded area is where plant moisture use has been observed, and the white area below is nearing crop lower limit and moisture stress. The black line is a summed total of soil water content from 30-100cm.
Figure 2. Summed soil moisture graph displaying plant water use through spring with high levels of plant available water Site at Raywood (Central Victoria), the red oval highlights the depletion of a near full profile of moisture recorded in July 2012 with a low decile spring. The level recorded in November was an estimated crop lower limit for that monitored soil type of clay loam. PAGE 43 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 3. Individual sensor stacked graph of the Raywood SMM site A zoomed view of the Raywood site of all the individual sensor depths (positioned every 10cm starting at 30cm to 100cm). Dotted ovals highlight the moisture use from the different soil depths. Early spring it is used from the easiest source point being the shallow sensors while in late September moisture is being used from 60cm down to 100cm. The dashed oval indicates that the sensors at that depth has reached crop lower limit with no change in the soil water content (flat line) during October. At this site with the dry spring, crop lower limit was reached down at 100cm. (NB. separate sensor graphs are defined by colour and sensor lines are arranged as per depth location, shallowest on top and deepest on the bottom).
Figure 4. Varied water use through two spring periods associated with different levels of plant available water. PAGE 44 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Summed soil moisture content from site at Speed (Mallee), dotted red oval highlights the depletion of a full profile of moisture in 2011 with a low decile spring. The soil moisture reserves were never built up to those levels again in the 2012 even after significant rain in July deposited moisture deep (dashed blue oval). Low moisture use in spring 2012 was attributed to low crop vigour and limited plant available moisture.
Participant evaluation In a targeted survey of those who received regular DEPI correspondence it was found that: 1. 50% of the audience found the soil moisture information generated relevant for them and less than 10% indicted it was not relevant. 2. 85% have improved their knowledge/ability to assess soil moisture compared to prior to the project, with the most used indicator of soil moisture still being rain fall records. 3. 100% have indicated they will continue to use some form of the technology to monitor soil moisture levels in the future. It is hoped that continued access to this information will assist informed input decisions. This will minimise input in low decile years with a low soil moisture base and maximise yield potential in more favourable conditions based on soil moisture and positive seasonal forecasts.
Discussion The challenge to the grains sector is to increase productivity growth. This may be from a number of options, some of which include: ›› Better farm decisions through improved information - not just more information ›› Potentially through the uptake of new technology. The soil moisture monitoring network has proven to successfully link to both of these with positive results. Challenge – Micro monitoring one probe for paddock size management. Different soil types across paddock and farm and how many probes are required versus the ability to vary paddock management. PAGE 45 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Online information and tools for farmers and land managers Andrew McAllister1
1 Agriculture Research Division, Department of Environment and Primary Industries, Tatura, Victoria, Australia [email protected]
Abstract One challenge for agricultural research and extension organisations in transferring knowledge to the farming sector is the ability to contextualise information to the local farming environment. Rapid expansion in on-line accessible spatial information and tools has opened opportunities to enhance decision support systems in farm resource management. The Victorian Department of Environment and Primary Industries is developing FarmWeb 2.0 a web-based spatial mapping and data management program which will support the agricultural industries with access to a range of decision support products that require contextualising to the farm. To demonstrate the application of on-line spatial information to enhance decision support DEPI is working with the dairy company Murray Goulburn Cooperative (MGC) Ltd to trial the application of a FarmWeb 2.0 based nutrient management decision support system with MGC suppliers and fertiliser agronomists.
Soil news – keep it real, not political and choose the right media Ron Aggs1
1 Former editor of Agriculture Today; climate and agricultural research and extension photo-journalist, PO Box 6114, Conder, ACT, 2906, [email protected] The ‘Big Ask’ From late-2012 until now the fact has been obscured from Australians generally (though not from specialists) that soil has been bulldozed around negligently, indiscriminately, in high stakes deceptive politics still not widely reported or understood. ‘Enough soil carbon to mitigate climate change is a big ask’ was a p2 definitive piece in the October 2012 edition of Agriculture Today, the NSW Department of Primary Industries’ flagship farm research, advisory and management newspaper. Expectations were unrealistic for delivering increases in sequestration of carbon in soil, said one of Australia’s most respected soil scientists Dr Mark Conyers, delivering the 2012 Harald Jensen lecture, hosted by the NSW Branch of the Australian Society of Soil Science. Dr Conyers was a Department of Primary Industries (DPI) principal research scientist based at Wagga Wagga Agricultural Institute and the Harald Jensen lecture is a key annual event on Australia’s soil science calendar. The content of Dr Conyers’ story in Agriculture Today has never been contradicted to me, nor to my knowledge, to anyone associated with it. PAGE 46 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Beyond its usual readership in The Land newspaper and online, the October 2012 Ag Today story: dovetailed with Lateline’s (ABC TV) reporting that the Federal Coalition’s - now Australian Government’s - climate policy could not demonstrate that storing carbon in Australian soils would achieve the major proportion of a target to reduce Australia’s greenhouse emissions by 5% (of 2000 levels) by 2020 and provided background and impetus for ABC FactCheck’s negative verdict that during the 2013 the federal election campaign, voters were not hearing “the full story on climate research”. Lateline had twice previously reported the issue along similar lines in April 2011. A concise and short timeline traces the disintegration of Direct Action’s credibility: After the 2013 Lateline report by Steve Cannane (April 18) when host Tony Jones interviewed now Environment Minister Greg Hunt, the Coalition went quiet about the potential for storing carbon in soils, then briefly flirted with reafforestation as the mainstay of Direct Action, which led to the critical verdict from the ABC FactCheck unit, then generally fell back only on saying they were confident of easily reaching emissions reduction targets. So the key question to Mr Hunt and the government should continue to be: “tell us exactly what methodologies you know will work to meet the 5% target, as rather than what you keep claiming you’re ‘confident’ about”. I questioned the new Federal Member for Fairfax, Clive Palmer at his two National Press Club appearances, on November 12 (2013, pictured left) and February 12 (2014) and emailed him and received responses from his staff after both events. Mr Palmer and his staff possess the key information about the fallacy that storing carbon in soil can be the major contributor to the Federal Coalition Government achieving its Direct Action target of 5% greenhouse gas reduction. What will his Palmer United Party [PUP] and other newly aligned Senators do mid-year with their balance of power in the Senate when legislation to repeal the Carbon Tax comes before Federal Parliament?
Clearly, Mr Palmer wants the carbon tax gone but on the Direct Action side of the equation he told the Press Club (February 12) “…we don’t know….we only just set up our party 8 weeks before the election, we didn’t look at enough things in enough detail which we’re looking at right now. And there’s a [PUP] working party looking at Direct Action now and seeing whether or not it will be the right sort of solution and whether something else is required”. PAGE 47 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
That’s a snapshot to the end of February – watch the space to see how, or if, soil science fact gets in the way of national politics in or before June. The nation’s Climate Authority wants a 19% greenhouse gas reduction on 2000 levels by 2020, the business community wants more. In the context of this Soil Change Matters conference, the broader and more relevant story is to ask what farmers, researchers, public and private sector advisory personnel and agribusiness people already know – that is: how to improve urban Australia’s poor understanding of, and lack of connection to, how land managers must maintain, and where possible, improve soil quality for food security and food production as we adapt to climate change? Communication of these messages should be a high priority for all Australian governments to the communities they represent. With world populations exploding, what a substantial survival job that is, given the rhetoric that Australia could become a major world food supplier. Leaving aside political rhetoric about climate policy, no-one here would debate the value of individual landowners doing whatever they could to cost-effectively store as much carbon as their soil permits, to enhance productivity.
How best to reach audiences? Traditional and new media technology now totally cater to metropolitan audiences and readers, and to regional population centres almost as well. But citing NSW as the example, it would be a major blind spot to assume farmers are migrating as rapidly to consume predominantly web-based news and current affairs in preference to print. I’d suggest that anyone who really wants to keep contact with a mass farmer audience, based on the NSW experience, should maintain a solid presence in newspapers and other print formats if you have it, and if you don’t, then get it, at least until there is clear market evidence that high speed internet access has convinced farmers to get online in preference to reading print…. and currently that is a slower process than among metro and regional urban audiences. (There is no debate in this context about the fact that farmers also rely heavily on radio and TV, particularly the ABC, as sources of news and information.) In NSW, the Quantitative Agricultural Research Survey 2012 (QARS) by McNair Ingenuity for Fairfax Agricultural Media (formerly Rural Press) released last December (2014), demonstrated farmers were not yet migrating to the web or social media as their preferred source of news. The QARS revealed the largest single farmer readership group of the print edition of The Land continued to be the 18-39 year olds – last year 90% of them read the paper and preferred print as the delivery medium. In fact the print version of The Land is the favourite way farmers in all other older age groups surveyed also wanted their news presented. They will still be likely to want news via print as a main source in the short to medium term, when Fairfax still expects The Land and corresponding interstate mastheads to remain profitable (even though profits are clearly decreasing). Part of the argument supporting the market dominance of print to this point has been PAGE 48 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
about farmers’ lifestyles. They still regard print as the most convenient format to get their information - they can read before they go out to the paddocks early in the morning, again at lunchtime if they come in, or in the evening. Or take it with them in their vehicle. The other part of the argument is that high speed access to the internet is still poor in many rural areas. So why would the NSW Department of Primary Industries axe, after 20 years of publication, its widest reaching news vehicle, the flagship farm research, advisory and management newspaper, Agriculture Today? Ag Today had published monthly in the print edition of The Land as a highly credible masthead until December 2012 and on the web since 2005. It had approximately 105,000 readers a month. Now, at a time of the largest-ever reforms to the state’s agriculture and natural resource management practices, which will be implemented by the NSW government’s new organisation, Local Land Services (LLS), the LLS website is the go-to source – a visually attractive shopfront but heavy on enthusiastic PR, which as you’d expect has received a high volume of hits. LLS started operating in January (2014), a combination of three previous organisations – the extension (field advisory) staff of NSW Agriculture, the NSW Catchment Management Authorities and the NSW Livestock Health and Pest Authorities. Its internet shopfront talks a big game on the future of land and water management in NSW. It is still way too early to predict the success and substance of the new enterprise and to assess how it will inform ratepayers and the public. To be fair, it’s early days – LLS has been operating 3 months. But as an example, the timeline and advice through Christmas and New Year was so tight for ratepayers to enrol to vote for Board Directors in their regions that in some cases, less than one in 10 people eligible to vote actually enrolled by the deadline. In the case of NSW DPI’s remaining research, biosecurity and other programs’, their main news vehicle is the department’s media release website, www.dpi.nsw.gov.au/aboutus/ news/all At the time budget cuts and rationalisation to create one new organisation and reconstitute another occurred, Ag Today was delivering approximately 350 targeted research and advisory stories a year at a cost to the former NSW DPI of a few cents per message per reader. This cost was less than the cost to DPI of an employee logging onto the internet each day and/or daily use of the NSW Department of Primary Industries Twitter account, before salaries of any media staff users or online content creators were factored in. Google Analytics in late February showed the entire DPI site had 331,000 hits in the previous month but that’s all pages, without knowing how many hits there were on the media releases pages, which essentially target journalists and news organisations rather than the wider “client” base. NSW DPI’s Twitter account by comparison at the end of May 2013 had 2000 followers. At the end of February 2014 it had 2964. Not big numbers. Is there something about government PR messaging that stutters clumsily in this abbreviated “conversation” format? PAGE 49 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Two more anecdotes: Farm Online - the web version of The Land newspaper – had 90,000 hits in January (i.e. one month), according to the general manager at Fairfax Agricultural Media’s North Richmond office, John Dwyer. The Audit Bureau of Circulations puts sales of the print edition of The Land above 40,000 a week, which translates – weekly - to a readership of more than 100,000. In a media release on the NSW DPI’s website dated February 13, according to Dr Cameron Archer AO, Principal at Tocal College, sales of agriculture-related books are booming, defying the shift to digital e-reading. “Sales of books from Tocal have never been higher and are up significantly on last year,” Dr Archer said. “In 2013 Tocal printed and distributed 13,000 publications Australia-wide and internationally on agriculture and related matters. The range of books includes more than 100 titles. In choosing methods of message delivery, one fundamental yardstick – so as to avoid discriminating against some people who will really want to receive your information - is to continually critically assess how fast high speed internet is reaching marginalised rural areas.
References Web newspaper article Aggs R, (October 4 2012) Enough soil carbon to mitigate climate change is a big ask, Agriculture Today, www.dpi.nsw.gov.au/archive/agriculture-today-stories/ag-today- archive/october-2012/enough-soil-carbon-to-mitigate-climate-change-is-a-big-ask TV and web current affairs Cannane, S (April 18 2011) Soil carbon targets difficult to reach, Lateline (ABC TV) www. abc.net.au/lateline/content/2013/s3740395.htm TV and web current affairs Barron J, Ferguson S et al (September 2013) Greg Hunt not giving the full story on climate research, FactCheck (ABC TV), www.abc.net.au/news/2013-08-30/greg-hunt- not-giving-full-story-climate-research/4923258 TV current affairs live interview Jones, T (April 18 2011) Coalition’s direct action plan, Lateline (ABC TV) www.abc.net.au/ lateline/content/2013/s3740397.htm Public forum transcript National Press Club, Canberra, (February 12 2014) National Press Club Address with Clive Palmer, Leader, Palmer United Party, p37 Market survey McNair Ingenuity, (2012) Quantitative Agricultural Research Survey (QARS), Fairfax Agricultural Media (formerly Rural Press), North Richmond Website Local Land Services (2014), www.lls.nsw.gov.au Web news article Bevan, P (February 13, 2014) Agriculture-related books in big demand, NSW Department of Primary Industries, www.dpi.nsw.gov.au/aboutus/news/all/2014/agriculture-books PAGE 50 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
TUESDAY 25 MARCH 2014 - TECHNICAL SESSION 1
A space-time observation system for soil organic carbon S. B. Karunaratne1, T. F. A. Bishop1, J.S. Lessels2, J.A. Baldock3 and I. O. A. Odeh1
1 Faculty of Agriculture and Environment, The University of Sydney, Sydney, NSW 2006, Australia. [email protected] 2 Northern Rivers Institute, School of Geosciences, University of Aberdeen, Aberdeen AB24 3UF, UK. 3 Sustainable Agriculture Flagship, CSIRO Land and Water, Adelaide, Australia.
Abstract In this paper we present a framework for an observation system for soil organic carbon (OS-SOC). The aim of the OS-SOC is to predict soil organic carbon in space and time. Thus, we propose that the RothC model be embedded within the OS-SOC driven by satellite-derived inputs, such as evapotranspiration and biomass additions to soil which characterizes space-time variations in land use and management. Furthermore, instead of fixed model parameters, we propose a Bayesian calibration for estimating the uncertainty associated with the model rate constants. When the model is properly calibrated and implemented, as new observations are acquired then the model simulated values could be updated using data assimilation techniques. Here we present initial results for the implementation of the proposed OS-SOC.
Introduction Modelling changes in soil organic carbon (SOC) has been given more attention in recent times. This is due to the fact that small changes in SOC can have a large impact on the global carbon cycle which could lead to significant changes in atmospheric concentrations of CO2 (Lal, 2004). Furthermore, the decline of SOC in agricultural soil due to improper agronomic practices and changes in climatic conditions may negatively affect productivity since SOC and its associated elements play an integral role in maintaining soil and processes (Baldock and Broos, 2011). Therefore, it is important to monitor the variation of SOC in space and time for effective management and security of agricultural production systems. There are two common approaches in current literature used to quantify the variation of SOC in space and time; (a) statistical approaches; (b) process / mechanistic approaches. In general terms statistical approaches estimate a change in carbon between two sampling points in space and time, and rely on observations. There is potential to combine these two approaches augmented by newly available satellite observed surrogates and here we outline a framework for an observation system for soil organic carbon (OS-SOC) and present a case study from northern New South Wales (NSW). We use the RothC model which has been incorporated into the Australian national carbon account system (NCAS) to simulate changes in SOC (Richards and Evans, 2004). Specific aims of this work are to introduce two components (a) the use of satellite-derived inputs for the RothC model; (b) Bayesian calibration of the rate constants of the RothC model. PAGE 51 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Framework for the Observation system for soil organic carbon In the OS-SOC it is proposed to utilize readily available satellite data and products as weather and land use inputs. For example it is proposed to utilize MODIS derived NPP to get C inputs while weather data derived from satellites such as rainfall (e.g. TRIMM) and ET (e.g. MODIS ET product) could be used in conjunction or instead of Bureau of Meteorology (BOM) weather stations. This approach allows near real-time modelling of SOC using near real-time observations of weather and land management practices. This is particularly useful for management practices as it bypasses the need to continually update land use and cover maps through time providing estimates of biomass inputs through look up tables associated with land uses. An advantage of this approach is that it is scalable from the sub-catchment to the national extent. We also propose the use of Bayesian calibration to estimate the uncertainty associated with the model parameters and inputs, and subsequent predictions. Once the model is calibrated and estimated the posterior distributions (PDF), they are used to carry out simulations which resulting in a distribution of values instead of one value. In addition, once the OS-SOC is calibrated and properly implemented, new observations can be incorporated by updating the model calibration and predictions as necessary. The framework for the proposed OS- SOC is given in the Figure 1.
Figure 1: Framework for the OS-SOC PAGE 52 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Case study – OS-SOC applied to the Cox’s Creek catchment, northern NSW, Australia Methods a) Study area The study was carried out in the Cox’s Creek catchment which is located in northern NSW of Australia. The catchment has a mixture of land uses comprising irrigated agriculture (4%), dry land cropping (35%), pasture (38%) and forest (20%). For the purpose of this case study we will present results for the dry land agriculture. b) Soil data Two sets of soil data were used in this study: the first set was the legacy data collected in 2000 by the NSW state government and the second set were new observations in 2010. To obtain the second set in 2010, the study area was re-sampled using a design-based sampling strategy. For all the samples measurable SOC fractions were predicted using newly developed spectroscopic models under the SCaRP project (Baldock et al., 2013). The newly collected data were mapped using digital soil mapping techniques and values extracted coincident to locations of the year 2000 dataset. The year 2000 dataset was used to initialize the model while year 2010 dataset was used to calibrate the model. In the case of the year 2000 dataset mass preservative splines were fitted to the soil profile data to extract the 0 – 0.3 m depth interval values for the SOC and SOC fractions. The RothC model was initialized with the measurable fractions namely particulate organic carbon (POC), humus organic carbon (HOC) and resistant organic carbon (ROC) fractions predicted by spectroscopic models and used to replace the conceptual RPM, HUM and IOM pools in the RothC model respectively. Our modelling focuses on the top 0 – 0.3 m of soil.
Weather data Interpolated monthly weather grids for rainfall, monthly evaporation (Class A pan), maximum temperature, and minimum temperature were acquired from the SILO database (source: Queensland Climate Change Centre of Excellence) covering the study area from 1970 to 2010 at a spatial resolution of 5 km. Namely two datasets were prepared (a) average annual weather data from 1970 to 2000; (b) monthly weather data from 2000 to 2010 with no averaging. The average weather data were used for equilibrium state runs of the model to derive the required C inputs to match with current measured SOC stock.
Temporal C inputs For this study, estimates of temporal C inputs were derived using temporal Net Primary Product (NPP) data using previous work as described by Smith et al., (2005) (equation 1). NPP , = t PI t PI t� 1 * NPPt�1 (1) where; PIt was the plant C input in the given year (C t ha yr-1), Pt-1was plant C input in the previous year (C t ha yr-1), NPPt was the NPP for the given year (C t ha yr-1) and NPPt-1was the NPP for the previous year (C t ha yr-1). The initial C inputs that were used to reach the measured SOC levels were obtained after running the model in the equilibrium state. Once the optimum C inputs were determined to act as C inputs for year 2000, a time series of C inputs were derived based on Equation 1. As temporal NPP PAGE 53 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
data, the MODIS-derived NPP data (MOD17A3) (https://lpdaac.usgs.gov/products/modis_ products_table/mod17a3) from 2000 to 2010 was used. The MODIS-derived annual NPP data were available at a 1 km spatial resolution. The original units of the NPP were recorded in C kg per m2 and converted to C t ha-1. This method enables the estimation of temporal changes in C inputs spatially across the study area. Performance of the RothC model with temporal C derived inputs (dynamic inputs for 10 years) derived from MODIS NPP were assessed against the C inputs derived using yield and harvest index as described by Skjemstad et al., (2004).
c) Calibration of the rate constants of the RothC model The model calibration was carried out using the recently developed DREAM algorithm (Vrugt et al., 2008) which has been shown to efficiently summaries the posterior distribution of process-based models which do not have a formal likelihood function. As the likelihood function root mean squired error (calc.rmse) implemented within the DREAM R package was used (Guillaume and Andrews, 2012). Since no prior distributions were available the ranges suggested by Scharnagl et al., (2010) for model rate constants were used. Simulation of SOC using RothC model The estimated posterior distributions for the respective rate constants were used to carry out the simulations. Once the simulations were carried out with entire posterior distributions of the considered rate constants, it was possible to calculate the uncertainty due to the rate constants. The 95 % confidence interval around the mean for simulated SOC was calculated from year 2000 to year 2010. As a result, it was possible to estimate the uncertainty related to rate constants from the start of the simulations to its end.
Results and Discussion Summary statistics Out of three measurable fractions, the highest mean was reported for HOC (26.65 C t ha-1) followed by ROC (17.95C t ha-1) and POC (6.84 C t ha-1). Performance of RothC model – dynamic C inputs derived from MODIS NPP vs. C inputs derived using yield and harvest index Since the simulated results were skewed we calculated the root median square error (RMeSE). We found that the model where C inputs derived from MODIS NPP recorded RMeSE of 9.27 C t ha-1 while value of 12.49C t ha-1 for the model where C inputs were derived based on yield and harvest index. This highlights the advantage of using MODIS derived C inputs when running the RothC model in the spatial context. Estimation of the posterior distributions of the rate constants The mean estimated rate constants are very close to default values recommended for the RothC model ( Coleman and Jenkinson, 1999). However there is a wide variation in the estimated posterior distributions for the rate constants. For example the 95 % confidence interval for the DPM rate constant for dry land agriculture ranges from 2.8721 year-1 to 17.1284 year-1. Therefore, the assessment of simulation uncertainty using this entire range of estimated posterior distributions for the rate constants is important. Figure 2.a depicts the estimated posterior PDF for each rate constant for dry land agriculture. The entire PDF were used to carry out the simulations from year 2000 to 2010. Figure 2.b depicts the simulation uncertainties associated with the rate constants. The red line for each site indicates the mean simulated value while grey area indicates the upper and lower boundaries of the 95 % confidence interval for the simulations (Figure 2.b). PAGE 54 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Conclusions In this case study we demonstrated some of the functionalities of the proposed OS- SOC and quantified the uncertainties related to RothC model rate constants when the model was run spatially across a catchment. We further demonstrated how to (a) obtain posterior distributions of the model rate constants; (b) use of those estimated posteriors distributions to carry out simulations; and (c) obtain C inputs to the soil based on NPP which was derived from the MODIS satellite product. The use of MODIS-derived C inputs via NPP reflects the effect of other abiotic and biotic factors which effect biomass production (e.g. drought). Therefore, it will give more reliable estimates of C inputs to the soil especially in space and time. Further work could extend the approach to modelling subsoil C using the RothPC-1 model (Jenkinson and Coleman, 2008) and further quantify the uncertainty associated with other aspects of the framework, for example, the spectroscopic estimates of C fractions or biomass inputs. Finally, we believe the initial results presented here hold much promise for a functional OS-SOC especially in a future where soil data becomes easier to obtain and remote sensing platforms improve their spatial, temporal, radiometric and spectral resolutions.
References Baldock JA, Broos K (2011) Soil Organic Matter In: Huang PM, Li Y, Sumner ME, (Eds.), Handbook of Soil Sciences: Properties and Processes. CRC Press/Taylor Francis Group, pp. 11.11-11.52. Baldock JA, Hawke B, Sanderman J, Macdonald L (2013) Predicting contents of carbon and its component fractions in Australian soils from diffuse reflectance mid-infrared spectra. Soil Research (accepted). Coleman K, Jenkinson DS (1999) ROTHC-26.3. A Model for the Turnover of Carbon in Soil. Model Description and Windows User’s Guide. Rothamsted Research Harpenden Herts AL5 2JQ Guillaume J, Andrews F (2012) dream: DiffeRential Evolution Adaptive Metropolis. R package version 0.4-2. Jenkinson DS, Coleman K (2008) The turnover of organic carbon in subsoils. Part 2. Modelling carbon turnover. European Journal of Soil Science 59, 400-413. Lal R (2004) Soil carbon sequestration to mitigate climate change. Geoderma 123, 1-22. Richards GP, Evans DMW (2004) Development of a carbon accounting model (FullCAM Vers. 1.0) for the australian continent. Australian Forestry 67, 277-283. Scharnagl B, Vrugt JA, Vereecken H, Herbst M (2010) Information content of incubation experiments for inverse estimation of pools in the Rothamsted carbon model: A Bayesian perspective. Biogeosciences7, 763-776. Skjemstad JO, Spouncer LR, Cowie B, Swift RS (2004) Calibration of the Rothamsted organic carbon turnover model (RothC ver. 26.3), using measurable soil organic carbon pools. Soil Research 42, 79-88. Smith J, Smith P, Wattenbach M, Zaehle S, Hiederer R, Jones RJA, Montanarella L, Rounsevell MDA, Reginster I, Ewert F (2005) Projected changes in mineral soil carbon of European croplands and grasslands, 1990-2080. Global Change Biology 11, 2141-2152. Vrugt JA, terBraak CJF, Clark MP, Hyman JM, Robinson BA (2008) Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation. Water Resources Research 44, W00B09. PAGE 55 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Modelling soil carbon in Australia: Cause for optimism Garry O’Leary 1, De Li Liu 2, James Nuttall 3, Muhuddin Rajin Anwar4 Fiona Robertson 5
1 Department of Environment and Primary Industries, PB 260 Horsham Vic, 3400, garry.o’[email protected] 2 Department of Primary Industries, Wagga Wagga Agricultural Institute, NSW 2650, [email protected] 3 Department of Environment and Primary Industries, PB 260 Horsham Vic, 3400, [email protected] 4 Department of Primary Industries, Wagga Wagga Agricultural Institute, NSW 2650, [email protected] 5 Department of Environment and Primary Industries, Mt Napier Rd Hamilton VIC 3300, [email protected]
Abstract Modelling is widely applied to explore practicable and theoretical options in various scientific disciplines. The very early modelling attempts of engineers to model the flow of mass and/or energy meant that the engineering disciplines like civil engineering, aerospace and mathematics nowadays have modelling as an integral part of any serious research enquiry. There are other disciplines that really need to catch up and of course there are others such as soil science that the best of modelling is yet to be shown. Simulation models provide robust and objective methods to extrapolate likely responses of crops and soils to climate change over different landscapes and time periods. Central to such simulation models are the soil carbon dynamics and suppling mineralised nutrients to crops through its linked crop and nutrient submodels. Free Air CO2 Enrichment (FACE) experiments have shown that crop yields are increased at elevated atmospheric CO2 concentrations, whilst nutrient levels (e.g. grain nitrogen concentration) are usually decreased. Progressive nitrogen limitation under elevated CO2 environment is likely to further limit grain protein in crops. Attention to soil processes therefore is an essential part of building robust production systems. To the farmer, focus must be on the productive capacity of the land and its rejuvenation to sustain production beyond the present. In a broader context of reducing atmospheric CO2 through soil C sequestration understanding soil processes and the immediate environment require attention to productivity issues. This is because without maintaining productivity a better understanding of soil C processes is unlikely to lead to increased soil C sequestration in Australia’s farming land. Modelling may also provide the opportunity to identify the best opportunities to increase soil organic carbon (SOC) within specific regions and agro-ecological zones. Some gaps in our knowledge of how to manage SOC are being addressed in a national research effort. To close such gaps, a focus on the scant measured data against which models can be tested is necessary. Nevertheless, continuing to apply models pushes the boundaries well beyond what can be achieved experimentally. Whilst this can be seen as problematic it does offer optimism for the future and is a strong advantage for modelling. PAGE 56 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
2.b Uncertainty of simulated changes in SOC stocks based on the posterior distributions of 2.a Estimated posterior distributions the rateconstants (plots are given for 8 sites) for the model rate constants
Experimentation The measurement of components of the soil involved in soil C dynamics is fundamental to any modelling effort. Standards need to be set so that others can repeat the observations. Hence the national focus on standardised measurement is crucial to advancing our knowledge and management of soil carbon. However, we don’t have the luxury of measuring what we would like. It’s too costly. But we can measure some things and an example is some of the work of Sale et al. (2012) where organic C is placed into the soil with resultant changes in soil properties and improved productivity. Is this cause for optimism or pessimism? Certainly it’s possible to be quite optimistic but it is also possible to pessimistically dismiss it on economic or life-cycle grounds. Reliance on long-term experiments containing irrelevant management practices may also be missing the mark.
Modelling Modelling offers the exploration of practices that have not been robustly tested. For example, soil carbon levels in a long term experiment have been observed to decrease under stubble retention on the soil surface in a semi-arid environment (Heenan et al. 2004). Subsequent simulation showed that such practice is only equivalent to 24% incorporation of wheat stubble into the soil (Liu et al 2009). Recent work from Liu et al. (2014) shows that it appears possible to increase SOC in rainfed wheat cropping soils of south eastern Australia by incorporating some threshold amount of wheat stubble into the soil. Whilst the suggested thresholds appear quite large at 50-70% practical methods need to be developed and tested to see if this can be achieved in the field. Liu et al. (2014) goes on to argue that if such levels of incorporation can be achieved in practice then large gains could be made that potentially could assist Australian meet is emission reduction targets. That modelling study proposed an optimistic approach to soil carbon management unlike another modelling study (Robertson and Nash 2013) that relied on historic practices where little scope was seen to increase soil C sequestration in Australia. As part of the national DAFF program of “Filling the Research Gaps” our team is setting out to document the performance of various soil carbon models that can be used to explore ways to profitable manage SOC. In this way experimental data is crucial to help predict accurately changes in soil C in current long-term experiments. This quest however, has been fraught with difficulties of data collation and reformatting for use PAGE 57 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
with models. In looking over some old data sets we have noticed that those that have already been modelled in the past (e.g. Hermitage, QLD) that the overhead of setting up input data files is much lower than those in multiple and typically non-modelling friendly formats.
A way forward From a modelling perspective we see that the performance of our contemporary models needs to be clearly documented. Whilst the deficiencies in model design will be apparent to some, the measured performance is an objective way to access the extent that any particular model can be relied on. It helps in both a scientific and legal way that any models are tested against measured data and the accuracy articulated. It was a surprise to see national soil C modelling efforts without objective measurements of accuracy (e.g. Skjemstad et al. 2004). With custodians of experimental data we plan to make a number of long-term data sets involving soil C available in a format that can be rapidly applied to modelling studies. We think that the Agricultural Production Simulator (APSIM) framework is best for this purpose as it allows a large suite of farming systems including grazing systems to be explored in a modelling context. Farming systems that have high productivity in terms of the accumulation of biomass that can find its way into the soil are clearly ones that will feature more in future farming systems. Whilst farming systems in the higher rainfall areas offer such potential modelling opportunities involving such systems are even greater and offer optimism for raising SOC levels in the Australian landscape. References Heenan, DP, Chan, KY, Knight, PG (2004) Long-term impact of rotation, tillage and stubble management on the loss of organic carbon and nitrogen from a Chromic Luvisol. Soil and Tillage Research 76, 59–68. Liu, DL, Chan, KY, Conyers, MK (2009) Simulation of soil organic carbon under different tillage and stubble management practices using the Rothamsted carbon model. Soil and Tillage Research 104, 65–73. Liu, DL, Anwar, MR, O’Leary, G, Conyers, MK (2014) Managing wheat stubble as an effective approach to sequester soil carbon in a semi-arid environment: Spatial modelling. Geoderma 214-215, 50-61 (http://dx.doi.org/10.1016/j.geoderma.2013.10.003). Robertson, F, Nash, D (2013) Limited potential for soil carbon accumulation using current cropping practices in Victoria, Australia. Agriculture Ecosystems & Environment 165, 130–140. Sale, P, Gill, J, Peries, R, Tang, C (2012) Subsoil manuring on problem clay soils: increasing crop yields to the next level. In: Capturing Opportunities and Overcoming Obstacles in Australian Agronomy. Edited by I. Yunusa. Proceedings of 16th Australian Agronomy Conference 2012, 14-18 October 2012, Armidale, NSW, Australia (http:// www.agronomy.org.au). (http://www.regional.org.au/au/asa/2012/soil-water- management/8105_salepw.htm). Internet address verified 3 February 2014. Skjemstad, JO, Spouncer, LR, Cowie, B, Swift, RS (2004) Calibration of the Rothamsted organic carbon turnover model (RothC ver. 26.3), using measurable soil organic carbon pools. Australian Journal of Soil Research 42, 79-88. PAGE 58 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Responses of soil carbon stocks to climate variability and extremes: future projections using RothC in an Australian agricultural landscape Mohsen Forouzangohar1*, D. Dugal Wallace2, Craig R. Nitschke 1, Lauren T. Bennett1
1 Department of Forest and Ecosystem Science, The University of Melbourne, 4 Water Street, Creswick, Victoria 3363, Australia. 2 Agriculture Productivity Group, Department of Environment & Primary Industries, 32 Lincoln Square, North Carlton, Victoria 3053, Australia. * [email protected]
Abstract Climate variables are known as primary factors controlling the balance of soil organic carbon (SOC) stocks. Soils are expected to lose SOC in response to global warming. Therefore, research is currently directed towards predicting the possible responses of SOC stocks to changes in average climate under various scenarios of climate change. What is underappreciated in this context is the likely significant impact of climate extremes and inter-annual variability on the long term balance between SOC and atmospheric CO2. In this study, we applied RothC for simulating SOC trends in response to scenarios of variable and extreme climate in a primarily agricultural landscape over 50 years. We found that climate variability, in general, and extreme rainfall, in particular, presents risks to the balance of SOC stocks by facilitating greater microbial decomposition. RothC simulations predicted that, on average, soils will lose 0.24 t/ha SOC under extreme wet years and will gain 0.1 t/ha under extreme dry years. Overall, it was estimated that SOC will decline by 11-13% under tested variable climates. Excess rainfall appeared to be overriding the effects of variability in average temperature in RothC simulations. Our results suggest the importance of accounting for added water in SOC estimations within irrigated systems. We concluded that climate variability and extremes are important elements for predicting future SOC under climate change projections, and should be incorporated in the modelling of SOC. PAGE 59 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Digging deeper: does it matter when assessing soil changes? Denis A. Angers1, Alex B. McBratney2, Budiman Minasny2
1 Agriculture and Agri-Food Canada, 2560 Hochelaga Blvd., Quebec City, QC, G1V 2J3, Canada, [email protected] 2 The University of Sydney, Biomedical Building C81, Suite 401, 1 Central Avenue, Eveleigh, NSW 2015
Abstract Although management-induced changes have long been considered to only take place in the surface soil layer (Ap horizon), recent studies show significant soil change deeper in the profile, even in the short term. Soil organic carbon (SOC) will be used to illustrate such changes. Despite their low C concentration, the subsoil horizons below the A horizon may represent a high proportion of SOC stored in the soil profile. Estimates vary from 40 to 60% at the global scale. In the past few years, our understanding of deep SOC has significantly improved but its potential for change under varying management practices remains largely unknown. Their usually low C concentration suggests that subsoil horizons may not be saturated in organic C, and thus offer potential for additional storage. However, the relatively stable SOC at depth may become available to microbial decomposition following labile C input. Moreover, conditions of decomposition such as water and nutrient availability and temperature may vary with depth, and either favour or slow down decomposition. Carbon inputs at depth can originate directly from plant roots and their exudates, or C can be transferred by leaching of soluble C, the burrowing action of biota or mechanically by tillage. Field examples of changes in SOC at depth induced by management practices will be used to illustrate some of these effects and the possible mechanisms at play.
Estimate soil erosion to attribute carbon storage to management practice Adrian Chappell1, Jeff Baldock2, Raphael Viscarra Rossel3, Jonathan Sanderman4
1 CSIRO Land & Water, GPO Box 1666, Canberra ACT 2601, [email protected] 2 CSIRO Land & Water, PMB 2, Glen Osmond, SA 5064, [email protected] 3 CSIRO Land & Water, GPO Box 1666, Canberra ACT 2601, [email protected] 4 CSIRO Land & Water, PMB 2, Glen Osmond, SA 5064, [email protected]
Abstract Soil erosion removes preferentially the fine, nutrient- and carbon-rich material from the soil surface which reduces its fertility and moisture holding capacity, increases its susceptibility to erosion and deteriorates the soil condition. Soil erosion dominates soil degradation and has caused a global deterioration in agricultural land, reduced its total productive area and contributed to greenhouse gas emission. For Australia, the long- term (1950s-1990) mean annual erosion of cultivated land is (-1.26 t ha-1 y-1) five times larger than uncultivated land. Although widely adopted soil conservation in the 1980s has likely caused an overall reduction of erosion in south-eastern Australia, erosion remains highly spatially variable. Dust emission continues unabated (between 2000- PAGE 60 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
2011) from this ‘dry continent’ and we estimate SOC dust emission of 1.59 Tg C ha-1 y-1. Unfortunately, soil erosion is excluded from the national carbon accounting system (NCAS) and the carbon farming initiative (CFI) which increases uncertainty and diminishes accuracy. Here we show how soil erosion can hide the true SOC sequestration potential and cause management practices to appear ineffective at storing carbon. We discuss a cheap and practical method to estimate soil erosion as part of the CFI and show how the inclusion of soil erosion in carbon cycling reduces CO2 emission.
Quantifying uncertainty and optimising predictions of soil carbon content and composition with mid-infrared spectroscopy. JA Baldock1 and B Hawke1
1 Sustainable Agriculture Flagship, CSIRO Land and Water, Adelaide, Australia.
Abstract Interest in soil orgnaic carbon (and its assocatied elements) exists due to the positive contribution it makes to many soil properities and processes and its potential to mitigate greenhouse gas emissions. Understanding the composition of soil organic carbon can provide important insights into its contribution to soil processes as well as its vulnerability to subsequent change. The recent soil carbon research program (SCaRP) built on previous work conducted in Australia to demonstrate the utility of mid- infrared spectroscopy when combined with partial least squares regression (MIR/PLSR) to provide rapid and low cost predictions of carbon content and composition. The SCaRP also demonstrated an improved accuracy of prediction when regional rather than national algorithms were produced. In this study we have investigated the ability of applying a locally weighted regression approach to the SCaRP spectral library in an effort to further improve the accuracy of MIR/PLSR predictions of soil carbon content and composition. We will also discuss the quantification of uncertainty assocaited with MIR/PLSR predictions. The presentation will be concluded with an assesment of what is required for Australia to develop a nationally consistent approach to the use of MIR/PLSR for predicting soil carbon content and composition to support the National Geenhouse Gas Inventory and the Carbon Farming Initiative. PAGE 61 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Environmental cost of two critical land use changes in Russia during last century Kurganova I. and Lopes de Gerenyu V.1
1 Institute of Physicochemical and Biological Problems in Soil Sciences of the Russian Academy of Sciences, Institutskaya street, 2, Pushchino, Moscow region, 142290, Russia, [email protected]
During last century, two critical changes in land use caused by the abrupt shifts in agricultural policy took place in Russia. Between 1954 and 1963, about 16.3 million ha of virgin lands were involved in agriculture. It was the largest land expansion in steppe regions of the eastern Russia and resulted in negative outcomes for the environment: soil erosion and degradation, deflation, disappearance of steppe landscapes, decrease of biodiversity, and climate aridization. The total loss of carbon (C) for the first 10 years owing to the intensive mineralization of soil organic matter exceeded 27 Tg C per year; this might result in stimulation of the greenhouse effect. After 1990 due to the collapse of collective farming in Russia, over 45 million ha of arable lands were abandoned. The withdrawal of croplands led to several benefits including carbon sequestration and increase of biodiversity of post-agrogenic ecosystems. The average C accumulation rate in the upper 20 cm of mineral soil was 0.96±0.08 Mg C/ha/yr for the first 20 years after abandonment and 0.19 ± 0.10 Mg C/ha/yr during the next 30 years of post-agrogenic evolution and establishment of natural vegetation. The amount of C sequestered over the period 1990-2009 accounted to 42.6±3.8 Tg C per year. This C sequestration equals to about 10% of the current fossil fuel emissions of Russia or 4% of global CO2 release due to deforestation and other land use disturbances. Therefore, two crucial land use changes have induced the significant outcomes for the global environment. PAGE 62 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
A base function rather than a baseline for soil organic carbon in a variable landscape? Ben Jones1, Michael Moodie2, Narelle Beattie3
1 Mallee Focus, Melbourne, Vic, 3000, [email protected] 2 Mallee Sustainable Farming, 2/152 Pine Avenue, Mildura Vic 3500, [email protected], 3 Mallee Catchment Management Authority, P.O. Box 5017, Mildura, Vic, 3500, [email protected]
Abstract Establishing a baseline for soil carbon across a catchment is a conceptual challenge, especially when soil types are highly variable. An aggregate measurement can be made across the catchment, but which soil does it apply to? This study explored several approaches to calculating soil carbon baselines for the Mallee Catchment Management Authority region, covering over 2.4 million hectares of agricultural land in the low rainfall Victorian Mallee. Soil organic carbon was measured at 155 one hectare focus sites over three years, across the region. Several possible co-factors were used (clay content, land system, land class, rainfall) to derive base functions for topsoil (0-10cm) organic carbon. Soil organic carbon varied from 0.17 to 2.2% across all sites (av. 0.79%). Soil organic carbon fitted a power relationship of clay content quite well (R2=57.8%) and remaining factors were tested to see if they improved on this. Geographic coordinates and sample transect fitted best; landform and land system improved little on clay content. Growing season (April-October) rainfall over recent years (2006-8) fitted almost as well as coordinates, however older measures (historic; preceding 10 years) fitted poorly. Using the data from this program several possible ‘base functions’ were constructed and would allow a specific ‘baseline’ to be calculated for a soil anywhere in the catchment, given clay content. The functions may be too complicated for general use but could readily be delivered in an online calculator. This project was supported by the Mallee Catchment Management Authority through funding from the Victorian Government. PAGE 63 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
TUESDAY 25 MARCH 2014 - TECHNICAL SESSION 2
Delivering solutions to questions regarding soil change – examples from the USDA National Cooperative Soil Survey Maxine Levin1, Susan Andrews, Michael Robotham, Lenore Vasilas and David Hoover, USDA-Natural Resources Conservation Service, National Soil Survey Center, Lincoln, NE
1 USDA Natural Resources Conservation Service, [email protected]
Abstract The US National Cooperative Soil Survey has been grappling specifically with the subject of Soil Change within the Interpretations Conference Committees and within the USDA- Natural Resources Conservation Service, Soil Science Division since the early 1990’s. Historically the US Soil Survey worked around questions of soil change to focus on building a national map, county by county, that would answer questions about land use and soil management (1). In the US, all these assessments of the potential uses of soils, from agricultural production to engineering properties, have become consolidated under the overarching category of Soil Interpretations. Soil Mapping, Soil Classification and Soil Taxonomy focused on the static qualities of the soil profile, attempting to make estimates and predictions of soil groupings based on soil characteristics that were stable beyond a 5 to 20 year cycle of use and management and potential anthropogenic change. From the early years of the soil survey through the development of computer databases in the 1970s, soil interpretations were based on written guides that were used by the soil surveyors to develop the interpretation tables contained in soil survey manuscripts published by the National Cooperative Soil Survey (NCSS). Tables contained use and management interpretations by map unit or component. Interpretive results for a tract of land could be determined by cross referencing the hard copy soil map and the interpretive table(s). Examples of common interpretations provided in soils surveys include: crop, forage and range suitability groups, and use limitations for recreation, building site development, and engineering uses. The question of soil change centering on the potential for erosion, salinity and sodicity, wetland drainage phases and soil contamination has been the realm of soil interpretations and trying to tie the map unit and individual soil characteristics to suitability or vulnerability indexes in separate tables or maps. In the mid 1980’s through 2000 the National Cooperative Soil Survey started to explore how we might map the soils to display more information by soil function and the capacity to recover from various anthropogenic forces. Farm and rangeland areas with drastic soil disturbances from native landscape were being mapped and classified within the boundary rules of soil classification, not their original genetic sources. The Survey also started mapping urban areas where all the soils in the survey area had anthropogenic characteristics that needed to be described, classified and interpreted for managed use. Soil change issues within policy forced us to build new systems of classification and description. The best example of this was Hydric soil indicators for wetland delineation as part of the Clean Water Act. Disturbed functioning wetlands and PAGE 64 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
drained systems were being regulated by other agencies and needed to be delineated and documented by interagency and disciplinary groups to be substantiated in courts. The National Cooperative Soil Survey partnership worked in hand with the established maps creating potential wetland maps with hydric soil lists but deferred regulation to onsite examination of hydric soil indicators to confirm active function of wetlands for policy management and regulatory statutes (2). In addition to the example of hydric soil policy implementation, there were examples for highly erodible land with wind or water soil erosion potential, salinity vulnerability and reclamation and subsidence vulnerability. The NCSS continues to identify ways to improve existing products and develop new products. Next steps for further work will be an increased focus on soil monitoring through National Resource Inventory studies, the Ecological Site Inventory and modeling to identify soil change and vulnerable landscapes in time (refer to presentation by Susan Andrews on Soil Change, Resistance, Resilience, State and Transition Models and dynamic Soil Property studies for details of new research and information platform for data;also refer to presentation of Lee Norfleet on Modelling, CEAP studies to identify carbon and soil change functions over time). The ongoing activities include improving the consistency of the soils data that underpins interpretations across political boundaries, developing a “minimum data set” of commonly used interpretations to include in (SSURGO), and working with local cooperators to assess and improve existing interpretations. The NCSS continues to seek new ways interpret soils and to make that information more easily accessible. As an example, practice specific soil interpretations are being explored to support NRCS conservation planning and practice implementation. Other areas of focus include the development of real-time interpretation systems that allow incorporation of site-specific information, and interpretation systems that will allow users to incorporate other spatially-referenced data sets including climate and landuse, to develop accurate and site-specific interpretive information products. On the distribution side, efforts are underway to look at how interpretive information can be effectively delivered through other avenues including via smart phones and tablets. All of these examples could lead to more effective sampling and interpretation of soil information to support monitoring and soil change attributes. By building on its long history and record of success in interpreting soils information to meet user needs and incorporating new ideas and technology, the Soil Science Division, the NRCS and the National Cooperative Soil Survey partnership are well positioned to continue to deliver interpretive information that meets customer needs now and on into the future.
References Soil Survey Staff, Natural Resources Conservation Service, United States Department of Agriculture. Soil Survey Geographic (SSURGO) Database for [Survey Area, State]. Available online at http://websoilsurvey.nrcs.usda.gov/. Accessed [2014]. Soil Survey Staff, Natural Resources Conservation Service. 2014. National soil survey handbook. Title 430-VI. U.S. Government Printing Office, Washington D.C. Sec 602. PAGE 65 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Land Management within Capability, a new scheme to guide sustainable land management in New South Wales, Australia Jonathan Gray1, Greg Chapman2, Brian Murphy3
1 NSW Office of Environment and Heritage, PO Box 3720, Parramatta NSW 2124 AUSTRALIA, [email protected] 2 NSW Office of Environment and Heritage, PO Box 3720, Parramatta NSW 2124 AUSTRALIA, [email protected] 3 NSW Office of Environment and Heritage, PO Box 445, Evans St, Cowra NSW 2794 AUSTRALIA, [email protected]
Abstract A new Land Management within Capability evaluation scheme, used to guide sustainable land management in New South Wales, Australia, is presented. The scheme quantifies the potential impacts of specific land management actions and compares these with the inherent physical capability of the land in relation to a range of land degradation hazards. This leads to the derivation of Land Management within Capability (LMwC) indices, which rate the sustainability of land management for individual sites up to broader spatial entities such as catchments. The scheme can be used to identify particular land management actions that place the land at greatest risk of degradation and therefore need to be addressed. The greater the degree of land management within capability operating at a site, then the greater is its resilience to withstand potentially degrading processes, and therefore ongoing sustainability. Results from the scheme can help to guide natural resource agencies at local, regional and State levels to target priorities and promote sustainable land management across their lands. It has particular application for regional soil condition monitoring programs, given that regional monitoring of land management is usually more practical and cost effective than monitoring of actual soil condition. Few other schemes that assess the sustainability of a given land management regime in a semi- quantitative manner are reported in the literature.
Introduction Land degradation is a consequence of failure to manage land in accordance with its inherent capability. Land capability refers to the inherent physical capacity of the land and its soils to sustain a range of land uses and management practices in the long term without degradation to soil, land, air and water resources (Dent & Young 1981). Land degradation may have impacts both on- and off-site, and ultimately leads to a reduction in the capacity of the land to deliver ecosystem services including agricultural productivity. Decisions on land management practices such as the intensity of tillage, length of bare fallow and maintenance of ground cover will all determine whether a site is being used sustainably and the likelihood of land degradation occurring. The management of land within its inherent capability is vital for sustainable land use. It is important to know the degree to which land is currently being managed within its inherent capability over different regions and identify particular issues or locations of greatest concern. Such information can be a powerful guide to correcting unsustainable land management practices throughout a region. An evaluation scheme that can simply and effectively provide such information is required. PAGE 66 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Most land evaluation schemes used in Australia and globally (Rowe et al 1988, van Gool 2008, OEH 2012, USDA 1983, FAO 1976) tend to deal with the physical characteristics and limitations of the land, with only broad reference to land use and management issues. Specific land management criteria have however been directly incorporated into a few schemes, including the FAO Framework for Evaluating Sustainable Land Management (FESLM) (FAO 1993) and the Victorian Land Use Impact Model (LUIM) (McNeill & MacEwan 2007). Various quantitative modelling and simulation systems incorporate the influence of different farm management practices with environmental factors, such as APSIM (Keating et al 2003). The Land Management within Capability (LMwC) scheme presented here represents a new approach to land evaluation that specifically incorporates land management data. It was developed as part of a recent monitoring, evaluation and reporting (MER) program carried out over the state of New South Wales (NSW), Australia (Gray et al 2011). This paper presents an outline of the scheme.
The Land Management within Capability scheme The LMwC scheme quantifies the potential impact of specific land management actions and compares this with the capability of the land in relation to a range of land degradation hazards. The scheme recognises that certain activities can have a large adverse impact on the soil and land, whilst others have lesser impact. Those activities with a large adverse impact require land of higher capability in order to be sustainable. Conversely, land management actions with lesser adverse impact can be carried out on a wider range of land types, including land with lower capability. The LMwC methodology involves a comparison of the estimated impact of current land management actions against the physical capability of the land and soil to derive LMwC indices. The resulting process as applied in the NSW MER project is described below and summarised in Figure 1. PAGE 67 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 1: The Land Management within Capability assessment process
Steps 1 & 2: Collect data Soil and land data were collected from each site within each soil monitoring units (SMU), these being areas of broadly uniform soil-landscape character and land management requirements. There were 10 SMUs for each catchment region. Details of land management practices implemented over the site were collected from the landholder, using a specially designed questionnaire. PAGE 68 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Step 3: Determine the capability of land The capability at each site was evaluated using the rules tabulated in the Land and Soil Capability (LSC) system. The LSC system is a comprehensive capability assessment scheme recently developed by the NSW natural resource agency (OEH 2012). It involves a series of rules to allocate biophysical land characteristics into one of eight classes for each of a range of land degradation hazards such as sheet erosion, wind erosion, acidification and structure decline. Each indicator is assessed separately. In the scheme, LSC Class 1 refers to land of highest capability, that is, least susceptible to degradation; while Class 8 refers to land with the lowest capability, that is, the most susceptible to degradation. The overall LSC rating for any site is taken as the worst capability rating, or the limiting factor, of any of the component hazards. Site data was applied to the rule sets, together with regional climate and other datasets, to derive individual LSC values for each land degradation issue at each site.
Step 4: Evaluate land management actions A framework was developed that considers the potential impact of a range of land management actions on the individual land degradation hazards that comprise the LSC classification (eg, sheet erosion, structure decline, etc). This allows individual actions at a site to be rated as having a low to very high impact on soil condition and allocated a corresponding “upper sustainable LSC class” (see examples in Table 1). In general, the higher the impact, the better the capability of the land must be for the activity to be practiced sustainably. The combined influence of each action was simply averaged to give the upper sustainable LSC class for each hazard ie, the rating of sustainability of management practices for that hazard for that site. PAGE 69 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS Upper Upper Sustain. LSCMGT ------Acidification Impact ------Upper Upper Sustain. LSCMGT 4 2 2 1 1 1 Impact Structure decline Structure M VH VH EH EH EH Upper Upper Sustain. LSCMGT 4 3 2 2 1 1 Wind erosion Impact M H VH VH EH EH Upper Upper Sustain. LSCMGT 4 3 2 2 1 1 Impact Gully erosion M H VH VH EH EH Upper Upper Sustain LSCMGT 4 3 2 2 1 1 Impact Sheet erosion M H VH VH EH EH
Specific action 0 1 2 3 4 >4 Land management practice high (EH) high (VH); Extremely (M); High (H); Very Moderate Tillages prior to sowing Table 1: Sample derivation of upper sustainable LSC class of 1: Sample derivation Table Step 5: Compare upper sustainable LSC with actual LSC of site for each hazard site for upper sustainable LSC with actual of Step 5: Compare PAGE 70 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The upper sustainable LSC class for each hazard was compared with the actual LSC class of the hazard at each site and the difference between the two derived (see Figure 2). Positive values indicate that the combined management actions are within the capability of the land for that hazard (sustainable state in Figure 2), negative values indicate management actions are beyond capability (unsustainable state of Figure 2). For example, a hot burn of stubble followed by long bare fallow will have a very high impact on sheet erosion, thus these practices are restricted to land with high capability for sheet erosion. Step 6: Derive LMwC indices for each hazard The LMwC index was calculated for each hazard by considering the results from the above comparison, using the rules given in the lower part of Figure 2. For example, where a site hazard is being managed at greater than 2 units within capability it has an LMwC index of 5 (very high sustainability), whereas if it is being managed at more than 1 unit beyond capability, it has in index of 1 (very low sustainability ).
Figure 2: Derivation of LMwC indices Steps 7 & 8: Derive LMwC indices for site, soil monitoring units, catchments and beyond The LMwC indices for each hazard at a site were combined to give the overall LMwC summary index for that site. These indices were then combined to give both overall and hazard specific LMwC indices for relevant spatial entities, such as the soil monitoring unit, the catchment and ultimately the entire province or State. The LMwC indices provide a broad indication of the level of sustainable land management over each of these entities.
Discussion The LMwC results provide useful land management information at the soil monitoring unit (SMU), regional and State levels. Results were presented in plots for each regional catchment in the 2010 NSW State of Catchment reports. Current data is presented in the 2012 NSW State of Environment report (EPA 2012) and in Gray & Chapman (in press). PAGE 71 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
For each hazard, data is provided on the overall LMwC index, the range of indices, the apparent trend, SMUs of concern (ie, those with an LMwC index of <3.0 for that hazard) and data confidence levels. Similar results were presented on the basis of SMUs, regional catchments and the State as a whole. The current data suggest that on a state wide basis, land in NSW is overall being managed at a level in accordance with its inherent physical capability, with an LMwC index of 3.6, however there are widespread issues of concern, with individual hazards being unsustainably managed over many areas. It has been found that over 50% of soil monitoring units examined across the State have a “poor” or “very poor” rating (with LMwC indices <3) for at least one hazard. In these areas, there is a risk of ongoing land degradation from particular hazards that are currently not being sustainably managed (EPA 2012). Outputs from the Land Management within Capability process highlight particular hazards and geographic areas of concern across individual regions. It provides a relative rating of the degree to which land is used or managed in comparison with its inherent physical capability, ie, the level of sustainable land management. Adoption of land management within capability strategies at local and regional scales will significantly contribute to achieving resilience in rural land use systems. The process can be used to identify particular land management activities that are contributing to degradation and need to be corrected. For example, in a particular region a low LMwC index (poor management) may have been identified in relation to acidification. Further data analysis may reveal over use of nitrogen fertilisers or over- irrigation as primary causes of the problem. This information can then help to direct future education and extension programs. It may serve as a useful tool for the ongoing monitoring of soil and land resources, given the expense and difficulties of on-site physical soil measurements (McKenzie 2008) and the greater ease of collection of land management data. More generally, results can help to guide regional and State Government natural resource agencies on priorities in promoting sustainable land management across their lands. Further development of the process will improve its reliability and may lead to a rapid field assessment procedure to guide sustainable management at the individual property level. The incorporation of easily accessible land management data collected by statistical agencies such as Australian Bureau of Statistics’ Land Management and Farming in Australian (ABS 2010) will further enhance the potential of this scheme to promote sustainable land management at regional, State or national levels.Heritage Trust is acknowledged.
References ABS (2010) Land Management and Farming in Australian 2010 Australian Bureau of Statistics, http://www.abs.gov.au/ausstats/[email protected]/mf/4627.0, accessed 1.3.2013. Dent D, Young, A (1981) Soil Survey and Land Evaluation, George Allen and Unwin, LondonEPA 2012 FAO (1976) A Framework for Land Evaluation, Soils Bulletin 32, Food and Agricultural Organisation of the United Nations 1976, Rome. FAO (1993) FELSM: An international framework for evaluating sustainable land management, World Soil Resources Report 73, Food and Agricultural Organisation of the United Nations, Rome. PAGE 72 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Gray JM, Chapman GA, Murphy BW (2011) Land Management within Capability, a NSW Monitoring, Evaluation and Reporting Project, Technical Report, NSW Office of Environment and Heritage, Sydney. Gray JM, Chapman GA (in press) NSW Monitoring, Evaluation and Reporting Program. 2008-2010 results, Technical Report Series, Office of Environment and Heritage, Sydney. Keating BA, Carberry PS, Hammer GL, Probert ME, Robertson MJ, et al (2003) An overview of APSIM, a model designed for farming systems simulation, European Journal of Agronomy 18, 267-288. McKenzie NJ (2008) Monitoring of soil and land conditions. In Guidelines for Surveying Soil and Land Resources, 2nd ed. (Eds NJ McKenzie, MJ Grundy, R Webster, AJ Ringrose- Voase) Australian Soil and Land Survey Handbook Series, pp 491-514. CSIRO Publishing, Melbourne. OEH (2012) The Land and Soil Capability Scheme – a general rural land evaluation scheme for NSW. Technical Report, Office of Environment and Heritage NSW, Sydney McNeill & MacEwan 2007, Rowe, RK, Howe, DF and Alley, NF (1988) Manual of guidelines for land capability assessment in Victoria, Victorian Department of Conservation, Forests and Lands USDA (1983) National agricultural land evaluation and site assessment, Soil Conservation Service, US Department of Agriculture, Washington DC van Gool D (2008), Conventional land evaluation, in Guidelines for Surveying Soil and Land Resources, 2nd ed, McKenzie NJ, Grundy MJ, Webster R, Ringrose-Voase AJ (eds), Australian Soil and Land Survey Handbook Series, CSIRO Publishing. PAGE 73 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Mapping the immediacy of soil condition reaching irreversible tipping points to prioritise catchment actions Chapman GA1, Lobry de Bruyn LA2 ,Barrett T3 and Maher T4
1 Land and Soil Capability 39 Valley Road Springwood NSW 2777, [email protected] 2 School of Environmental and Rural Science, University of New England, Armidale, NSW,2351 [email protected] 3 Office of Environment and Heritage, C/- UNE Armidale NSW 2351 [email protected] 4 Natural Resources Commission GPO Box 4206 Sydney NSW 2001 [email protected]
Abstract NSW Catchment Management Authorities are encouraged to use resilience thinking and spatial analysis to target natural resource management investment and activities. Soil condition resilience reduces as soil condition approaches thresholds where soil ecosystem services become unviable. The threshold of very poor or degraded soil condition represents an irreversible tipping point from which recovery is problematic. We mapped soils for immediacy of reaching irreversible tipping point by dividing the proximity of reaching very poor condition by land management within capability trajectory. A rule and value based five class classification for soil condition (very good to very poor) in NSW has been allocated to 850 soil condition monitoring sites for various indicators eg, carbon, structure and pH. At most sites the degree to which land management practices are within land capability (LMwC) which has also been classified using a rules based, five class classification. Mean scores by land use and soil type combinations for both soil condition, LMwC and their reliability (using mean/SD and number of sites) were calculated. Maps for each soil condition indicator show the relative immediacies of reaching irreversible tipping points. When combined with maps of soil ecosystem service values such as provision of clean water, productivity and protection of soil biodiversity, priorities can be spatially allocated towards conserving the functions of highest value soils under greatest threat. Whilst the method has some promise, soil condition and land management practise data is sparse and often sporadically distributed. The method highlights soil type and land use combinations with insufficient confidence for determining priority actions, but useful for targeting data collection priorities. PAGE 74 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Some insights into the health of Dermosols around Tasmania Chris Grose1
1 Department of Primary Industries, Parks, Water and Environment, Prospect, Tasmania
Introduction The Department of Primary Industries, Parks Water and Environment (DPIPWE) is the primary agency in Tasmania responsible for the collection and interpretation of soil resource information across the state. In 2004 DPIPWE established a project to measure and monitor the health of a variety of soil types under a variety of land uses around the State. Implemented under the “Caring for our Country” banner and supported by the three Tasmanian Natural Resource Management (NRM) regions this project collected the baseline data to enable future quantifiable assessments of trend against a range of soil health indicators that could be used in NRM and State of the Environment reporting. Each site is to be revisited and resampled every five years to provide further data for comparison against the baseline dataset. This presentation summarises the results of two separate sampling rounds, five years apart, for one Tasmanian soil type. Soils provide a wide range of services to us and the planet, many of which are often forgotten in our busy daily routines. Soils provide the medium in which much of our food and other organic resources are grown; they support a range of environmental and biodiversity systems, filtering the water that flows into streams and rivers, providing homes, habitats and resources for a range of flora and fauna and contribute to the overall diversity of our planet. It should be reasonable, therefore, for us to understand the impacts that our use of this resource is having upon its health and this project was established to provide hard data on trends in the health of our soils over time. Dermosols, according to the Australian Soil Classification (Isbell 2002) represent those soils that have a moderate or better structure within the B2 horizon and are relatively uniform in texture. In Tasmania Dermosols typically comprise clay loam topsoils and increase in clay content with depth. Cotching et al (2009) report that, based on an interpretation of historical land systems information, Dermosols are probably the dominant soil order that occur in Tasmania (24% compared to the next most common, organosols, at almost 15%). Similar work by Cotching et al (2002) assessed 15 dermosols sites for changes associated with three forms of agricultural management, long term pasture, cropping with shallow tillage and cropping with more rigorous tillage. This work identified significant changes in a variety of soil parameters that could be attributed to land management and to duration of cropping history. While considered fairly robust Cotching et al consider that the Dermosols are generally less well drained and less robust than the ferrosols yet they are often cultivated as intensively. Chilvers (1996) has reported on management guidelines for both these soil types and impacts of intensive cropping on ferrosols have been reported in Sparrow (1999) which also identified correlations between organic carbon content and length of cropping history. The Soil Condition Evaluation and Monitoring (SCEAM) project was initiated to establish base line data on soil condition for a variety of soil-land use combinations around the state and to repeat sampling at each site every five years to develop a dataset that would allow identification of trends in soil condition over time under current management practices. Based on discussions at the national level at the time of inception, a number of PAGE 75 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
key soil properties were identified as being important indicators of soil health and these are the ones that are reported on in this paper. These soil properties are soil pH, organic carbon content, bulk density and aggregate stability (other indicators, such as EC and ESP were also measured but the data have yet to be analysed). This paper reports on changes that have occurred in the Dermosols between the baseline data collection and the repeat sampling five years later.
Method The SCEAM project sampled 285 sites around Tasmania between May 2004 and October 2008 encompassing a wide range of soil type and land use combinations. Of these sites 51 are recorded as being Dermosols under an agricultural land use while a further 24 are recorded as under plantation or native forest. Soil sites suitable for investigation under SCEAM were identified using available soil maps of Tasmania plus local knowledge and contacts in agricultural industries around the State. The relatively large number of sites on Dermosols recognises the contribution made by this soil type to agriculture around the State. Of the 51 agricultural sites sampled 12 are under long term pasture (six organic pasture, two dryland pasture and four on north facing slopes), 20 are under intensive cropping and 19 under perennial horticulture. The sites were sampled over a three year period and at different times of the year. Once a suitable paddock has been identified a point in the paddock is precisely located using differential GPS. A 50m transect is then run out across the slope and the bearing from the origin recorded. Two or three quick holes were dug close to the transect to confirm uniformity of soil type along the length of the transect. Three types of soil sample were collected from each selected site. Close to the start of the transect a soil pit was excavated to 100cm and described using standard terminology (McDonald et al 2009). Samples were collected by horizon unless the soil layer exceeded 30cm thickness in which case the layer was subdivided into layers of equal thickness of less than 30cm in thickness and sampled accordingly. The purpose of the soil pit was to confirm the soil classification and provide a full profile description and supporting analysis for the site. Standard suite of nutrient analysis was performed on each depth sample by a commercial laboratory in Western Australia, including the key indicators of soil health pH, organic carbon, electrical conductivity and ESP. A second set of samples was collected from two depths every two meters along the transect. Samples from each depth were bulked, air dried, sub-sampled using a splitter and sent for analysis. The depths of the two samples were 0-75mm and a subsoil depth 75mm thick at some depth from 100-300mm such that sampling across soil layers could be avoided. Subsoil sampling depth was consistent at each site but differed between sites depending on the thickness of topsoil layers. Finally, at three points along the transect, 12.5m, 25m and 37.5m, samples were taken from each of the same two depth layers for bulk density and aggregate stability assessment. At each point one sample for each of bulk density and aggregate stability was collected from the top soil and the subsoil layer. Bulk density samples were collected using a cylinder 75mm deep by approximately 75mm in diameter carefully knocked into the soil using a suitable driver. Aggregate stability samples were collected using a spade and carefully packed for return to the office. These samples were processed in house. PAGE 76 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Sampling was repeated at each site approximately five years after the initial sampling by returning to the same GPS coordinates and running the 50m transect out on the same bearing as for the original sampling.
Results Figure 1 presents the percentage change in organic carbon content in the topsoil at each cropping site against the percent change in bulk density for corresponding samples. Two observations are clearly evident. First, there appears to be no correlation between the change in soil carbon content and the change in bulk density. Secondly, the majority of samples (90%) show an increase in bulk density over the five year period while a smaller number (64%) show a decrease in organic carbon. Similar values were observed for the sub soil.
Figure 1. Change in organic carbon compared to change in bulk density between sampling rounds 1 and 2 for topsoil samples from Dermosols under cropping. Figures 2 and 3 show similar graphs for dermosols under pasture and perennial horticulture respectively. Even under pasture organic carbon content appears to have declined in 92% of sites while bulk density has increased in a similar proportion of sites. Under perennial horticulture, typically stone fruit, apples or vines, 74% showed an increase in bulk density and 63% showed a decrease in organic carbon.
Figure 2. Change in organic carbon compared to change in bulk density between sampling rounds 1 and 2 for topsoil samples from Dermosols under pasture. PAGE 77 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Dermosols under horticulture - topsoil 60 40 20 0 -80 -60 -40 -20 -20 0 20 40 -40
% change in bulk density bulk in change% % change in organic carbon
Figure 3. Change in organic carbon compared to change in bulk density between sampling rounds 1 and 2 for topsoil samples from Dermosols under perennial horticulture. A paired sample t-Test was used to determine if changes that occurred between the two sampling rounds were significant for each of the four indicators of soil health (pH organic carbon, bulk density and aggregate stability). For the Dermosols under cropping the t-Test indicated that the decrease in organic carbon, increase in bulk density, increase in pH and decrease in aggregate stability were all significant within the topsoil samples. Similar changes occurred within the subsoil except that the difference in aggregate stability was not considered significant. For Dermosols under pasture a decrease in organic carbon and an increase in bulk density were considered significant within the topsoil. There were no significant changes within the subsoil. For Dermosols under perennial horticulture a decrease in aggregate stability in both topsoil and subsoil samples was identified as significant, together with an increase in the bulk density of subsoil samples.
Discussion The results suggest that, for Dermsols sampled as part of the SCEAM project, soil health appears to be declining under the three land use types investigated, as indicated by significant adverse change in one or more of the soil health indicators of pH, organic carbon content, bulk density and aggregate stability. Increase in top soil and sub soil pH under cropping continues an anecdotal observation that many Tasmanian cropping soils are not suffering from pH decline. This can probably be attributed to the long history of liming that is undertaken by many farmers around the State. Organic carbon levels are recorded as declining significantly under both cropping and pasture management although only within the topsoil of pasture sites. The decline under pasture comes as something of a surprise and the data require further investigation and analysis to determine the reasons. Bulk density is also up in in both sub soil and top soil for pasture and cropping but only the subsoil under perennial horticulture. Increased bulk density is not necessarily an issue if previous bulk densities have been low. Mean topsoil bulk density under cropping varied from 0.99g/cm3 in the first round of sampling to 1.2g/cm3 in the second round, with PAGE 78 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
corresponding values of 0.87g/cm3 and 1.01g/cm3under pasture. While current topsoil bulk density averages under both cropping and pasture are not excessive the concern is that the general trend appears to be upwards. Declining aggregate stability under cropping suggests that continued use and cultivation of the land is weakening the bonds holding aggregates together. The reasons for change in the subsoil under perennial horticulture are less clear and need further investigation. It is acknowledged that the measurement of some of the soil properties described is frequently unreliable. Other sources of ‘error’ may relate to the resampling occurring at a different time of the year to baseline data collection (although we tried to minimise this as much as possible), or that the paddock was in a different phase of crop rotation or soil management or just that soil conditions were different due to natural seasonal variation in climate or soil moisture content. It is also unreasonable to make judgements on trends in soil condition based on two points on a graph and therefore interpretation of the above results needs to be handled with care. However, many of the initial results support findings of other writers (e.g. Cotching et al 2002, Sparrow et al 1999) that soil condition in Tasmania continues to decline and the results should be considered as an early warning to possible future trends.
References Chilvers WJ (1996) ‘Managing Tasmania’s Cropping Soil – a practical guide for farmers” Department Of Primary Industries and Fisheries, Tasmania. Cotching WE, Cooper J, Sparrow LA, McCorkell BE and Rowley W (2002) Effects of Agricultural Management on dermosols in northern Tasmania. Australian Journal of Soil Research 40, 65-79. Isbell RF (2002) ‘The Australian soil classification (revised edition).’ (CSIRO publishing). McDonald RC, Isbell RF, Speight JG, Walker J and Hopkins MS (2009) ‘Australian Soil and Land Survey Field Handbook, 3rd edition.’ (CSIRO publishing). Sparrow LA, Cotching WE, Cooper J, Rowley W (1999) Attributes of Tasmanian Ferrosols under different agricultural management. Australian Journal of Soil Research 37, 603- 622.
Soil monitoring – a basic tool for protection of soils and sustainable land use in Slovakia Jozef Kobza1
1 Soil Science and Conservation Research Institute Bratislava, Regional working place Banska Bystrica, Slovakia, e-mail: [email protected]
Abstract A soil monitoring system in Slovakia was implemented in 1993 and monitoring of 318 sites on agricultural land is carried out on five year cycles. Its importance consists of providing soil information on changing with space and time as well the development of soil quality in topsoil and subsoil. The soil monitoring network in Slovakia is constructed on ecological principles – all main soil types and subtypes, soil substrates, climatic PAGE 79 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
regions, emission regions, polluted and non-polluted regions as well as various land use are included. Soil properties are evaluated according to the main threats to soil relating to European Commission recommendation for European soil monitoring performance as follows: soil contamination, soil salinization and sodification, decline in soil organic matter, soil compaction and erosion. The most significant changes have been determined for physical degradation processes – soil compaction and erosion where about 50 % of agricultural land is potentially affected by soil erosion in Slovakia. In addition, decline in soil organic matter and available nutrients indicate the serious facts on evaluation and extension of soil degradation processes during last period in Slovakia. Obtained measured data and required outputs are reported to the Joint Research Centre (JRC ) in Ispra (Italy) and European Environmental Agency ( EEA) in Copenhagen (Denmark). Finally, soil monitoring system thus becomes a basic tool for protection of soils and sustainable land use as well as for creation of legislative controls not only in Slovakia, but in the EU too.
Introduction Soil monitoring system in Slovakia is a part of monitoring of environment which includes 10 partial monitoring systems: Soil, Water, Air, Meteorology and Climatology, Waste, Radioactivity, Forests, Foreign components in foods and fodders, Geological factors and Biota. Soil monitoring system in Slovakia has been running consistently since 1993. Its importance consists of providing actual and objective information on temporal trends in important soil properties according to threats to soil as well as soil erosion, soil compaction, decline in soil organic matter, soil salinization and sodification and soil contamination.
Material and Methods Soil monitoring network in Slovakia is constructed on ecological principles – all main soil types and subtypes, soil substrates, climatic regions, emission regions, polluted and non-polluted regions as well as various land use. There are 318 monitoring sites on agricultural and alpine land in Slovakia. All soil monitoring sites are located in WGS 84 coordinates. The monitoring site is of circular shape, with a radius of 10 m and an area of 314 m2. The standard depths of 0–0.10 m, 0.20–0.30 m and 0.35–0.45 m on soils under grassland and 0–0.10 m and 0.35–0.45 m on arable land are sampled, but the depth is adjusted to characterize the main soil horizons. The soil monitoring in Slovakia is carried out on 5 year cycles. The most important soil indicators concerning threats to soil are included in the soil monitoring system in Slovakia according to the recommendation of the European Commission (EC) for united soil monitoring system in Europe (Van – Camp et al. 2004). Methodical and analytical procedures are carried out according to publication of Kobza et al ( 2011).
Results and discussion Soil erosion Soil erosion belongs to the most environmental problems and the most extended degradation process in Slovakia. Erosion is measured on soil transects using 137CS profile distribution. In addition, the area of soil erosion distribution is determined by using a predictive model for erosion where the USLE equation is included (Wischmeier PAGE 80 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
and Smith, 1978). This interactive and predictive erosive model was created for farmers. They can find this model on www.podnemapy.sk. This application is very helpful for the information on soil erosion intensity and its area distribution. Results showed that there is actually about 50% of agricultural soils affected by soil erosion in Slovakia (including abandoned soils) (Table 1).
Table 1 Soil erosion distribution on agricultural land in Slovakia
Erosivity categories (soil loss) Area in ha % of farm land Non or slight (0–4t/ha/year) 1 225 238 50.92 Medium (4–10t/ha/year) 338 106 14.05 High (10–30t/ha/year) 428 379 17.80 Extremly high (>30t/ha/year) 414 248 17.22 Total 405 971 100.00
Soil compaction Soil compaction is monitored in the soil monitoring network, but only on arable land (topsoil and subsoil). The parameters monitored include bulk density, porosity, and texture. Trends in physical properties are different according to textural categories. Results showed that Luvisols and Chernozems (WRB, 2007) – intensive agriculturally exploited soils - have a negative trend in physical properties, especially on the bottom of the arable layer. Cambisols (mainly subtype Stagnic) and Rendzic Leptosols are sensitive to soil compaction. Contrastingly, Fluvisols and Mollic Fluvisols belong to the more resistant soils in relation to soil compaction. Concerning obtained results there are 200 kha of actually compacted soils (8.3% of agricultural land) and 500 kha of potentially compacted soils (20.7% of agricultural land) in Slovakia. Decline in soil organic matter Quantitative and qualitative indicators of soil organic matter (SOM) are permanently monitored in soil monitoring network in Slovakia. Originally, after slight decline in soil organic carbon (SOC) on the beginning of soil monitoring system in Slovakia (1990s), its increase has been indicated on all arable soils during last period. It could be probably caused by subsidies of Slovak Government for increasing of soil organic matter in soil. Qualitative indicators (CHA/CFA/Q46, fractional composition of HA) of soil humus are without significant trend. However, the measured values are running in the range which is characteristic for the concrete soil type as well as for the chemical structure of humic acids (HA). These indicators seem to be a result of soil genesis. Soil salinization and sodification On the basis of obtained results it was indicated that these processes are running more or less in paralell, but the sodification process seems to be dominant under soil-climatic PAGE 81 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
conditions of Slovakia (mostly in south-western and south-eastern part of Slovakia) – Kobza et al. (2009). There are an estimated 5000 ha of salty soils in Slovakia (0.2% of agricultural land of Slovakia). Soil contamination The significant change in concentration of inorganic (Cd, Pb, Cu, Zn, Cr, Ni, Co, Se, As, Hg, F) and organic (PAHs, PCB) contaminants was not indicated during monitored period of 20 years. It means that the soils which were contaminated at the beginning of soil monitoring, are still contaminated at present. There are about 25 kha contamined soils in Slovakia (1 % of agricultural soils of Slovakia), but predominant part of them is contaminated by geogenic influence. Anthropogenic influence of soil contamination has only very slight decreasing trend (see an example of fluorine development near aluminium smelter) during last period.
Figure 1 Development of Fluorine near Aluminium smelter The soil has a specific role in environment. Namely, when the concentration of fluorine in air emissions has been rapidly decreased (improvement of aluminium processing technology during monitored period), the concentration of fluorine in soil has only slightly decreased. Finally, the soils can be easily polluted, but natural remediation is a long to very long-term process.
Soil monitoring database of Slovakia and data reporting The user interface of the database also allows entry and viewing of data, their selection and printing a table of data archived for soil monitoring sites (profiles). Structural harmonization of the database and migration to an ORACLE platform are proceeding at present. In addition, a network interface database for accommodating the new data has been created. The main functional services for publishing spatial information of soil monitoring are online, with OGC – WMS & WFS specification to deliver the standard reporting data formats (Kobza et al., 2013). The reporting context for soil monitoring data is illustrated in Figure 2. PAGE 82 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 2 Implementation of soil monitoring system with monitoring of environment in Slovakia and EU
Conclusion The soil monitoring system supports the comparable and objective data on actual state and development of soils in Slovakia. The obtained results are useful above all in decision-making sphere (e.g. for soil protection) and in various branches of national economy as well as in research institutes and universities with environmental education. In addition, reporting of obtained data to EEA in Copenhagen (Denmark) and to JRC in Ispra (Italy) contributes to creation of important outputs concerning actual state and development of soil cover in EU.
References Kobza, J., Barančíková, G., Čumová, L., Dodok, R., Hrivňáková, K., Makovníková, J., Náčiniaková–Bezáková, Z., Pálka, B., Pavlenda, P., Schlosserová, J., Styk, J., Širáň, M., Tóthová, G. (2009) Monitoring pôd Slovenskej republiky (Soil monitoring of Slovak Republic) (SSCRI Publishing: Bratislava), 200 pp. ISBN 978–80–89128–54–9. Kobza, J., Barančíková, G., Bezák, P., Dodok, R., Grečo, V., Hrivňáková, K., Chlpík, J., Lištjak, M., Makovníková, J., Mališ, J., Píš, V., Schlosserová, J., Slávik, O., Styk, J., Širáň, M. (2011) Jednotné pracovné postupy rozborov pôd (Uniform analytical procedures for soil) (SSCRI Publishing: Bratislava), 136 pp. ISBN 978–80–89128–89–1. Kobza, J., Barančíková, G., Dodok, R., Hrivňáková, K., Makovníková, J., Pálka, B., Styk, J., Širáň, M. (2013) Soil monitoring of Slovakia. SSCRI Bratislava, 26 pp. Van–Camp, L., Bujarrabal, B., Gentile, A–R., Jones, R.J.A., Montanarella, L., Olazabal, C. And Selvaradjou, S–K. (2004) Reports of the Technical Working Groups Established under the Thematic Strategy for Soil Protection. EUR 21319 EN/5, 872 pp, Office for Official Publications of the European Communities, Luxembourg. Wischmeier, W.H., Smith, D.D. (1978) Predicting rainfall erosion losses – Guide to conservation planning, Agricultural Handbook 537, (USDA Publishing: Washington), 58 pp. PAGE 83 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The Conservation Effects Assessment Project – operation of a national scheme for soil and environmental monitoring Mari-Vaughn V. Johnson1, M. Lee Norfleet1, Jay D. Atwood1, James R. Kiniry2, Jeff G. Arnold2, Mike J. White2, Jimmy Williams3
1 USDA-Natural Resources Conservation Service; Grassland, Soil, and Water Research Laboratory; 808 East Blackland Road; Temple, Texas 76502 USA. Presenting Author email: mjohnson@brc. tamus.edu 2 USDA-Agricultural Research Service; Grassland, Soil, and Water Research Laboratory; 808 East Blackland Road; Temple, Texas 76502 USA 3 Texas A&M Agri-Life; Blackland Research and Extension Center; 720 East Blackland Road; Temple, Texas 76502 USA
Abstract The Conservation Effects Assessment Project (CEAP) is a complex project involving numerous entities and activities ranging from scientific research, monitoring programs, and data analysis, to model development and simulation of management optimization, all with the ultimate intent of quantifying environmental benefits of conservation dollars spent. CEAP has five major Components. The Cropland Component and Grazingland Components provide national and large regional assessments of conservation practice impacts on U.S. cropland and grazing land through 1. extensive monitoring and analysis of land management and conservation practice adoption and 2. interpreting these data on large watershed scales. The Wetlands and Wildlife Components focus on conservation practice impacts on wetlands and wildlife, with Component regions delineated along geo-ecological boundaries rather than watersheds. The Watershed Component provides highly detailed, site-specific information on selected watersheds to provide a finer level of analysis about lands of particular interest. Although all of the CEAP Components are interrelated, the focus of this discussion will be on the CEAP Cropland Component.
CEAP: History In the 2002 Farm Bill the United States Congress increased the funding levels for conservation programs and initiatives to over $38 billion ($42 billion AUD), an 80 percent increase compared to funding levels in the previous Farm Bill, which was enacted 6 years earlier, in 1996. This single most significant investment of Federal resources to conservation efforts on private lands in U.S. history was accompanied by a request for greater accountability of per dollar benefits. The windfall budget and logical Congressional request for accountability exposed a significant gap in scientific knowledge related to conservation practices. While field-scale experiments generally supported the widespread assumption that conservation practices provide positive benefits, there was no extant quantification of conservation benefits on a large regional or national scale. In response to the identified need for greater scientific inquiry into and analysis of conservation practice effects, the U.S. Department of Agriculture’s Natural Resources Conservation Service (USDA-NRCS) initiated the Conservation Effects Assessment Project (CEAP) in conjunction with the USDA Agricultural Research Service (ARS), a number of other government organizations, and university scientists. CEAP’s goal was to quantify the conservation benefits of Federal dollars at the national and large watershed scales. PAGE 84 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
CEAP reports were designed to inform Congress and other governmental entities of conservation successes and opportunities to improve development and implementation of future conservation programs and initiatives.
Cropland CEAP: Data Collection The Cropland National Assessment Component of CEAP has produced large watershed- scale reports of conservation impacts and needs on cultivated cropland in seven large watersheds in the United States, with reports for seven additional watersheds planned for future publication (http://www.nrcs.usda.gov/wps/portal/nrcs/detail/national/technical/ nra/ceap/?cid=nrcs143_014144). Cropland CEAP relies on a statistical approach to data collection and a modeling approach to estimate the impacts of various reported conservation practices and approaches to agricultural land management on natural resources and some ecosystem services. Independent of the CEAP effort, the NRCS assesses the Nation’s natural resources via the National Resource Inventory (NRI), a statistically sound sampling protocol that represents land use and associated trends in natural resources across the United States and its territories (Figure 1). Between 1977 and 2000, statistically representative points on a permanent, scientifically based sampling frame were assayed on a five-year cycle. The first NRI report was published in 1982 and each iteration of the sampling and analysis has led to improvements in data collection and interpretation. In 2000, an annual sampling protocol was adopted to better respond to trends in natural resource dynamics.
Figure 1. Trends in sheet and rill erosion rates, as captured by NRI data (USDA, 2013). PAGE 85 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The CEAP effort relies heavily on the NRI statistical design, which allows CEAP to select a subsample of representative points in any given watershed. The NRI was designed to complement the NRCS Soil Survey Program. Each NRI sampling point is linked to NRCS soil survey data. The soil survey databases include both soil characteristics and climate databases associated with each point. CEAP collects more information about each NRI point included in the CEAP survey through a “NRI-CEAP Cropland Survey”, which is accomplished by training non-Federal survey enumerators to collect land management data directly from the farmers. Participation in the survey is not mandated by any regulations or rewarded by any federal incentives. Rather, the data is voluntarily provided by land owners, with no penalties imposed on landowners who do not wish to participate. The survey collects information on land management (crops grown, planting and harvesting dates, nutrient and pesticide applications, tillage applications, and conservation practice use, etc.) over the previous three years, which allows the survey to capture information on crop rotation management.
Cropland CEAP: Modeling Approach Four data sources support the CEAP modeling effort (Figure 2). The statistically valid NRI-based sampling approach described above captures the diversity of land uses, soils, climate, and topography across the area of interest. The NRI-CEAP farmer surveys also provide valuable information on site-specific farming and conservation practices over a period of three years. Data on land management is also collected from NRCS field offices, where specialists work directly with land-owners and land-managers to develop conservation plans that are soil, climate, and land-use appropriate. The final source of model input data is provided by the USDA – Farm Service Agency (USDA-FSA), which provides NRCS with information on lands enrolled in various USDA supported conservation programs. The USDA-NRCS leads the CEAP modeling effort, with technical support provided by USDA-ARS and Texas A&M AgriLife. The CEAP modeling team applies the daily time-step, process based Agricultural Policy Environmental Extender (APEX) model to estimate the impacts of reported land management and conservation practice adoption on soil quality on farmed fields. The majority of scientific research on conservation practice effects has been conducted at the field scale upon which APEX operates, enabling modelers to develop well calibrated and validated estimates. APEX simulates the interactions between plant growth; tillage; grazing; and water, nutrient, and pesticide dynamics to estimate yield stability and nitrogen, phosphorus, sediment, and pesticide losses at the edge-of- field, including below the root-zone. APEX also simulates carbon dynamics associated with conservation practice adoption. APEX is useful for understanding field-scale impacts of changing conservation and land management practices. CEAP is also able to answer larger scale policy relevant questions, including conservation practice impacts within watersheds. The rigorous NRI statistical framework enables modelers to aggregate the edge-of-field modeled results and apply the Soil and Water Assessment Tool (SWAT) to provide estimates for regional and national conservation practice impacts. SWAT uses APEX output as input representing farmland contributions to instream loads of nutrients, pesticides, and sediments. SWAT also uses measured data to simulate non-point source loads from land uses other than cropland (urban, forested, etc.); the model routes instream loads from one watershed to the next, enabling modelers to estimate the impacts of conservation practice adoption may across regional scales. PAGE 86 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 2.The CEAP modeling framework
Cropland CEAP and Soil Change CEAP is a complex project, which relies on the honest and willing participation of landowners as well as good science and evolving modeling systems. The capacity of models to represent reality is largely dependent on data availability and scientific understanding of complex ecosystem processes. In the agro-ecosystem, soils and soil change impact yield potentials, as well as the efficacy of conservation practices, which has environmental impacts. More monitoring work done in collaboration with modelers is needed to better characterize soil change dynamics. When research scientists do not work closely with modelers, they may collect data that cannot be used to inform the model due to the way that agro-ecological processes are coded into the model and the way the model simulates interactions between components. Through research scientists- modeler collaboration, models can be better built, better calibrated and better validated to simulate soil dynamics realistically. Soil health, including biotic, chemical, and physical interactions, has been suggested as a good indicator for predicting soil resistance and soil resilience. However, there is currently insufficient data for soil health. Few quantitative metrics developed to elucidate soil health. The metrics upon which the scientific community eventually settles for characterizing soil health need to be not only quantitative, but also readily transferrable across all potential agricultural soils. Finally, the Conservation Effects Assessment Project was designed to determine the impacts of conservation practices adoption across large watersheds in the United States. Since its inception, people have attempted to use CEAP for purposes other than those for which it was designed. Namely, CEAP was not designed to predict impacts on biodiversity or other ecological services, other than nitrogen, phosphorus, carbon, and sediment dynamics at the field scale (APEX) or the watershed scale (SWAT). As mentioned previously, CEAP relies on the NRI statistical framework, which is applicable at various scales, but adjustments to data interpretation must be made when scales are changed in order to remain statistically valid. Therefore, many of the scalar assumptions extrapolated from CEAP should not be considered valid without further analysis.
References U.S. Department of Agriculture (2013) Summary Report: 2010 National Resources Inventory, Natural Resources Conservation Service, Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa. http://www.nrcs. usda.gov/Internet/FSE_DOCUMENTS/stelprdb1167354.pdf PAGE 87 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The influence of land cover history on productivity Kathryn Sheffield1, Elizabeth Morse-McNabb2
1 Agriculture Research Division, Department of Environment and Primary Industries, Parkville, Victoria, Australia. [email protected] 2 Agriculture Research Division, Department of Environment and Primary Industries, Bendigo, Victoria, Australia. [email protected]
Abstract Remote sensing of land cover history can provide broad indications of productivity over time, that can potentially be used as an indication of soil condition change. The aim of this research is to understand the influence of land cover history and environmental factors on current soil condition using a study area located in the Wimmera region of north-west Victoria. This investigation undertook a very long time series analysis of Landsat imagery (approximately 40 years) and used novel datasets and approaches to assess areas based on history and cover condition. Normalised Difference Vegetation Index (NDVI) and land cover maps based on NDVI thresholds were used to identify trends in vegetation cover change at a landscape scale, and their relationship with factors such as land use intensification history, and land use. This work has improved the broad, baseline understanding of production variation across the landscape, while also providing a practical demonstration of the integration of a range of disparate data sources.
Introduction Farming systems and land use history, including production intensity and land management, can result in soil changes that compromise current and future capacity for primary production and provision of ecosystem services. Differences in vegetation cover linked to agricultural production in farming landscapes, can be attributed to a number of factors, including climatic conditions, soil properties, geomorphological factors, that occur as gradations across the landscape, and also land cover and management, occurring as sharp transitions aligning with land parcels (Hill and Donald 2003; Sumfleth and Duttmann 2008; Wen, Yang et al. 2012). An improved understanding of land cover history will increase our understanding of soil condition, and linkages between soil health and the productive capacity of Victorian agricultural landscapes. There are many different spatial data sets available that can contribute to an understanding of land use and production history in Victoria, including soil and landform information, climatic data, such as rainfall, long-term production history (Sinclair, White et al. 2012) and remotely sensed data. Information derived from remotely sensed data, such as the Normalised Difference Vegetation Index (NDVI), can be used as an indicator of green vegetation cover. Investigations into the relationships between these types of data have improved understanding of the linkages between remotely sensed data and landscape and agricultural information such as crop rotation, productivity, soil properties and land management practices (Dang, Pringle et al. 2011; Hill and Donald 2003; Sumfleth and Duttmann 2008). This study aims to improve our understanding of the influence of land cover history and various environmental factors on soil condition in the Wimmera region of north-west Victoria, by examining linkages between vegetation cover over time and factors such as rainfall, landform and production history. This paper presents a preliminary analysis of production history and vegetation cover at the landscape scale. Rainfall and landform factors will be integrated and analysed at the paddock level in future work. PAGE 88 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Methods Data This study used 97 Landsat images acquired between 1973 and 2012, sourced from the USGS Landsat archive (http://earthexplorer.usgs.gov.au/). Landsat images have a swath (approximately 170 km x 180 km) large enough to deliver landscape scale information, and a spatial resolution (approximately 30 m – 60 m) which is able to deliver information at a paddock and sub-paddock scale. The duration of the image series (i.e. over 40 years) enables longer-term trends of vegetation growth and land use change to be detected. Current advancements in processing of the Landsat image archive have led to a reliably calibrated data set, with a consistent absolute radiometric scale between sensors and total uncertainties of under 10% for most sensors and bands (Hansen and Loveland 2012; Markham and Helder 2012). Imagery derived from Landsat sensors 1-7 were used in this study, with each image calibrated to top of atmosphere reflectance using coefficients published in Chander, Markham et al. (2009). Further geo-rectification was undertaken as required and all images resampled to a 30 m spatial resolution to provide a consistent spatial unit across the time series. Clouds and shadow were masked from each image and NDVI derived using red and near-infrared (NIR) reflectance:
Where: NIR = near-infrared reflectance Red = red reflectance
Each year was considered in terms of seasons: Summer (December to February), Autumn (March to May), Winter (June to August) and Spring (September to November) and, where available, one cloud-free image per season was acquired. This process generated a series of 97 NDVI images spanning 1973 to 2012. Over the 40 year time period, 29 images were acquired in Summer, 20 images in Autumn, 18 images during Winter and 30 images in Spring. This study examined spring vegetation cover. The time series of 30 Spring images was grouped into five-year time intervals, and the percentage of time a pixel was covered by green vegetation (NDVI >0.65) was calculated, producing a measure of high NDVI frequency. Other spatial data used included land privatisation (i.e. when land tenure changed from Crown Land to private), and land use in 2005, estimated from Landsat images acquired between 1989 and 2005 (Sinclair, White et al. 2012). Land privatisation was considered in three time periods: before 1866, (early pastoralism), between 1866 and 1888, and after 1888 (Victorian gold rush and agricultural expansion). The development of this data set, and sources of information used to compile this map, are detailed in Sinclair, White et al. (2012).
Study area The study area, located in the Wimmera region of Victoria (Figure 1), is dominated by rain-fed cropping production systems, pasture and grasslands, with remnant woody native vegetation found in small blocks, Crown Land reserves and along watercourses. PAGE 89 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 1: Study area in the Wimmera region of Victoria, Australia
Results and Discussion Private land in the study area is dominated by dryland agriculture (90% of study area) and grazing (7.5% of study area), with smaller areas of irrigated agriculture and timber production. Approximately 1.8% of the study area remains as Crown Land. Figure 2 illustrates fluctuations in vegetation cover over time, and differences between land uses. Dryland agriculture and grazing show similar patterns with small fluctuations over time, while irrigated agriculture shows more variability in high NDVI frequency, most likely related to water availability. Timber production shows less frequent fluctuations, related to harvesting and planting cycles within the plantations.
Figure 2: Average percentage of time pixels had an NDVI value of greater than 0.65 during Spring (based on a five-year average), grouped by land use, in 2005 (from Sinclair, White et al. (2012)) PAGE 90 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 3 shows changes in higher vegetation cover frequency based on the period of land privatisation. Vegetation located on Crown Land, which is predominantly native woody scrub vegetation, had a low average percentage of time with a Spring NDVI value greater than 0.65 (Figures 2 and 3), suggesting that vegetation cover on Crown Land, while typically lower than agricultural land, has remained relatively steady over the past four decades. This also illustrates how agriculture has changed the vegetation cover trends within the landscape. Generally, land privatised after 1888 experienced high Spring NDVI values, less frequently, compared with land privatised either before 1866 or between 1866 and 1888. There was also less variation in the frequency of high NDVI values on land privatised after 1888, suggesting these sections of the landscape may have lower, but more consistent, vegetation cover compared with areas privatised before 1866. This may be related to the quality of agriculture land, in terms of soil quality and fertility, which was privatised at during the different time periods. Land management, including land use, such as inter-annual crop rotations compared with grazing pastures, and practices such as stubble retention may also be a factor.
Figure 3: Average percentage of time pixels had an NDVI value of greater than 0.65 during Spring (based on a five-year average) grouped by privatisation date. Changes in denser vegetation cover, based on period of land privatisation, were investigated for specific land uses, such as dryland agriculture (Figure 4). In some five year periods (such as 1978-1982), all dryland agriculture areas showed similar frequencies of higher NDVI, while in other periods (such as 1983-1987 and 2003-2007) there are large differences, dependent on when the land was privatised. This may be due to factors such as the type of agriculture undertaken on these paddocks, rainfall and soil properties. Further analysis, including additional factors such as soil properties, will determine the significance of these observations in relation to soil condition and management. PAGE 91 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 4: Average percentage of time dryland agriculture pixels (based on estimated land use in 2005 from Sinclair, White et al (2012)) had an NDVI value of greater than 0.65 during Spring (based on a five-year average) grouped by privatisation date. The work presented in this paper has shown relationships between production and selected landscape scale factors. These relationships begin to show the influence of land management on production and vegetation cover. Further work will develop relationships at a paddock scale, involving additional factors such as soil properties and land use intensification, that provide indicators of the length of time land has been used for production purposes( see Sinclair, White et al (2012) for dataset details); geomorphological units, that provide information on the underlying land systems (Department of Environment and Primary Industries 2013), and rainfall. Spatial relationships between factors across the landscape will also be explored to improve our broad, baseline understanding of variability in production across landscapes.
References Chander G, Markham BL, Helder DL (2009) Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and E0-1 ALI sensors. Remote Sensing of Environment 113, 893-903. Dang YP, Pringle MJ, Schmidt M, Dalal RC, Apan A (2011) Identifying the spatial variability of soil constraints using multi-year remote sensing. Field Crops Research 123, 248-258. Department of Environment and Primary Industries (2013) Victorian Resources Online: Geomorphology. http://vro.depi.vic.gov.au/dpi/vro/vrosite.nsf/pages/landform_ geomorphology, Date accessed: 24/1/2014 Hansen M, Loveland TR (2012) A review of large area monitoring of land cover change using Landsat data. Remote Sensing of Environment 122, 66-74. Hill MJ, Donald GE (2003) Estimating spatio-temporal patterns of agricultural productivity in fragmented landscapes using AVHRR NDVI time series. Remote Sensing of Environment 84, 367-384. PAGE 92 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Markham BL, Helder DL (2012) Forty-year calibrated record of earth-reflected radiance from Landsat: A review. Remote Sensing of Environment 122, 30-40. Sinclair SJ, White MD, Medley J, Smith E, Newell GR (2012) Mapping the past: Constructing a digital land-use history map for Victoria, Australia. Proceedings of the Royal Society of Victoria 124(3), 193-206. Sumfleth K, Duttmann R (2008) Prediction of soil property distribution in paddy soil landscapes using terrain data and satellite information as indicators. Ecological Indicators 8, 485-501. Wen L, Yang X, Saintilan N (2012) Local climate determines the NDVI-based primary productivity and flooding creates heterogeneity in semi-arid floodplain ecosystem. Ecological Modelling (242), 116-126.
Monitoring ground cover for Australian agriculture using MODIS and Landsat imagery Jane Stewart1, Juan Guerschman2, Peter Scarth3, Lucy Randall1, Jasmine Rickards1
1 Australian Bureau of Agricultural and Resource Economics and Sciences, Department of Agriculture, GPO Box 1563, Canberra, ACT 2601, [email protected] 2 CSIRO Land and Water, GPO Box 1666, Canberra, ACT 2601, [email protected] 3 Department of Science, Information Technology, Innovation and the Arts, 41 Boggo Road, Dutton Park, QLD 4102, [email protected]
Abstract Through collaboration a national approach to monitor ground cover has been implemented for Australia. This involved the Department of Agriculture, state and territory agencies, CSIRO and the Terrestrial Ecosystem Research Network. Standardised methods are used to measure ground cover and site data is compiled into a national database for calibration and validation of satellite derived ground cover estimates. The outcome is an operational, combined Landsat/MODIS ground cover product for Australia. This fractional cover product separates ground cover into the components of photosynthetic and non-photosynthetic vegetation as well as the bare soil component. This combined Landsat/MODIS product enables ground cover to be monitored through time (back to 1985 for Landsat and 2000 for MODIS) and at scales (30 metre and 500 metre pixel sizes) appropriate for national, state, regional and property decision making. The product is being used to assess Australia’s soil resources and agricultural productivity, particularly grazing in the rangelands, and as an input to erosion models. The current accuracy and future developments of the combined Landsat/MODIS fractional cover product are given along with measures for reporting ground cover trend. PAGE 93 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
WEDNESDAY 26 MARCH 2014 - SOIL CHANGE MATTERS WORKSHOP
Understanding Soil Processes in natural Soils – a Base for assessing Soil Change Daniela Sauer1
1 Dresden University of Technology, Helmholtzstr. 10, D-01069 Dresden, Germany, E-mail: daniela. [email protected]
Abstract Which soil-forming processes are taking place in different environments? At which rates are they operating? In which way do they change soil fertility? These questions need to be addressed for estimating the soils’ responses to climatic changes. Three examples of soil development will be introduced: 1) from a gravelly semi-desert (Patagonia); 2) from the Mediterranean region (S Italy), and from a humid-temperate region (S Norway). A major process in the semi-desert is dust entrapment between gravel; this dust carries carbonates which are dissolved in moist winters and re-precipitated 30-60 cm below the surface; carbonates lead to strong cementation on the long term. Dust accumulation has important implications for water storage, and it provides rooting space and nutrients. Main processes in the Mediterranean region include 1) carbonate leaching and accumulation at some depth, 2) weathering; 3) clay formation and translocation; 4) iron oxide formation, leading first to brown, later to progressively reddish soils; 5) increase of soil thickness (T) with terrace age, resulting from the interplay of soil deepening (D) through weathering; soil upbuilding (U) due to accretion of material; soil removal (R): T = f (D + U – R) (Johnson, 1985). Two directions of soil development in humid-temperate climate will be introduced: 1) formation of Podzols on sandy beach deposits, involving leaching of nutrients and acidification; 2) formation of Albeluvisols on loamy marine sediments, leading to more fertile soils possibly affected by perched water. Possible responses of the soils to climatic changes will be addressed for each of the examples.
Global soil change, anthro-pedostratigraphy, and the Anthropocene Daniel deB. Richter1, Allan R. Bacon 1, Zachary Brecheisen1
1 Duke University, Box 90328, LSRC, NSOE, Durham, NC 27708 USA, [email protected]
Abstract The Anthropocene is a useful concept that describes the transformation of the planet from a natural to a human-natural system. Geologically, i.e., stratigraphically, the Anthropocene remains a hypothetical name for contemporary geologic time, with the issues involved in renaming the Holocene are being actively considered. For about 50 years, pedo-stratigraphy, i.e., the science of interpreting soils information in stratigraphy, has been critical to understanding Quaternary and older environments. With more PAGE 94 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
than half of Earth’s 13 billion hectares of soil today actively under human management, human-altered soils are developing characteristic spatial structures, processes, and temporal dynamics. Such global soil change is recorded in Earth’s litho-, chemo-, and bio-stratigraphies, and is significant to Earth-system evolution and human well-being. Litho-stratigraphic signals include human-altered native soil profiles mixed with, sealed, or buried by new materials and deposits. Chemo-stratigraphic signals are being imprinted on soils worldwide, via land applications of fertilizers and wastes, irrigation waters, and long-distance atmospheric transport of pollutants. Bio-stratigraphic signals accumulate in soils as native vegetation is converted to that domesticated species and due to human- associated global mixing and extinctions of plants and animals. Fossilized products of biota including bones, pollen, and other biotic parts are being transformed on a global scale. All considered, pedological perspectives suggest it is entirely reasonable to consider that the Earth’s pedostratigraphy is being transformed into an anthro- pedostratigraphy and that this transition supports the validity of the transition from the Holocene to the Anthropocene Epoch.
Modelling soil change over millennial time scales Uta Stockmann1, Tom Vanwalleghem2, Budiman Minasny3, Alex. B. McBratney4
1 Department of Environmental Sciences, Faculty of Agriculture and Environment, The University of Sydney, Biomedical Building C81, 1 Central Avenue, ATP, Eveleigh, NSW, 2015, Australia, uta. [email protected] 2 Department of Agronomy, Campus de Rabanales, University of Cordoba, 14010 Cordoba, Spain, [email protected] 3 Department of Environmental Sciences, Faculty of Agriculture and Environment, The University of Sydney, Biomedical Building C81, 1 Central Avenue, ATP, Eveleigh, NSW, 2015, Australia, [email protected] 4 Department of Environmental Sciences, Faculty of Agriculture and Environment, The University of Sydney, Biomedical Building C81, 1 Central Avenue, ATP, Eveleigh, NSW, 2015, Australia, [email protected]
Soil plays a fundamental role in natural ecosystems. The soil body provides many ecosystem services that are beneficial for humankind. Ecosystem changes such as change in land use and land management practices can affect the functioning of soil systems, and in turn can have a major impact on the global water and energy balance. Therefore it is important to understand how fast a soil can form, especially in response to the consideration of soil as a renewable resource. Work on quantifying the processes that form a soil profile is still minimal, but the availability of new sophisticated laboratory techniques has opened up the possibility of addressing the demand of quantifying processes of soil landscape evolution in the critical zone. Therefore, a field study was conducted along three principal toposequences in the Werrikimbe National Park situated in northeastern NSW, Australia, to actually measure and quantify the rate of soil formation. To investigate soil formation processes terrestrial cosmogenic nuclides were used to derive production rates of soil in mm kyr-1 and optically stimulated luminescence was applied to examine vertical mixing rates of soil. The outcome of this work has advanced the fundamental knowledge in soil science and provides basic data which can be used to model the stability of the soil-landscape. We therefore implemented the derived rates of pedogenic processes in a quantitative-mechanistic soil formation and landscape evolution model, MILESD, and tested its ability to predict the current soil landscape of the subcatchment analyzed at Werrikimbe National Park. PAGE 95 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil change an important aspect of soil security Damien Field1, Alex McBratney2, Budiman Minasny3
1 Dept. Environmental Sciences, Fac. of Agriculture and Environment, The University of Sydney, [email protected] 2 Dept. Environmental Sciences, Fac. of Agriculture and Environment, The University of Sydney, alex. [email protected] 3 Dept. Environmental Sciences, Fac. of Agriculture and Environment, The University of Sydney, [email protected]
A multi-dimensional concept of soil security has been developed to address the global challenges that affect the long term sustainable development, including; food, water and energy security, adapting to a variable climate, and maintaining biodiversity and ecosystem services. There are five dimensions of soil security addressing the biophysical aspects assessed as the soil’s capability and condition, and also considers the socio- economic concerns through the dimensions capital, codification and connectivity. This overarching concept of soil security is broader than previous ideas such as soil quality, soil health and the more recent concept of soil change. The dimensions of capability and condition recognize that soil change occurs over geological as well as the human timescales and is affected by the soil’s use and management. Like soil change soil security recognizes that the use of soil results in disturbances and the change in the soil’s condition is determined by its resistance and resilience, but soil security also requires the establishment of local reference state, i.e. its capability. This reference state needs to be recognized in a soil classification framework and addresses the question, ‘what functions can a soil perform’. As called for by soil change the opportunity to effectively share knowledge and information beyond the soil science community needs to be embraced. The soil security concept explores how a ‘value’ needs to be placed on the soil, contributing to the assessment of the soil’s capital, and the need to support the development of soil policy based on relevant soil knowledge and understand how people are connected to the soil.
Soil knowledge matters Neil McKenzie1
1 CSIRO, Australia. Abstract There is an increasing awareness of the finite nature of soil resources and the significant threats to soil function in most countries regardless of their wealth. The national and global issues are reasonably clear. Unprecedented demands are being placed on the world’s soil resources and by 2050 they need to support increased food production of >70%. However, arable land is finite and soil degradation widespread. Major crops are reaching yield plateaux and better soil management is needed to conserve nutrients, improve water-use and reduce emissions. Climate change compounds the situation. Some of Australia’s soil management challenges are immediate, obvious and serious – they arise partly because of the nature of our soils and the history of land management. Other problems are more subtle but equally important in the long term – they require vigilance and a sustained response over decades. At present, the country does not have PAGE 96 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
the necessary information and response systems in place to deal with these problems. This state of affairs has arisen because of the long-term nature of the problem, a range of institutional factors, and the reality that the costs and benefits associated with improved soil management have both a private and public component. It is argued that ten key actions are necessary. 1. Build an enduring capability for mapping, monitoring and forecasting the functional properties of Australian soils. 2. Produce regular assessments showing where land management systems can operate sustainably within the constraints set by changing climate, weather and soils. 3. Provide reliable and locally appropriate advice on sustainable soil management throughout Australia and beyond. 4. Deal with the key threats to soil function in Australia, especially acidification. 5. Identify the potential for sequestration of carbon in soils at scales that usefully inform land managers, markets and governments. 6. Develop land management systems that are sustainable and improve the resilience of landscapes. 7. Rethink the institutions, especially those responsible for soil information. 8. Provide stable and attractive career paths for soil and land resource specialists. 9. Train a new generation of soil scientists and ensure ecologists, foresters, agronomists and related professionals have a good understanding of soil function. 10. Engage globally and understand the scale of the soil security challenge. These actions should ensure that Australia’s soil resources provide the ecosystem benefits and wealth that are prerequisites for a secure and prosperous future.
Introduction to Policy Perspectives Andrea Koch1
1 Program Leader, Soil Carbon Initiative, United States Studies Centre at the University of Sydney, Institute Building (H03), City Road, UNIVERSITY OF SYDNEY NSW 2006, [email protected]. au
Abstract The international community is responding to global soil degradation through an array of initiatives including the UN FAO Global Soil Partnership, the Global Soil Forum in Germany, the UNCCD’s push for a land degradation neutral target, and UNEP recognition of the importance of increasing soil carbon. Following Rio+20, the UN aims to establish new sustainable development goals to be established in 2015, intensifying the focus on land and soil management at the international level. At the national level, many countries are developing new policy approaches and strategies to address soil, including the USA and European countries. The EU is still negotiating for the establishment of a European Soil Protection Framework, after eight years of discussion and delay. PAGE 97 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
In Australia, the development of the National Soil Research Development and Extension Strategy brings a new and welcome focus to soil policy, through a coordinated national approach. At the same time, the Coalition Government’s Direct Action Plan for Climate Change will see a much stronger emphasis on soil carbon sequestration for climate change abatement. How do these Australian soil policy initiatives compare to other initiatives around the world? What could and should Australia offer in the international arena, and what can we learn from other nations? What does this mean for Australia’s soil science effort and how does Australian soil science contribute to international policy initiatives? This introduction will lay out the international landscape for soil policy development and provide context for Australia’s soil policy agenda.
Policy and effective action for soil security. Johan Bouma1
1 em. Prof. Wageningen University, the Netherlands. [email protected]
Abstract Soil science is a vital scientific discipline producing cutting-edge research in its various sub-disciplines. Its role in studying major environmental issues, often defined as food security, water and energy availability, climate change and biodiversity loss, is, however, less obvious to stakeholders, policy makers and the public at large. Taking a pro-active interdisciplinary approach in demonstrating the role of soils when studying these issues is advocated in the context of Sustainable Development Goals. Soil change matters not only in a negative way when referring to erosion and degradation but also, and particularly, to soil improvement. But only successfully completed programs in practice will be convincing and that’s why an active role of soil scientists, acting as knowledge brokers, is advocated in transdisciplinary programs including more emphasis on preparation and implementation than is allowed in current programs. A case study is presented and a narrative is used to link the five major environmental issues in a logical sequence, showing their interdependance. Current research and education programs should reflect demands made by inter- and transdisciplinary approaches.
Introduction The title of this conference is well chosen. Mankind is facing a number of major environmental challenges in the coming decades as world population is likely to exceed 9 billion. Strategic reports on the state of the environment usually list food security, freshwater and energy shortages, climate change and biodiversity loss as key environmental issues. The preservation of soils, as such, is usually not listed as a key issue but the way the soil resource is changing as a result of human (mis) management certainly is such a key issue, because it has major effects on each of the five recognized environmental issues mentioned above. Perhaps soil science has been too much inward looking, paying inadequate attention to documenting these major effects. Also, when scanning soil science literature, emphasis tends to be on erosion,- degradation and pollution of soils, representing soil change in the wrong direction. But soil research has also been quite successful in developing soil conservation and soil improvement PAGE 98 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
techniques that can turn around degradation processes by applying innovative soil management. Even though much knowledge and expertise is available, soil degradation in various forms still proceeds at an alarming scale. How to raise more awareness and achieve soil change in a positive direction? The objective of this paper is to discuss this issue by analysing: (i) the way in which various stakeholders, policy makers and citizens- at-large experience environmental issues and concerns; (ii) the role soil science plays in framing these issues; (iii) ways to effectively participate in inter- and transdisciplinary environmental research programs, and (iv) a case study where an attempt was made to apply some of the ideas being propagated here.
Environmental awareness in society Environmental awareness has increased considerably during the last 50 years, starting with the landmark book by Rachel Carson: ”Silent Spring” in the 1960’s, the Club of Rome in the 1970’s, emphasis on Sustainable Development following the Brundtland report in the 1980’s and the activities of the IPPC, studying climate change, also since the 1980’s. The, at least partly successful, UN Millennium Goals of the 2000’s have now been succeeded by the UN Sustainable Development Goals (SDG’s) and Targets for 2030. They reflect a broad societal context in which to consider environmental issues. When discussing soil change, I would propose we do so in the broad context of the SDG’s. The ten Goals can be summarized as follows( http://sustainabledevelopment.un.org): Goal 1 End extreme poverty including hunger Goal 2 Achieve development within planetary boundaries Goal 3 Ensure effective learning for all children and youth for life and livelihood Goal 4 Achieve gender equality, social inclusion and human rights for all Goal 5 Achieve health and wellbeing at all ages Goal 6 Improve agricultural systems and raise rural prosperity Goal 7 Empower inclusive , productive and resilient cities Goal 8 Curb human-induced climate change and ensure sustainable energy Goal 9 Secure ecosystem services and biodiversity and ensure good management of water and other natural resources. Goal 10 Transform governance for sustainable development. Each goal is specified by three specific targets, that are not mentioned here. It is easy to be cynical about what can be seen as unrealistic, wide ranging “shopping lists”. Still, they provide a context and specific targets and this is particularly valuable, it would seem, for a relatively small profession as soil science that appears to suffer from a lack of recognition in terms of their role in studying the major environmental problems of the immediate future. Earlier, in the introduction of this paper, we mentioned key environmental issues that are being recognized in international strategic documents. They all come back here, be it in a broad socio-economic context. Food security is key in goals 1 and 6, while of obvious importance for goals 2, 5, 8 and 9. Water is mentioned in goal 9 , where one can wonder what is meant by “other natural resources”. Could that be soils, perhaps? Climate and energy show up in goal 8 and biodiversity in goal 9. In summary, the suggestion is to frame our activities in the context of the SDG’s which offer a development context for the world up to 2030. This avoids looking back and puts emphasis on what can and must be done in future. PAGE 99 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Framing soil science in the SDG context Earlier, Bouma and Mc Bratney ( 2013) and Bouma ( 2014) have emphasized the need to more clearly demonstrate the important role that soils can play to ensure food security and freshwater availability and to combat climate change, energy shortages and biodiversity loss. This is not a difficult challenge: no soils, no crops. Soils are very effective in purifying percolating wastewater thereby protecting aquifers. Next to the oceans, soils contain more Carbon ( C ) than the vegetation of the world and increasing soil-C is a major route for climate change mitigation. Biofuels grow on soils and improved agricultural practices, avoiding losses of agrochemicals to air and water, are crucial for biodiversity. Rather than emphasize the importance of soils as such, it helps to emphasize the importance of soils for crop production, water purification, climate mitigation and biodiversity preservation. But only talking about this or presenting conceptual and theoretical papers is not going to carry the day. Only specific examples out there in the field that demonstrate successful efforts will be convincing. We need more of those. The SDG’s, as mentioned above, add an additional socio-economic dimension and it would be wise to keep the SDG’s in mind when reporting soil studies. For example, female farmers are dominant in many developing countries and agricultural development will therefore be favourable for gender equality ( goal 4). When considering the SDG’s, the various roles played by agriculture are quite prominent. Agriculture is crucial to end hunger and extreme poverty, it strongly affects the planetary boundaries of Rockstrom et al ( 2006) , population health, climate change, ecosystem services and governance, the latter if only by ownership issues related to land. This discussion is important for soil science as some soil scientists feel that the profession has been too closely related to agriculture in the past and advocate detachment from agriculture and a closer association with the geosciences. This can be done, but when striving for an increased societal profile for the profession, linking with the SDG’s would be wise and that implies, as discussed, a clear focus on agriculture. Finally, Bouma ( 2014) pointed out that the five key environmental issues, discussed above, should not be seen in isolation: improving food production ( the first issue among equals) may have positive effects on the other four issues and he advocates development of narratives in which all the issues are linked in a logical manner.
How to reach out to other disciplines and with stakeholders? Interdisciplinarity Too many papers and strategic documents contain abstract pleas for inter- and transdisciplinarity, the first term relating to work with other disciplines and the second to interaction with various stakeholders and policy makers. Of course, the SDG goals cannot be reached with disciplinary research alone but only talking about the issue of interdisciplinarity is not enough and we should face up to what this really means. It helps to realize that the hydrological, climatological, agricultural and ecological research communities are tightly organized, sometimes resembling closed shops. They, like us, have their own journals, meetings and research programs. Funds are scarce everywhere and colleagues from other disciplines are only being invited as members of interdisciplinary research teams if they can make crucial contributions. The challenge, therefore, is to produce such contributions. How to do that? In order to make predictions of future conditions, use of simulation models is common in all disciplines dealing with major environmental issues. Such models are gross simplifications of reality and simplifying input of adjacent disciplines is easier accomplished than simplifying its PAGE 100 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
own input. Soil scientists are often not members of the interdisciplinary teams and are therefore easy victims of simplification. The approach to take by soil scientists needs more discussion but should go beyond expressing a need for interdisciplinarity. I would suggest ( see also Bouma, 2014) to run their models with and without good soil data and compare the results. For example, many current hydrological basin-models don’t contain soil data or use it only in a elementary form. The hypothesis would be that using up-to-date soil data produces better results. But the verdict is still out as there appear to be no studies so far taking this approach. And important: use their models and not your own! Models are a symbol of scientific manhood and recognizing this will be helpful in communicating with colleagues in other disciplines. Transdisciplinarity Working with stakeholders, citizen groups, NGO’s and policy makers is complicated (see also Sayer et al, 2013). The old paradigm of scientists studying self-defined problems and producing “solutions” that are passed on to grateful citizens to be applied in practice is not realistic anymore, nor has it ever been. The relation between science and society is changing dramatically in the 21th century if only because of modern information technology. Lessons were learned in a major research program in the Netherlands on Sustainable Agriculture, as reported by Bouma et al ( 2011). The SDG’s and their specific targets present “wicked” problems for which no single, “magic” solutions exist. Complicated and cumbersome compromises have to be found balancing economic, social and economic considerations. Two major conclusions were reached in the Dutch study: (i) before starting any project, more time should be spent in defining stakeholders involved and their opinions, interests and hidden motives. As is, researchers often spend much time on formulating project proposals with minor contacts with stakeholders, followed by a quick start once the project has been funded; (ii) what really counts in the end are specific results. Just delivering reports and writing papers is not enough. The important role of “knowledge brokers” was identified in trying to realize implementation of plans. They are, ideally, members of the scientific team and have a high social intelligence being able to make the right injections of knowledge at the right time and place to move things along. As is, we often ignore this implementation phase and move on to the next project. As many , if not all, environmental issues are land related, soil scientists would appear to be in an excellent position to act as “knowledge brokers”, the more so since soil surveys in the early phase of the profession involved extensive contacts with land users.
A case study reflecting an inter- and transdisciplinary struggle. In the early 1990’s dairy farmers in the Northern Frisian Woodlands (NFW) in the Netherlands challenged environmental legislation on ammonia emission from manure spread at the soil surface, that was believed to have adverse effects on nature quality in adjacent nature areas. The NFW area was designated as a “national landscape” because of its small scale character with hedgerows separating relatively small, elongated fields. To reduce emissions, farmers were required to inject liquid manure in the soil and they did not like that because they expected damage to soil structure and loss of control because the heavy machinery to be used was operated by independent contractors. Besides, operating large machines on small fields offered operational problems. They proposed an alternative by feeding their cows a low-protein diet resulting in manure with less ammonia, as had been proved to be feasible elsewhere ( Sonneveld et al, 2008). This way they could realize comparable ammonia emissions as found when spreading manure at the surface. However, this was against the law and several farmers received fines of thousands of euro’s when they refused to inject their manure. This resulted in PAGE 101 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
a decade long struggle with legislators and researchers as reported by Bouma et al ( 2011) as case study no.1. In the late nineties the government allowed experimentation for a limited number of farmers but results were inconclusive, partly because of disagreements among researchers. As time went by, the farmers increasingly broadened the scope of their activities, embracing a “cradle-to-cradle” approach that was studied in 2010 with Life Cycle Analysis (LCA), results of which are to be published shortly. The “cradle-to-cradle” approach included use of less chemical fertilizers and external inputs and an increased application of animal manure, produced on-farm. A dramatic point was reached in July 2013 when the government decided that experiments would be terminated and that farmers had to abide by the law. Following an intensive political campaign , all parties (!) in the Dutch Parliament approved a motion asking the government to extend the experimental period for another five years for a large number of farmers. This is seen as defacto approval of the cradle-to-cradle approach and illustrates the importance of this issue on national level. The LCA study compared seven cradle-to-cradle farms (C) with seven comparable farms using more traditional procedures (T). In summary, both populations had very large standard deviations but results showed that: (i) C had significantly higher soil organic matter contents ( 186 versus 156 tons/ha); (ii) C used significantly less energy than T (4.9 MJ/kg milk) versus 4.3MJ; (iii) Values for global warming-, acidification- and eutrofication potential were lower for C but not significantly so. Average nitrate contents of groundwater were 12 mg/l for (C) and 22 mg/l for ( T). No significant difference and both below the EU threshold of 50 mg/.l, and (iv) Average income for C farms was 30% higher than for (T) but standard deviations were very high and the difference was not significant. The NFW study allows a narrative, as mentioned above, in terms of an efficient production system of milk providing a higher income to farmers, partly because of lower production costs, while water quality is improved, and the soil organic matter content increases, providing a contribution to climate mitigation. Energy use was significantly lower and lower emission of nitrogen compounds was favourable for biodiversity in surrounding nature areas. In addition ,hedgerows , that were maintained in the (C ) system, contributed significantly to biodiversity. In summary, all major environmental issues , as mentioned in the introduction, are being served by the ( C) system under the overall banner of food production and this conclusion can be extended to the SDG’s by recognizing that contributions are made to stay within the planetary boundaries of Rockstrom et al (2006) (Goal 2), that rural prosperity has increased ( goal 6), while serving goals 8 in terms of climate and energy and goal 9 in terms of biodiversity, water and soil ( another “natural resource”) . Goal 10 is intriguing. The NFW study showed government to be highly risk averse and unable to respond to bottom-up initiatives that, if embraced in a timely manner, would have supported government objectives. Also, concepts of environmental indicators, thresholds and proxies were poorly defined, requiring more transparancy (e.g. Bouma, 2011). The research community was divided and provided poor advice to regulatory agencies. In retrospect, more transparency and clarity of objectives could have resulted in better governance. But overall, soil change matters and it is clearly positive here because soil quality has improved as the higher organic matter content leads to a higher moisture supply capacity, better filtering capacity and higher soil biodiversity.
Lessons learned Soil science is a vital scientific discipline producing cutting-edge research in its various sub-disciplines. Its role in studying major environmental issues, often defined as food security, water and energy availability, climate change and biodiversity loss, appears PAGE 102 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
to be, however, less obvious to stakeholders, policy makers and the public at large. This should be improved because the societal relevance of soil science is not as high as it could and should be. The following suggestions are made (see also Bouma and McBratney, 2013, Bouma, 2014): (i) Taking a pro-active interdisciplinary approach in demonstrating the role of soils when studying the five major environmental issues, is advocated in the context of Sustainable Development Goals. (ii) Taking an active role in transdisciplinary efforts by developing a role as knowledge broker. (iii) Reconnecting the knowledge chain, linking tacit knowledge and field experience to basic soil research in both directions. This link was well developed in the past but appears to be broken now. (iv) Judging scientific quality not only by current quality indicators but also by its societal relevance, reflecting current discussions about the relation between science and society in the 21th century. (www.scienceintransition.nl) (v) Reflecting requirements of inter- and transdisciplinarity in soil education. (vi) Showing that Soil Change Matters by not only emphasizing negative changes, following erosion and degradation, but also by showing successful applications of innovative management leading to positive changes in soil quality. Only real examples in the field, such as the NFW study, are convincing.
References Bouma J (2011) Applying indicators, threshold values and proxies in environmental legislation: A case study for Dutch dairy farming. Env.Sci.and Policy 14, 231-238. Bouma J (2014) Soil science contributions towards Sustainable Development Goals and their implementation: linking soil functions with ecosystem services. J.Soil Fertility &Soil Sci. ( in press) Bouma J and McBratney AB (2013). Framing soils as an actor when dealing with wicked environmental problems. Geoderma 200-201, 130-139. Bouma J, Van Altvorst AC, Eweg R, Smeets PJAM, Van Latesteijn HC (2011). The role of knowledge when studying innovation and the associated wicked sustainability problems in agriculture. Advances in Agronomy 113, 285-314. Academic Press. USA. Rockstrom. J. et al (2009). A safe operating space for humanity. Nature 461: 472-475. Sayer J, Sunderland T, Ghazoul J, Pfund JL, Sheil D, Meijaard E, Venter M,Boedhihartono AG, Day M, Garcia C, Van Ooster C, and Buck LA (2013). Ten principles for a landscape approach to reconciling agriculture, conservation and other competing land uses. Proc. Nat. Acad. Sci. 110, 8349-8356. Sonneveld MPW, Schroder JJ, De Vos JA, Monteny GJ, Musquera J, Hol J, Lantinga EA, Verhoeven F, and Bouma J. (2008).A Whole-Farm Strategy to J. of Env. Qual., 37 (3), 333-337. PAGE 103 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The Global Soil Partnership Luca Montanarella1
1 European Commission - DG JRC, Via E. Fermi, 2749, I-21027 Ispra (VA), ITALY, luca.montanarella@ jrc.ec.europa.eu
Introduction The Global Soil Partnership (GSP) has been established, following an intensive preparatory work of the Food and Agriculture Organization of the United Nations (FAO)CL 145/LI in collaborationM/7 Rev.1 with the European Commission (EC), as a voluntary partnership 3 coordinated by the FAO in September 2011. The GSP is open to all interested stakeholders: Governments (FAO Member States), Universities, Research Organizations, c) support the acquisition of relevant soil knowledge and the implementation of targeted Civil Societyrese arOrganizations,ch in accordanc Industrye with nat andiona lprivate conditi ocompanies.ns and needs tIto isadd a rvoluntaryess applied partnership challenges on aiming towardsthe ground providing; a platform for active engagement in sustainable soil managementd) prom otande li nksoils be protectiontween exis tating all m ulscales:tilatera local,l initiat national,ives and bod regionalies to adv andanc global.e know lAsedg ae and “coalitionsc ofie nttheifi cwilling” underst anditowardsng of soilsoil iprotection,ssues, captur eit sattemptsynergies, whito makele taki ngprogress into acco inunt reversing the soil degradationexisting andwith ong thoseoing partnersworks and that effo rhavets that a ar genuinee being under will oftak protectingen at the mul soilstilater foral lourevel , and without duplicating or prejudging the work under the competent fora. futuree) generations. develop sust Itai nopenlyable so iaimsl mana towardsgement g creatinguidelines anfor enablingthe differen environment,t soils consider despiteing their the resistance of a minority of national governments, for effective soil protection in the large majority ofdev theelopm countriesent objec thattives areand genuinelydecisions; concerned about the rapid depletion of their limitedf) prsoilom resources.ote access to soil information and advocate the need for new soil surveys and data collection; GSP ing) Actionpromote investment and technical cooperation (including technology transfer) in all related soil matters to address fundamental issues in different regions; The GSPh) prpartnersomote in arestitu thetiona ownersl strengt henof theing andGSP. c aThereforepacity devel theopm GSPent of Plenary soil ins tassemblyitutions at local, is the mainnat idecisiononal, regi onamakingl and ibodynterreg ofional the lpartnership.evels; and It meets annually and is tasked with alli) pr relevantomote th decisionse necessary for publ theic aGSP,nd gov likeernm theent approval awarene ssof of the soi variousls throug planh rec ogofni action,tion of a World Soil Day and celebration of an International Year of Soils. the nomination of the Intergovernmental Technical Panel on Soils (ITPS) and the 5.coordination Compositio nof a nthed G Regionalovernanc Partnerships.e The GSP is supported by a Secretariat hosted 9.by FAOG (fig.over nanc1). e of the Global Soil Partnership is proposed to be composed of the following elements:
5.1 Partners 10. The GSP is a voluntary partnership, open to governments, international and regional organizations, institutions, and other stakeholders. 5.2 I ntergovernmental Technical Panel on Soils (I TPS) 11. The Intergovernmental Technical Panel on Soils (ITPS) shall provide scientific and technical advice on global soil issues to the GSP. 12. Members of the ITPS shall be experts appointed by the GSP Plenary Assembly for a term of 2 years, renewable for one additional term (with agreement of the GSP Plenary Assembly). The ITPS
PAGE 104 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 1: Organizational structure of the GSP.
The GSP is active in 5 main pillars of action: 1. Promote sustainable management of soil resources for soil protection, conservation and sustainable productivity 2. Encourage investment, technical cooperation, policy, education awareness and extension in soil 3. Promote targeted soil research and development focusing on identified gaps and priorities and synergies with related productive, environmental and social development actions 4. Enhance the quantity and quality of soil data and information: data collection (generation), analysis, validation, reporting, monitoring and integration with other disciplines 5. Harmonization of methods, measurements and indicators for the sustainable management and protection of soil resources
For each of these pillars detailed plans of action are developed with the help of specific working groups assembling the most relevant stakeholders and experts in relation to each of the relevant subjects. The highest scientific and technical level is guaranteed by the Intergovernmental Technical Panel on Soils (ITPS), as a core component of the GSP.
The Intergovernmental Technical Panel on Soils (ITPS) The ITPS is a scientific panel of 27 high level scientists selected according to a regional balance as follows: ›› five from Africa ›› five from Asia ›› five from Europe ›› five from Latin America and the Caribbean ›› three from Near East ›› two from North America ›› two from South West Pacific
Given its intergovernmental nature, panel members are nominated by the FAO Member Countries on the basis of a shortlist prepared by the GSP Secretariat (FAO). The mandate of the panel members is of two years and can be renewed for an additional term of two years (fig. 2). Panel members are acting as independent experts and are not bound to respond to their respective governments. They provide the necessary scientific and technical advice to the GSP as well as to other UN bodies requiring soil related expertise, like the UNFCCC, the CBD and the UNCCD. As a science policy interface, the ITPS operates as similar bodies for other policy areas, like IPCC for climate change, IPBES for biodiversity and the new SPI for desertification. PAGE 105 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 2: Members of the 1st Intergovernmental Technical Panel on Soils (ITPS) 2013-2015.
The ITPS, at its first meeting in July 2013 has identified its main activities and planned deliverables for its mandate of two years: Development of the background scientific and technical evidence towards the inclusion of soils within the sustainable development agenda post-2015, especially in the frame of the outcome of the Rio+20 sustainable development conference. Particularly the development of a soil related sustainable development goal (SDG) is addressed by this scientific and technical support activity. Revision and update of the World Soil Charter. The FAO World Soil Charter was adopted in 1981 by the 21st session of the FAO Conference and establishes a set of principles for the optimum use of the world’s soil resources, for improvement of their productivity, and for their conservation for future generations. The basic principles of the Charter are still valid today, but need in some parts some revision and updates in light of the new scientific findings and the rapidly changing environmental and social conditions of the world. A new updated version of the Charter will be submitted for endorsement in 2015. Compilation of the new World Soil Resources Report (WSRR). This report is intended as the flagship publication of the ITPS and should be aiming towards the establishment of a fully updated assessment of the status and trends of the global soil resources. It should be forming the reference for future policy decisions at global scale concerning soils and will provide an updated assessment of soil degradation processes after the last Global Assessment of the Status of human-Induced Soil Degradation (GLASOD) of 1990. It is planned to release the new WSRR in 2015, 25 years after GLASOD and in occasion of the UN International Year of Soils 2015. Endorsement of the plan of action for the 5 pillars of the GSP. As one of its main institutional tasks, the ITPS is mandated by the GSP plenary to develop and finalize the detailed plans of action for the five main pillars of the GSP. The plans of action PAGE 106 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
form the basis for the GSP activities and require a very detailed analysis of the current status and the necessary immediate actions on the priority areas identified by the GSP partners. The draft plans of action are prepared by specific working groups established by the GSP Secretariat that submit then the first draft versions to the ITPS for finalization. Once finalized and endorsed by the ITPS the GSP Secretariat (FAO) submits the five action plans to the GSP Plenary for final formal approval and subsequent implementation. Donors are encouraged to provide the necessary financial support to their implementation. A specific trust fund has been established by the GSP to facilitate the financial support by potential donors. Establishing relations with other science-policy interfaces like IPCC, IPBES and the SPI of UNCCD. The ITPS has the ambition to provide policy relevant scientific advice and information not only top the GSP, but also to other related UN bodies and conventions. It is well recognized by the climate change community that soils play a crucial role in the global greenhouse gas balance, especially as the major terrestrial carbon sink. Soil related scientific knowledge and information is needed by the UNFCCC to properly take decisions in relation to soils. The need for such data and information clearly emerged in the recent debate around the LULUCF issue. The special attention that this convention is now dedicating to organic soils, like peatlands, will require a substantial scientific input in the near future. Similar requirements are emerging also in other policy areas, like biodiversity and desertification. The recently established IPBES will require substantial support on soil related issues, especially on the recently decided fast track assessment on land degradation. The same applies for the recently established science-policy interface (SPI) of the UNCCD. The ITPS is open to collaborations with all these various organizations and stakeholders and formal contacts are on going between the GSP Secretariat and the various bodies on this matter.
Regional Soil Partnerships The actual implementation on the ground of the various plan of action of the GSP will have to be coordinated at Regional level. Soil properties and drivers and pressures are very different in the various parts of the world. Therefore there cannot be “one solution fits it all” approaches. Each action needs to be tailored and adapted to the local conditions. Therefore it is of crucial importance to activate Regional partnerships that can mobilize the existing stakeholders and organizations at national and local level. Seven Regional Soil Partnerships are proposed, following the FAO regional structure. But it is already evident that sub-regional partnerships will have to emerge in order to get closer to the actual realities at local level. Regional and Sub-Regional Partnerships have already been established for Europe, Eurasia, Asia, Near East and North Africa, West and Central Africa, East and South Africa, Mexico, Central America and the Caribbean, South America (fig. 3). More will follow in the near future, as the implementation of the GSP will get more and more operational.
Figure 3: Current areas of world covered by Regional Soil Partnerships (status February 2014). PAGE 107 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Conclusions and way forward The Global Soil Partnership is by now in its full operational phase. In less then two years an effective voluntary partnership has emerged facilitating all those stakeholders and governments that are genuinely committed to soil protection and sustainable soil management. This initiative has proved once more that if there is the will, actions can be implemented also at international and global level. Compared to lengthy processes in other more formal frameworks, this simple voluntary partnership was already able to get into operational mode in a very short time. All its planned bodies and structures are in place and fully operational, its action plans are in their final development phase and are soon ready for full implementation and substantial financial means have been already mobilized by donors. There is great hope that the GSP will trigger the needed pro-active thinking by all involved organizations towards achieving measurable progress in a short time frame on the vital issue of protecting our limited soil resources for future generations. Failing to do so will have catastrophic consequences for all of us. Recognizing that soil resources are of crucial importance for food security, climate change, biodiversity and many other ecosystem services is a first step towards soil protection and sustainable soil management. The recent endorsement by the UN General Assembly of the UN World Soil Day (5th of December) and the UN International Year of Soils 2015 is a major step towards building the framework for global awareness raising initiatives on this crucial issue. As a first achievement of the GSP, this is already an outstanding result of this very young, but already very effective partnership.
References Luca Montanarella, Ronald Vargas, Global governance of soil resources as a necessary condition for sustainable development, Current Opinion in Environmental Sustainability, Volume 4, Issue 5, November 2012, Pages 559-564, ISSN 1877-3435, http://dx.doi. org/10.1016/j.cosust.2012.06.007. PAGE 108 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Policy and science of soil change – a Victorian perspective Michael Crawford1, Jane Fisher2
1 Department of Environment and Primary Industries, PO Box 3100 Bendigo DC, Victoria 3554, michael.crawford@depi,vic.gov.au 2 Department of Environment and Primary Industries, PO Box 500 Melbourne, Victoria, 8002, jane. [email protected] Introduction The Victorian Government recognises that Victoria’s soils are vital to our continuing environmental, social and economic prosperity. Our soil currently generates over $9 billion worth of agricultural exports each year, accounting for 29% of Australia’s total agricultural exports and enhancing regional and local economies across the state. Victoria has a strong track record of soil management and the Victorian Government remains committed to maintaining and enhancing our soil to support our primary industries. The government also recognizes that healthy soil is the foundation of many key ecosystem services, including the protection of land, water and biodiversity and contributes directly to the habitat for Victoria’s celebrated flora and fauna. In short, Victoria’s soil resources generate many benefits and make a significant contribution to Victoria’s economic, environmental and social wellbeing. These benefits are both private and public in nature, and are transgenerational, in that the benefits exist for both current and future generations. As such, there are many parties that have a stake in the sustainable use, management and protection of Victoria’s soil resources. Not surprisingly, the interests of these parties can sometimes be in conflict, or the subject of market failure. Both of these situations justify government considering intervention. And when there is government intervention, or the possibility of it, there arises a need for policy to guide that intervention.
A Victorian perspective informed by history Victorian government policy, and in turn, investment and action, in relation to soil change and soil health outcomes has evolved over the decades in response to different drivers and philosophies. The current Department of Environment and Primary Industries (DEPI) was established in 2013 through the merger of the Department of Primary Industries (DPI) and the Department of Sustainability and Environment (DSE). Both of these departments had a role in soil health outcomes which in turn, partly reflected the predecessor organisations from which they arose. Roughly two thirds of Victoria’s land is managed privately, and the majority of this is in some form of agricultural land use. As well as its lead role in the management of public land, DEPI has a lead government agency role for management issues on private land, including soil health. DEPI makes significant investments in soil health in policy, research and practice change. There are strong linkages between policy, research and practice change in that all three components inform each other. The way in which research (or science) informs policy, and vice versa, in a Victorian context, is the main focus of this paper. The Department of Primary Industries (and historically the Department of Agriculture) always had a focus on supporting productive agriculture, and developing soil management practices that supported productive crop and pasture production. In doing so, there was often a focus on addressing production constraints, and various issues rose and fell in prominence over time (i.e. trace element deficiencies (1950s), acidification and waterlogging (1980s), soil structure (1990s), subsoil constraints (2000s), soil organic PAGE 109 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
matter (2010s)). The Department of Sustainability and Environment (and historically the Soil Conservation Authority) had more of a focus on soil protection and land degradation issues, partly from a productivity (private good) perspective, but just as importantly, from an environmental (public good) perspective. The soil erosion focus of the 1950s, a consequence of wind and water events which led to the formation of the Soil Conservation Authority), was superseded by the dryland salinity focus of the 1990s and the concurrent Decade of Landcare and subsequent various action plans for salinity. In the last decade, there has been a shift away from addressing problems to managing soils as complex systems that produce services that we benefit from: an ecosystem services approach (DSE 2012). This approach has been reinforced by the emergence of the Catchment Management Authorities (CMAs) as important delivery partners and determinants of government intervention at a local and regional scale. A recent seminal event in this policy evolution was a Parliamentary Inquiry conducted in 2004 by the Victorian Parliament’s Environment and Natural Resources Committee (ENRC) into the impacts and trends in soil acidity (ENRC 2004). This inquiry concluded that acid soils were a major economic and environmental challenge, and that a comprehensive acid soils management strategy was needed (with associated investment). The government response concluded that it was unwise to focus on a single issue (acidity) as that approach led to a single strategy response (liming), while other issues remained unresolved or unmanaged. The response, informed by close collaboration between DPI’s policy and science staff established that a better approach was to consider the soil resource holistically by focussing on the issue of soil health. The response postulated that the underlying causes of decline in soil health on private land, their impact on productivity, and interactions between soil, water and wind are often extremely complex and variable. It recognised that there is no universal formula for managing soil health and that an adaptive management approach is needed. These conclusions led to the development of DPI’s Soil Health Policy Framework for Productive Agriculture (DPI 2012). The purpose of this framework was to clarify the role of the former DPI in soil management for productivity on private land and to encourage the practical application of good policy principles to investment decisions by DPI. Along with the DSE Soil Health Strategy (DSE 2012), it continues to set the basis for DEPI’s interventions and investment in soil and soil health related research, extension, regulation and policy.
The current context The historical analysis shows that government intervention in soil management has been strongly influenced by the need to address problems, or to put it more strongly, crises (e.g. erosion, salinity). In some cases, government intervention, along with private action, has helped to ensure that the problem is likely to be less significant now than it was in the past. A key example of this is erosion. Work by DPI in measuring and promoting conservation farming practices in northwest Victoria reveal that the proportion of paddocks under PAGE 110 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
conservation management nearly doubled from around 44 per cent in 1996 to over 82 per cent in 2009 (and up from 0 per cent in the 1950s). This has greatly reduced the risk of wind erosion from these lands, and the occurrence of dust storms is now so much less than it was in past decades. In other cases, the crisis was either overstated (maybe because the science was wrong?) or the ability to address the crisis was overestimated. A key example of this is dryland salinity. Two decades ago, the ‘rising flood’ of dryland salinity was forecast to overtake large areas of productive agricultural land, and widespread efforts were instigated to prevent this. Today, in response to improved knowledge of groundwater systems and processes, the forecasts have been revised down significantly, and there is a better understanding of our ability (or inability) to influence these processes through soil and land management. The current accepted view is that salinity is a part of the landscape and that we need to learn how to live with salinity (DSE 2012). Government activity, informed by policy and science, now concentrates on realising opportunity and potential. In response to a range of drivers, Victorian government agricultural investment is driven by the policy of doubling food and fibre production by 2030. Whilst this may appear an audacious goal, analysis has shown that it is possible in some sectors, and that the benefits to the economy will be significant. However, to allow this opportunity to be realised, we need to ensure that all the contributing components are in place – and this includes a productive soil resource. There is also a realisation by most stakeholders that there are many win-win solutions in soil management. An output from better soil management (e.g. increased soil organic matter, reduced acidity) may lead to an outcome of improved productivity, whilst often also having a positive impact on environmental outcomes (e.g. increased soil organic matter means carbon has been sequestered in the soil). Further, because the majority of Victoria’s soil is under some form of agricultural management, achieving environmental outcomes requires interventions to be done in an agricultural context. Government interventions on private land must address productivity drivers to achieve both productivity and environmental outcomes.
The future context A key element of future activity in soil science in Australia is the National Soil Research, Development and Extension (RD&E) Strategy. Developed under the auspices of the National Primary Industries Research, Development and Extension Framework, this strategy brings together the various investor and provider partners in Australian soils RD&E. It has sought to identify what will be some of the ‘big’ policy issues in Australia and how they can be informed by improved soil data, information and knowledge. In turn, the strategy helps to identify the research priorities, the capabilities needed to address these priorities, and the current state of our capabilities in response to these needs. From a Victorian perspective, this is a key document in helping to shape the ongoing relationship between soil science and policy. PAGE 111 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
What does policy want from (soil) science? There are many theoretical approaches to the development of public policy, but they largely represent variations on the following seven-step framework (DPI, 2012): 1. Establish the nature and significance of the problem or opportunity 2. Identify potential market failures 3. Identify objectives and policy options to achieve them 4. Identify implications for private action 5. Assess policy options and developing key performance indicators 6. Identify and design delivery methods and prioritise resources 7. Design and conduct annual reporting and tailored evaluations. Government can use a variety of instruments and approaches to manage soil health. These include information generation and provision (R&D and extension), regulation, direct government provision of on-ground works, grants and cost-sharing arrangements (subsidies) and other market-based approaches (e.g. fees, levies, trading). Often a mix of these instruments is most appropriate. The distinction between legacy issues and current practices can also be important in determining which mix of policy tools is most appropriate. Regional variations in land use, soil condition, values and issues, along with partner resources and capabilities, mean there is likely to be benefit in setting specific objectives and developing responses in a differentiated manner in different regions. In Victoria, this is undertaken through the Regional Catchment Strategy (RCS) process. To be fully effective, it is necessary to have data, information and knowledge at the appropriate scale. When implemented fully and systematically, this approach represents an evidence- based, defendable approach to government intervention (or absence of) and leads to the effective and efficient use of taxpayer funds. However, there are some caveats to this statement, as often the full achievement of some of these steps, even the first one, is limited by the under application or poor application of scientific knowledge. Systematically working through the seven-step framework readily identifies the steps where (in a soils context) scientific knowledge can make a contribution to the policy problem. This process can generate questions such as in the list below. Policy wants to have answers to the questions. The questions can be asked at any time, and generally answers are expected as soon as possible. Potential questions include: ›› Where are the problems (or opportunities)? ›› Are the problems new and emerging, or have they been around for a long time? ›› How big (how widespread) is the problem? ›› Which industries are affected? What does the soil health problem cost agricultural productivity? ›› What is the problem’s past, current and future impact on public assets? ›› Does a solution exist, or does one need to be developed? PAGE 112 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
›› Is the solution cost effective? ›› Is the solution adoptable? What are the barriers to adoption? ›› Can data on soil condition be accessed quickly and easily? Can it be interpreted quickly and easily? ›› Is a visual (spatial) representation of problems by region and industry readily available? ›› How is the problem changing over time? How has it responded to past interventions? The practising soil scientist (and science manager) will quickly identify that the answers to some of these questions are not always readily available, often cost money and time to answer comprehensively, and frequently lead to more questions. The challenge therefore is to identify which of these questions is worth answering, and do they require that an answer be developed (i.e. through research and investigation) or are they of a nature that could be readily answered if the data and information was already available (i.e. through monitoring and reporting). An ongoing dialogue between policy advisors and scientists is critical to informing these decisions. Policymakers also look for consistency of knowledge and the ability of scientists to simplify the complex! Too often, soil scientists bamboozle policymakers with inconsistent and diverse approaches to issues (soil analysis, classification and mapping being classic examples) or provide answers that are laden with so many caveats and exceptions to be of no use at all.
What does (soil) science want from policy? Whilst science helps to inform policy, policy helps to inform science direction and intent, in that it provides the strategic context for the research. Scientific funding and effort is very much driven by clearly identified priorities that lead to the achievement of agreed outcomes. In addition to the strategic context that policy provides for science investment, policy also directly engages with science as described in the section above, and asks specific questions of science. What is helpful to science is if questions are specific, relevant and answerable (albeit with more research), as opposed to vague, unfocussed and unanswerable. Whilst some questions can be answered immediately, others take time. The identification of ‘emerging’ policy questions and the issues that are likely to be of increased concern in two to three years can help science to prepare to answer the specific questions when they need to be answered. Again, an ongoing dialogue between policy advisors and scientists is critical. To effectively support this, an ongoing investment in soil resources such as mapping, monitoring, archives and long term experiments is critical. Whilst Victoria has had good support in some areas of soil investment, it is behind ‘best practice’ in others, and this will impact upon the ability of science to adequately answer the questions of policy in the future. In Victoria, funding for initiatives like research into soil health is provided by a robust and competitive budget process managed through Cabinet. Ultimately, funding for soil research competes with other government priorities like health, education and infrastructure. Funding bids based on sound evidence about the nature of the problem and the potential benefit from the solution is critical to success in a competitive funding environment. PAGE 113 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Developing and maintaining ongoing collaborative discussions between soil scientists and policy makers is of mutual benefit as it is critical to ensuring that science is informed by policy needs, and that policy is aware of, and can maximise the benefit from new knowledge. Too often policy is informed of new scientific developments post-hoc and ad-hoc. Ideally, policy and science would form strong collaborative partnerships. The challenge for us all is how to do this, and by what metrics do we measure success?
References ENRC (2004) Inquiry on the Impact and Trends in Soil Acidity, Environment and Natural Resources Committee. Parliamentary Paper No. 59 of session 2003-2004, March. Victoria Parliament, DPI (2012) Soil Health Policy for Productive Agriculture. Department of Primary Industries, State of Victoria, Melbourne. DSE (2012) Soil Health Strategy – Protecting soil health for environmental values on public and private land. Department of Sustainability and Environment, State of Victoria, Melbourne. PAGE 114 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
WEDNESDAY 26 MARCH 2014 - TECHNICAL SESSION 3
Does soil organic matter influence functional soil properties? – A review of published information Brian Murphy123
1 In association with the Commonwealth of Australia (Department of the Environment) 2 In association with Australian Grains Research and Development Corporation 3 Honorary Scientific Fellow, NSW Office of Environment and Heritage, Cowra, NSW. brian.murphy@ environment.nsw.gov.au
Abstract A review has been undertaken into how soil organic matter affects a range of soil properties. The effect of varying the amount of soil organic matter on a range of individual soil properties was investigated using a literature search of published information largely from Australia, but also included relevant information from overseas. The soil properties considered included aggregate stability, bulk density, water holding capacity, soil erodibility, soil thermal properties, soil colour, soil strength, compaction characteristics, friability, nutrient cycling, cation exchange capacity, soil acidity and buffering capacity, capacity to form ligands and complexes, salinity and the interaction of soil organic matter with soil biology. Soil organic matter had clear effects on water holding capacity, cation exchange capacity, aggregate stability and buffering capacity to acidification. Soil organic matter also had a definite effect on the compaction and strength characteristics of soils which in combination with friability can determine how the soil responds to traffic and tillage. Soil organic matter was an important factor in providing a nutrient supply and in nutrient cycling, especially of nitrogen, but also of significant proportions of phosphorus and sulphur and other micronutrients. The relative effects of soil organic matter varied with texture, with soil organic matter generally being more critical in soils with lower clay contents.
Introduction Much of the recent interest in soil organic matter has been as a vehicle to increase soil carbon levels in order to trade carbon or to gain benefits in market based instruments for carbon. However, it can be argued that the potential effects of soil organic matter on the productive capacity of soils are also of practical and economic importance and of significant interest to many in the agricultural community. It is the objective of this review, to concentrate on the capacity of soil organic matter to be an agent to maintain and improve soil condition and soil health and so the productive capacity of the soil. While several reviews have been undertaken and suggest it is difficult to find quantitative evidence that soil organic matter improves soil properties, the approach taken here has been to investigate more directly the effects of soil organic matter on individual soil properties, especially through any pedotransfer functions that relate soil organic matter to the values of individual soil properties. It is clear that soil organic matter affects a range of soil properties and therefore any assessment of the influence of soil organic matter on productivity needs to consider this broad range of effects. In a final conclusion a paper that evaluates the overall effect of soil organic matter on overall productivity is reviewed. The detailed review is presented in Murphy (2013). PAGE 115 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil organic matter and functional soil properties A summary of the effects of soil organic matter on some of the major functional soil properties based on the review is given below. The summary is based on an evaluation of the published information and pedotransfer functions and relationships relating the soil properties to soil organic matter levels. Water holding capacity A range of pedotransfer functions that included soil organic matter as one of the variables were evaluated (eg Williams et al. 1992; Rawls et al. 1992). Several features of the relationship between soil organic matter and soil water holding capacity became evident. ›› As a general guide the plant available water between 10kPa and 1500kPa increased more for sandy soils with increasing soil organic carbon than for clayey soils. The general trend is shown below but this is only a very general guide, with more detailed data required for making specific recommendations: ›› Sandy soils by about 3 mm/100mm of soil for every increase of 1% soil organic carbon ›› Loam soils about 2.5 mm/100mm of soil for every increase of 1% soil organic carbon ›› Clayey soils about 2 mm/100 mm for every increase of 1% soil organic carbon. ›› The amount of water held at 10 kPa and at 1500 kPa increased with increasing soil organic matter although the amount of water held at 10kPa increased at a greater rate. ›› The increases in soil organic matter tend to be limited to the surface soils and probably the top 5 to 10 cm. This limits the capacity of increases in soil organic matter to increase the overall water holding capacity of a soil profile. ›› The changes in water holding capacity associated with increasing soil organic matter are complicated by the effects of soil organic matter on bulk density. ›› The predictions of the effects of soil organic matter on water holding capacity varied within the published literature and this is a reflection of the different data set and soils used to derive the different pedostransfer functions. Aggregate Stability While not a functional property itself, aggregate stability is a fundamental physical property of the soil that influences many other soil physical properties. Good aggregate stability is generally considered to be required to maintain good soil structure and a suitable soil physical condition of the soil for plant growth, infiltration and control of erosion. In general a level of soil organic carbon of 2 to 2.5% is considered necessary to maintain good aggregate stability (Kay and Angers 1999) and aggregate stability is considered to deteriorate rapidly when SOC falls below 1.2 to 1.5% (Kay and Angers 1999). Compaction characteristics and friability Several published relationships showed that the compaction characteristics of soils, friability and the Atterberg Limits (strong indicators of the compaction and tillage characteristics of soils) are all strongly affected by soil organic matter and from an agricultural and cropping viewpoint are improved by increasing soil organic matter. In general, a value of less than 1% soil organic carbon tended to result in a significant change in the compaction and friability characteristics of soils, but this requires further investigation as the characteristics improved with increasing soil organic carbon. PAGE 116 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil erodibility Soil erodibility is only one factor in determining the potential for water erosion with rainfall erosivity, length and degree of slope and land management factors such as cover and the presence of loosely tilled soil being other factors. However it when soil organic matter falls below 2% (SOC < 1.2%) soil erodibility is considered to increase (Rosewell and Loch 2002). Loch and Foley (1994) also showed the importance of good aggregate stability in reducing erosion and so this is related to soil organic matter and especially the amount of aggregates > 125 m. Nutrient Cycling A major feature of soil organic matter is that it holds a relatively constant amount of the different nutrients (Kirkby et al. 2011). As an example, this ratio of nutrients requires that for every 10 tonnes of soil organic carbon sequestered that it may be necessary to have 833 kg of N, 200kg of P and 143 kg of S (Himes 1997). This does vary between soils and with history of crops, pastures and land management, for example the amounts suggested by Williams and Donald (1957) for long term pastures are 645 kg of N, 44 kg of P and 254 kg of S. In the decomposition of the soil organic matter these nutrients can be released into the soil although how much becomes available to plants depends on a range of factors. In the formation and then in the decomposition of soil organic matter, a flux of nutrients or a recycling of nutrients is an outcome. Thus soil organic matter can be an important sink and source of nutrients in agricultural production. Nitrogen is a dynamic nutrient being continuously recycled between the atmosphere, the soil solution, soil organic matter, plant material and soil organisms. Soil organisms are responsible for the transformation of nitrogen between these different pools. Mineralisation in which nitrogen in soil organisms, plant material and soil organic matter is converted to mineral N (nitrate and ammonium ions) which is readily available to plants. How effective mineralisation is depends on the chemical characteristics of the substrate material including its C:N ratio and its lignin content. Substrates with high C:N ratios (generally cereals) are likely to result in N being fixed in the soil organism pool and unavailable to plants. Anaerobic conditions tend to favour the conversion of N to N20 and N2 gas rather than mineral N. A major source of N for plants and the formation of soil organic matter is the fixation of N by legumes. Legumes are able to fix between 70 to 180kg/ha/yr of N (Unkovitch et al.2010). When the plant materials from these legumes are added to the soil they can increase soil N levels. The legumes are only able to add these amounts of N to the soil if there are no limitations associated with soil acidity and P deficiency. Phosphorus has several sources and in many soils the P from soil organic matter may be only one source. It estimated that P from soil organic matter is generally about 40%, but this can vary from 20% to 80% depending on the soil type. Soil organic matter can prevent P being fixed into unavailable forms by the Fe and Al minerals (Moody and Bolland 1999) keeping it in a form which remains in the available pool for plants, even if this organic P needs to be mineralised before it becomes immediately available to plants. The mineralisation of plants and soil organic matter can be a major source of P for plants (McLaughlin et al. 2011). Sulfur is predominately supplied from the soil organic matter in most soils although some soils have a high mineral level of S such as those high in gypsum and some volcanic soils and some soils associated with marine deposits having been associated with acid sulphate soils. PAGE 117 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
The soil microbial population can be a critical factor in the effectiveness of nutrient cycling through soil organic matter. Maintaining soil biological health is an important factor in the functioning of soil organic matter (Watt et al.2006). Land management can influence the microbial populations in soils and their effectiveness in nutrient cycling to maximise the amount of nutrients that become available to plants (Watt et al. 2006; Murphy et al. 2011). Cation exchange capacity It is clear from the published information that soil organic matter can have an effect on cation exchange capacity but it is complex as it is dependent on the texture of the soil, not surprising), but also on the pH range of the soil. Much of the soil organic matter fraction that contributes to the cation exchange capacity has variable charge and this is why the effect of soil organic matter on CEC is pH dependent. Below pH 5.5, it appears that soil organic matter does not contribute greatly to CEC. For well-defined sets of soils, Hallsworth and Wilkinson (1957) and Chan et al. (1992) were able to define relationships between soil organic matter and CEC. The relationships were different for the different soil groups. For a chernozem group of soils Hallsworth and Wilkinson predicted that the CEC of the soil organic matter was 297 cmol(+)/kg, 134 for a group of acid soils, and Chan et al (1992) estimated the soil organic matter for red earths was 172. It would appear that the soil organic matter contributing most the charge contributing to the CEC is from the humus fraction. Buffering capacity to acidification The capacity of soil organic matter to buffer the soil against soil acidification has been recognised in both Queensland and NSW (Aitken et al. 1990; Helyar et al. 1990). Its capacity to buffer against acidification is also dependent on texture and clay content. While increasing the soil organic matter can increase the buffering capacity against soil acidification (possibly up to 3 tonnes of calcium carbonate by increasing soil carbon by 1 to 2%), in general this will only provide temporary relief from the acidification as the acidification rates will generally be sufficient to overcome the increased buffering by higher soil organic matter levels. However, this increased buffering capacity may be important in the short term for agronomic or economic reasons. Assessing the overall impact An estimate of the economic value of soil organic matter was undertaken by Ringrose Voase et al. (1997). In an investigation which included yield data from crops and pastures as well as soil data from 80 paddocks , the costs, returns, gross margins and soil properties were compared over 3 years (1992 – 1995). They looked at a series of soil landscape types that included soil types such as Red Chromosols derived from metasediments and Red Chromosols derived from parna. A standardised gross margin for each paddock year was used to estimate the relative influence of different biophysical variables on productivity. Overall growing season rainfall was found to be the strongest effect on gross margins, but soil organic matter was also found to have an effect in some of the major soil landscape types. In the aeolian soil landscape type with the parna derived Red Chromosols, for every 0.1% increase in soil organic carbon, the average gross margin was estimated to increase by $36.59 ± 26.51 /ha/yr based on current day values over the range of 0.9 to 1.9% soil organic carbon. These paddocks were largely cropping (70% cropping and 30% pasture) and crops included wheat triticale, lupins, field peas and canola. In the erosional soil landscape type with Red Chromosols on metasediments and granites on rolling hills PAGE 118 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
the main land use is pasture (68% pasture and 30% cropping). Analysis showed that for every 0.1% increase in soil organic carbon the average gross margins was estimated to increase by $18.86±16.59/ha/yr based on current day values, over the range of 0.9 to 2.6% soil organic carbon. The grazing enterprises included prime lambs, sheep for wool and cattle. The higher amount of cropping in the aeolian soil landscape type is probably the reason for the higher response to soil organic carbon compared to the erosional landscape type.
References Aitken, RL, Moody, PW and McKinley PG (1990). Lime requirement of acidic Queensland soils. I relationships between soil properties and pH buffer capacity. Australian Journal of Soil Research 28, 695 Chan, KY, Roberts, WP, Heenan, DP (1992). Organic carbon and associated soil properties of a red earth after 10 years of rotation under different stubble and tillage practices. Australian Journal of Soil Research 30, 71 – 83. Hallsworth, EG and Wilkinson, GK (1958). The contribution of clay and organic matter to the cation exchange capacity of soils. Journal of Agricultural Science 51, 1 – 3. Helyar, KR, Cregan, PD and Godyn, DL (1990). Soil acidity in New South Wales – current pH values and estimates of acidification rates. Australian Journal of Soil Research. 28, 523 – 37.Kay BD and Angers, DA 1999. Soil Structure. In ‘Handbook of Soil Science’. (Ed ME Sumner). ppA229 – A276. (CRC Press:Boca Raton, USA). Himes, FL, (1998). Nitrogen, sulphur and phosphorus and the sequestration of carbon. In R Lal et al. (eds) Soil Processes and the Carbon Cycle. CRC Press, Boca Raton, FL. Pp315 – 319. Kirkby, CA, Kirkegaard, Richardson, AE, Wade, LJ, Blanchard, C and Batten, G (2011). Stable soil organic matter: A comparison of C:N:P:S ratios in Australian and other world soils. Geoderma, 163, 197-208.McLaughlin, MJ, McBeath, TM Smernik, R, Stacey, SP, Ajiboye, B and Guppy, C (2011). The chemical nature of P accumulation in agricultural soils – implications for fertiliser management and design: an Australian perspective. Plant Soil 349, 69 – 87. Murphy, DV, Cookson, WR, Brainbridge, M, Marschner, P, Jones, DL, Stockdale, EA and Abbott, LK (2011). Relationships between soil organic matter and the soil microbial biomass (size, functional diversity and community structure) in crop and pasture systems in a semi-arid environment). Soil Research 49, 582 – 594. Murphy BW (2013). Soil Organic Matter and Soil Function – Review of the Literature and Underlying Data. Effects of soil organic matter on functional soil properties. Grains Research and Development Corporation and Commonwealth of Australia (Department of the Environment). Canberra, Australia. Rawls, WJ, Ahuja, LR and Brakensiek, DL (1992). Estimating soil hydraulic properties from soils data. In M. Th. Van Genuchten and FJ Leij (eds). Indirect methods for estimating hydraulic properties of unsaturated soils. Proceedings of International Workshop on Indirect Methods for Estimating the Hydraulic Properties of Unsaturated Soils. Riverside, California, October 11 – 13, 1989. US salinity Laboratory, Agricultural research Service, US Department of Agriculture, Riverside, California. Ringrose-Voase, AJ, Geeves, GW, Merry, RH and Wood, JT (1997). Soil indicators of changing land quality and capital value. CSIRO (Australia). Land and Water Technical Report 17/97. PAGE 119 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Rosewell, CJ and Loch, RJ (2002). Estimation of the RUSLE soil erodibility factor. In Neil McKenzie, Kep Coughlan and Hamish Cresswell (eds). Soil Physical Measurement and Interpretation for Land Evaluation. CSIRO Publishing, Collingwood Australia. Watt, M., Kirkegaard, JA, Passioura, JB (2006). Rhizosphere biology and crop productivity—a review Australian Journal of Soil Research 44, 299 – 317.Williams, CH and Donald, CM (1957). Changes in organic matter and pH in a podzolic soil as influenced by subterranean clover and superphosphate. Australian Journal of Agricultural Research 8, 179-189. Williams, J, Ross, P and Bristow, K. (1992). Prediction of the Campbell water retention function from texture, structure and organic matter. In M. Th. Van Genuchten and FJ Leij (eds). Indirect methods for estimating hydraulic properties of unsaturated soils. Proceedings of International Workshop on Indirect Methods for Estimating the Hydraulic Properties of Unsaturated Soils. Riverside, California, October 11 – 13, 1989. US salinity Laboratory, Agricultural research Service, US Department of Agriculture, Riverside, California.
Temperature sensitivity of CO2 emission from soil fractions: Implication for warming response in relation to soil texture Fan Ding1, Wenjuan Sun2, Yao Huang3
1 State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, No.20 Nanxincun, Xiangshan, Beijing, China, 100093, dingfan1985@ hotmail.com 2 State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, No.20 Nanxincun, Xiangshan, Beijing, China, 100093, [email protected] 3 State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, No.20 Nanxincun, Xiangshan, Beijing, China, 100093, huangyao@ibcas. ac.cn
Abstract It is well recognized that global warming promotes soil organic carbon (SOC)
decomposition, and thus soils emit more CO2 into the atmosphere under the warming, yet the response of SOC decomposition in different soil textures to the warming is
unclear. This limits our projection of SOC turnover and CO2 emission from the soils
under future warming. To investigate CO2 mission from soils with different textures, we conducted a 107-day incubation experiment. The soils were taken from forest and grassland. The incubation was conducted with three short-term manipulation cycles of changing temperature from 5ºC to 30ºC with an interval of 5ºC. Our results indicated
that CO2 emissions from sand (>50 μm), silt (2-50 μm) and clay (<2 μm) fractions
increased exponentially with increasing temperature. Sand fraction emitted more CO2
(CO2-C per unit of fraction-C) than the silt and clay fractions in both forest and grassland
soils. Temperature sensitivity of CO2 emission from soil fractions, expressed as Q10, decreased in the order clay>silt>sand. A further analysis of the incubation data showed an exponential decrease of Q10 with increasing temperature. Our results suggest that the decomposition of organic carbon in fine-textured soils rich in clay or silt is more sensitive to warming than those in coarse sandy soils, and the SOC would be more vulnerable in boreal and temperate regions than in subtropical and tropical regions under future warming. PAGE 120 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Carbon cycle in various ecosystems of Central Russia: vulnerability to repeated heat waves and droughts Lopes de Gerenyu V., Kurganova I., Myakshina T., Sapronov D., and Kudeyarov V.1
1 Institute of Physicochemical and Biological Problems in Soil Sciences of the Russian Academy of Sciences, Institutskaya street, 2, Pushchino, Moscow region, 142290, Russia, [email protected]
Under some scenarios, the regional climate projections show the amplification of drier conditions over most area of European Russia and Southern Europe, Asia and Northern America. The repeatability of soil droughts in these regions will increase due to the rise of air temperature and the decline in soil moisture during spring and summer. Based on observations provided by the Russian meteorological network, we carried out the analysis of medium (28-yr; 1973-2010) and short (14-yr; 1998-2012) climatic trends and estimated weather anomalies for the whole Central Federal district of Russia and separately for Moscow region. The experimental plots for carbon balance observations were located on the south of Moscow region (54°20-55’N, 37°34-37’E). The long-term soil CO2 emission (Esoil) was monitored weekly through 1998-2012 in 5 ecosystems different by land use and soil type (sandy Albeluvisols and loamy Phaeozems). The normalized difference vegetation index (NDVI) was used as a predictor of net primary production (NPP) for the same period (week resolution). Our calculations showed more significant increase of aridity during last 14 yrs compared to 28-yr period both for Central Russia and for the area studied. From the analysis fulfilled we conclude that the observed climate aridity in Central Russia may lead to the increase of C sink in ecosystems due to the decline in Esoil which is more significant than the decrease in NPP values. The soil texture and land use are the key drivers of vulnerability of carbon cycle in the temperate ecosystems to the current climate changes.
The effect of a changing rainfall pattern on soil carbon and nitrogen cycling in soil from pasture and afforested pasture Marianne Hoogmoed1, Shaun Cunningham 1, Patrick Baker2, Jason Beringer3, Timothy Cavagnaro4
1 School of Biological Sciences, Monash University, Victoria 3800, Australia. 2 Department of Forest and Ecosystem Science, Burnley campus, Melbourne University, Victoria 3121, Australia 3 School of Geography and Environmental Science, Monash University, Victoria 3800, Australia 4 School of Agriculture, Food and Wine, University of Adelaide, Waite campus, Glen Osmond, 5064, SA, Australia Email presenting author: [email protected]
Wetting and drying cycles of the soil are important drivers of soil carbon dynamics. As the climate is changing, longer dry periods are predicted while the individual rainfall events will be more intense. In attempt to mitigate climate change, afforestation is increasingly implemented throughout the world. This land-use change impacts the soil by changing the soil microbial community, nutrient status and soil structure, and thus potentially the soil’s response to wet/dry cycles. To be able to make predictions about the effect of afforestation on soil carbon sequestration, in a changing climate, we need to know how the soils carbon dynamics are affected by both land-use change, as well as wet/dry cycles. This paper presents results of a 3 month incubation study measuring CO2 respiration, C and N dynamics, and changes in the soil microbial community as affected by both land-use change and a reduction in wet/dry cycles. PAGE 121 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil biotic communities under shelter belts and the effect of plant functional type: what is driving soil change? Daniela Carnovale1,2, Geoff Baker2 Peter Thrall3 , Andrew Bissett3
1 Australian National University, College of Medicine, Biology & Environment, Fenner School of Environment and Society, Forestry Building 48 Linnaeus Way, Canberra ACT 0200, Australia 2 CSIRO Ecosystem Sciences, GPO Box, 1600, Canberra, ACT, 2601, Australia 3 CSIRO Plant Industry, GPO Box, 1600, Canberra, ACT, 2601, Australia
In Southeastern Australia, linear strips of planted trees and shrubs (shelter belts) are frequently established to restore ecosystem services that have been altered due to agriculture. Despite their wide use little is known about the effect of shelter belts on soil biotic diversity and how they alter aboveground – belowground interactions. This study aims to: 1) understand how shelter belts affect soil biotic communities and how time since establishment may drive soil community structure 2) explore aboveground - belowground linkages by investigating soil communities under dominate shelter belt tree genera (Acacia and Eucalyptus). We compared the abundance and structure of bacterial and fungal communities using quantitative PCR (qPCR) and terminal restriction fragment length polymorphism (T- RFLP), earthworm community composition and biomass, microbial biomass (C and N) and soil physiochemical properties between shelter belts, adjacent pasture and remnant vegetation. Findings demonstrate significant differences in the soil biotic composition in shelter belts compared with adjacent pasture sites. In terms of heterogeneity within shelter belts tree genus may contribute to community composition within the first 10 cm of the soil profile, while at greater depths soil biota tend to be influenced by abiotic soil properties and soil depth. These findings suggest that the conversion of agricultural land to shelter belt systems is driving changes in the soil biological community and in turn changes in edaphic soil properties. The high variability within belts and change in biotic communities with restoration reflects a range of soil and environmental conditions, which impacts on soil processes.
The characterisation of phospholipids and fatty acids in soil water extracts from a dairy farm in Gippsland, Australia using metabolomics. David Nash1, Michael William Heaven1, Thusitha Rupasinghe2, David De Souza2, Amsha Nahid3, Dedreia Tull2, Malcolm McConville2, Mark Watkins1, Murray Hannah1
1 Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Victoria, 3821, [email protected]; [email protected]; [email protected]; [email protected]. 2 Metabolomics Australia, Bio21 Molecular Science and Biotechnology Institute, 30 Flemington Road, The University of Melbourne Victoria, 3010, [email protected]; desouzad@ unimelb.edu.au; [email protected]; [email protected]. 3 Centre for Comparative Genomics, Murdoch University, 90 South Street, Murdoch, Western Australia, 6150, [email protected].
Abstract Organic phosphorus (P) makes up to 60% of total P in soils and accessing this reserve could lessen the need for inorganic orthophosphate fertiliser in agriculture PAGE 122 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
(Stutter, Shand et al. 2012). While the broad classes of organic P in soil are known (e.g. phospholipids, inositol phosphates), specific understanding of the composition of organic P in soil and soil water, and the associated biogeochemistry, hinders our progress towards more efficient fertiliser use. Metabolomics is a technique that can be used identify key compounds in soil water and therein the underlying P biogeochemistry in agricultural soils. Metabolomics were used to characterise phospholipids and associated fatty acids in soil water extracted from a field experiment in Gippsland. Based on a Grey Dermosol with Dermosolic Hydrosol soil in lower lying areas, the experiment had a resolvable incomplete block design with cultivated/uncultivated plots, two pasture types, three rates of P fertiliser addition and three blocks (36 plots). Using liquid chromatography/mass spectrometry (LC-MS) and gas chromatography/mass spectrometry (GC-MS) over 100 phospholipids compounds were characterised. Statistical analyses of the phospholipids identified 30 compounds with different concentrations (P<0.05) due to changes in cultivation practice but not pasture type or fertiliser addition. Phospholipids with14 to 38 carbon atoms and up to 6 double bonds were identified. Phospholipid classes in soil water included phosphatidylcholines (PC), the most common form of phospholipid in animals and plants, phosphatidylglycerols (PG), phosphatidylinositiols, and phosphatidylethanolamines (PE), a major form of phospholipids usually found in bacteria. Twelve phospholipids had odd numbered chains of carbon atoms, another indicator of a microbial or fungal source. It was shown that metabolomics could be used to identify phospholipids that could potentially act as markers of differing cultivation practices, different pasture types, and whether the plot was aerobic or anaerobic (i.e. bogged).
Introduction Phosphorus (P) in soils is used by plants for growth. While inorganic P (i.e. orthophosphate) has been extensively researched, organic P, which can be up to 60% of soil P (Stutter, Shand et al. 2012), is still relatively uncharacterised. “Organic P” is a large variety of different chemical species generated within plants, soil and soil biota. However, the characterisation of organic P has mainly been limited to 31P-NMR (Nash, Haygarth et al. 2014), which only broadly characterises molecular species (e.g. inositol phosphates). Metabolomic techniques offer the possibility of identifying key organic P species and their concentrations, thereby providing information on the underlying biogeochemistry Metabolomics combines advanced analytical analyses such as GC-MS and LC-MS with statistical techniques that can identify differences between samples or treatments, and the chemical species that are contributing to that separation. Moreover, the large amounts of data produced in metabolic studies can be data mined so that specific questions can be answered. In this research, we analyse soil water, extracted from soil from a dairy farm in Gippsland, south eastern Australia, for phospholipids and associated fatty acids. Phospholipids are an essential component of biological systems, commonly found in cell walls. Phospholipids consist of a phosphate head and one to two fatty acid arms (Fig. 1). While phospholipids are ubiquitous to living things, their structure (i.e. fatty acid chain length, functional group connected to the phosphate moiety) is characteristic of the organisms and processes by which they were created. For instance, generally phospholipids with odd-numbered fatty acid chains will have come from microbes (Řezanka and Sigler 2009). To understand how the concentration and types of PAGE 123 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
phospholipids change in response to different agricultural management this study utilised a field site where a range of soil conditions had been created using cultivation/no cultivation , different pasture types and different rates of P fertilizer addition.
O
N+ R= R1 O O Phosphatidylcholine (PC)
P R + NH3 O O O Phosphatidylethanolamine (PE) R2 O
, = R1 R2 CH3(CH2)x OH Fatty acid chain O OH
Phosphatidylglycerol (PG)
Fig. 1. Phospholipids identified in this study.
Material and Methods A resolvable incomplete block designed experiment was established (36 plots, 12 × 6 m) on a Grey Dermosol with Dermosolic Hydrosol soil in lower lying areas, on a commercial dairy farm at Poowong, Victoria, Australia (Fig. 2, (Watkins, Castlehouse et al. 2012)). The treatments were: Cultivation (C, mouldboard ploughed to 150 mm)/No cultivation (NC); Ryegrass (R, Lolium perenne, 25 kg/ha)/Mixed sward (M) of ryegrass (25 kg/ha) and clover (Trifolium repens, 5 kg/ha); and, P fertilizer Rate (Triple Super) (10, 35 or 100 kg P/ha annually).
During November 2012, ≥50 cores (20 mm depth) were taken from each plot. Soil water was extracted from each sample for physicochemical, LC-MS and GC-MS analyses. Physiochemical analyses included dissolved reactive P (DRP), total dissolved P (TDP), total dissolved N (TDN), ammonia (NH3), nitrate (NO3) and total N (TN) using flow injection analyses. Dissolved unreactive P (DUP) was calculated as the difference between TDP and DRP. Soil moisture, Oxidative-Reductive Potential (ORP), Dissolved Organic Carbon (DOC), pH and electrical conductivity (EC) were also performed on samples. PAGE 124 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Fig. 2. Resolvable incomplete block design used to create treatments at the experimental site. For metabolomic analyses, samples were freeze dried and a solution containing internal standards (1:1 butanol:H2O with internal standards (LC-MS: N-palmitoyl-d31-D-erythro- sphingosylphosphorylcholine, 1-palmitoyl-d31-2-oleoyl-sn-glycero-3-[phospho-L-serine], 1-palmitoyl-d31-2-oleoyl-sn-glycero-3-PC, 1-palmitoyl-d31-2-oleoyl-sn-glycero-3-PE; GC-MS: 13C-Myristic acid, 13C-Stearic acid)) was used to extract compounds from the powder. For LC-MS analyses, 500 µl of solution was dried in vacuo at 35°C and then resuspended in 100 µl of 50% 1:1 methanol:butanol + 10 mM ammonium formate. 2-5 µl of each sample was injected into a Agilent 6460 QQQ or 6550 Q-TOF LC-MS. For GC-MS analyses, 500 µl of solution was dried in vacuo at 35°C. Removal of water using methanol (100 µl) and drying steps occurred before samples were reconstituted (45 μL 2:1 chloroform:methanol) and placed in an Agilent 7890 GC coupled to a 7000 QQQ MS for analyses. Statistical analyses included ANOVA, principle component analyses (PCA) and partial least squares-differential analyses (PLS-DA), T-tests, boxplots and correlation charts. Tentative identification of the phospholipids and fatty acids was by comparison with internal databases using retention time and mass. Results and Discussion Of the 111 phospholipids found using LC-MS, 30 phospholipids were tentatively identified that were significantly different between cultivation treatments. There was no difference in phospholipids between plots with different pasture type or annual P fertilisation rate. The phospholipids were either PC or PE’s (14 each), along with two PG’s (Fig. 1 and Fig. 3, left). The concentration of the phospholipids ranged from 9.4-389 nM for PE’s and PG’s, and 98 nM-26 µM for PC’s. Most of the phospholipids had carbon chain lengths (14-38 carbon atoms) and number of double bonds (0-6 double bonds) assigned, though for four phospholipids, only mass could be determined. There were 12 odd chained phospholipids with the majority of these being greater in concentration in the non- cultivated plots. Odd-chained phospholipids are generally produced through microbial versus plant or animal sources (Řezanka and Sigler 2009). This is surprising as it is generally thought cultivation mixes up the soil, exposing organic matter and increasing microbial activity. There were an additional six even-chained PE, (i.e. phospholipids typically associated with microbes), with three of these higher in concentration in PAGE 125 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
NC versus C plots. With 60% of phospholipids being possibly microbially based, this suggests that cultivation may lower concentrations of possibly beneficial microbes at the surface (0-20 mm) while other sources of phospholipid (i.e. plants) are not significantly affected. This follows the same trend as studies of orthophosphate (i.e. inorganic P) where cultivation is known to bury P away from the surface (Nash, Webb et al. 2007). Further studies linking phospholipid concentrations with microbial composition would be needed to confirm this (e.g. quantitative Polymerase Chain Reaction (qPCR) analyses) hypothesis. A correlation chart of LC-MS versus physiochemical data revealed several relationships between nutrients and phospholipids (Fig. 3, left). A block of predominantly even- chained phospholipids (along with PE(35:5), PC(35:5) and PC(35:6)) were found to have a negative correlation with P and soil moisture. Likewise, the ratios of TDP/TP and TDN/TN were negatively correlated with phospholipids outside of this first group. The strongest positive correlation (0.6) was between PE(30:1) and DRP, TP and TDP. PE(30:1) is a common phospholipid created by bacteria (via syntheses of pentadecanoic acid, (Annous, Becker et al. 1997)) but the reason why it would be strongly associated with DRP is unknown, although it’s association with TDP and TP could be due to the organic P component of these analyses. The PCA of phospholipids found using LC-MS shows how they are segregated by treatment. The majority of samples can be separated by cultivation method (blue and orange circles) but not by pasture type, with R and M plots for both C and NC plots overlapping (Fig. 3, right). Of note, the PCA also revealed an outlier for one sample, as can be seen at the bottom right of the PCA plot (MC). This plot was boggy compared to other plots on the site and possibly had a different consortium of microbes. Analyses confirmed that the PE and PG phospholipids were lower in concentration for this plot. As these phospholipids are generally microbially sourced, it suggests that the sampling was preferential to aerobic rather than anaerobic microbes. Using GC-MS, a total of 20 fatty acids were identified from the samples. There were 14 even chained and 6 odd chained fatty acids with up to 5 double bonds. Concentrations ranged from 1.9 to 524 mmol. Fatty acids are breakdown products of phospholipids. The fatty acids were used to confirm the identity of the phospholipids by comparing fatty acid type and concentrations with phospholipid chain lengths and concentrations. PAGE 126 Color Key SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
-0.6 0 0.6 Value
PE (28:0) PE651 PE (29:0) PC (30:0) PE649 PC (31:0) PE (30:1) PC (36:5) PC (29:1) PE (14:0) PC (37:2) PC (35:2) e s t
PC (33:2) i PE647
PE (29:1) a b o l t
PC (35:6) e
PE (34:3) M PC (35:5) PC (36:1) PC 789 PC (38:6) PC (30:2) PC (32:1) PG (32:1) PG (32:2) PE (36:3) PE (36:5) PE (35:5) e 3 3 P P P P P P N N N H C C r T T H p T T O D D R R E u D O / / t T T N T N D O P s D / N i P D o D T R T m D
Physiochemical Measurements
Fig. 3. (Left) Correlation chart depicting correlations between phospholipids identified by LC-MS and physiochemical analyses. (Right) PCA of LC-MS data showing separation of NC(blue circle) and C (orange circle) treatments but not R and M treatments. In summary, a series of phospholipids have been shown to be different in concentration and type depending on cultivation methods. Further research into identifying the source of the phospholipids though a combination of metabolomic techniques and biological analyses (e.g. qPCR). Such studiesmay facilitate the identification of phospholipids that could act as biomarkers of the health of plants and/or the microbial communities within the soil matrix. Such markers could be used to monitor the effectiveness of changes to agricultural practices (e.g. the use of novel P “fertilisers” with complexes that extract organic P from the soil matrix (Stutter, Shand et al. 2012)).
References Annous BA, Becker LA, Bayles DO, Labeda DP, Wilkinson BJ (1997) Critical role of anteiso-C(15:0) fatty acid in the growth of Listeria monocytogenes at low temperatures. Applied and Environmental Microbiology 63(10), 3887-3894. Nash D, Haygarth PM, Turner BL, Condron LM, McDowell R, Richardson AE, Watkins M, Heaven MW (2014) Using organic phosphorus to sustain pasture productivity: a perspective. Geoderma In Press. Nash D, Webb B, Hannah M, Adeloju S, Toifl M, Barlow K, Robertson F, Roddick F, Porter N (2007) Changes in nitrogen and phosphorus concentrations in soil, soil water and surface run-off following grading of irrigation bays used for intensive grazing. Soil Use and Management 23(4), 374-383. Řezanka T, Sigler K (2009) Odd-numbered very-long-chain fatty acids from the microbial, animal and plant kingdoms. Progress in Lipid Research 48(3–4), 206-238. Stutter MI, Shand CA, et al. (2012) Recovering Phosphorus from Soil: A Root Solution? Environmental Science & Technology 46(4), 1977-1978. Watkins M, Castlehouse H, Hannah M, Nash D (2012) Nitrogen and phosphorus changes in soil and soil water after cultivation. Applied and Environmental Soil Science, 10. PAGE 127 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Using metabolomic techniques to identify compounds in soil water affected by changes to cultivation or pasture David M Nash1, Michael William Heaven1, James Pyke2, David De Souza2, Amsha Nahid3, Dedreia Tullb, Malcolm McConville2, Mark Watkins1, Murray Hannah1
1 Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Victoria, 3821, [email protected]; [email protected]; [email protected]; [email protected]. 2 Metabolomics Australia, Bio21 Molecular Science and Biotechnology Institute, 30 Flemington Road, The University of Melbourne Victoria, 3010, [email protected]; [email protected]; [email protected]; [email protected]. 3 Centre for Comparative Genomics, Murdoch University, 90 South Street, Murdoch, Western Australia, 6150, [email protected].
Soil water is the medium through which compounds in the soil interact with plants. However, there is relatively little research identifying compounds and the chemical and biological transformations occurring in soil water. This information could be used to promote soil health and optimise beneficial reactions. For instance, identifying organic phosphorus (P) compounds in soil water could help in developing technologies for accessing otherwise unavailable P. Metabolomics is commonly used in medical science for identifying key chemical and biological processes. Metabolomic techniques have been used to identify compounds in soil water extracted from a field experiment in Gippsland, Victoria. Based on a Grey Dermosol with Dermosolic Hydrosol soil in lower lying areas, the experiment had a resolvable incomplete block design with cultivated/uncultivated plots, two pasture types, three rates of P fertiliser addition and three blocks (36 plots). Using gas- and liquid-chromatography/ mass spectrometry over 2000 organic compounds were identified in the soil water samples. Twenty-two compounds were found to be significantly different between cultivation/ no cultivation and/or pasture types. A majority of these compounds were tentatively identified to have carbonyl (81%), amine (65%) or hydroxyl (56%) groups. Correlations between physicochemical properties and the relative concentrations of metabolites included positive correlation with P measurements and phosphates in regards to cultivation methods, and between electrical conductivity and floridoside, a sugar associated with plant heat stress. Metabolomic techniques were shown to be able to identify marker compounds that could possibly be used to estimate the health of the biological consortium in soil.
Introduction In agricultural production systems, soil water generally facilitates the transition of chemicals from the solid phase to the biological phase and vice versa. The chemical composition of soil (e.g. inositols) has been extensively studied. However, there is little research identifying compounds in soil water. Metabolomics combines advanced analytical chemistry and data mining techniques to identify chemical compounds (i.e. metabolites). This research utilised metabolomics to determine the effect of cultivation, pasture type and P fertiliser application rates on the chemical composition of soil water. Small organic molecules (m/z ≤ 1700 amu), which were thought to be indicative of biological and microbial processes were identified using liquid chromatography-mass spectrometry (LC-MS) and gas chromatography- mass spectrometry (GC-MS) and compared with physicochemical analytical results. PAGE 128 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Material and Methods A resolvable incomplete block designed experiment was established (36 plots, 12 × 6 m) on a Grey Dermosol with Dermosolic Hydrosol soil in lower lying areas, on a commercial dairy farm at Poowong, Victoria, Australia (Fig. 1). The treatments were: Cultivation (C)/No cultivation (NC): C plots were mouldboard ploughed to 150 mm; Pasture type: ryegrass (Lolium perenne, 25 kg/ha) (R), or a mixed sward of ryegrass (25 kg/ha) and clover (Trifolium repens, 5 kg/ha) (M); and, Annual fertilizer application rate: P fertilizer (Triple Super) added annually at a rate of 10, 35 or 100 kg P/ha. During September 2011, ≥50 cores (20 mm depth) were taken from each plot. Soil water was extracted from each sample for physicochemical, LC-MS and GC-MS analyses. Physiochemical analyses included dissolved reactive P (DRP), total dissolved P (TDP), total dissolved N (TDN), ammonia (NH3), nitrate (NO3) and total N (TN) using flow injection analyses. Dissolved unreactive P (DUP) was calculated as the difference between TDP and DRP. pH and electrical conductivity (EC) were also performed on samples.
Fig. 1. Resolvable incomplete block design used to create treatments at the experimental site. For metabolomic analyses, samples were freeze dried and a solution containing internal standards (200 µl butanol: 200 µl H2O with internal standards (13C-myristic acid, 13C6- sorbitol, 13C5,15N-valine)) was used to extract compounds from the powder. For LC-MS analyses, 400 µl of solution was dried in vacuo at 35°C and then resuspended in 100 µl of 50% methanol, 0.1% formic acid. 2 µl of each sample was injected into a Tandem 6520 Accurate Mass Q-TOF LC-MS. For GC-MS analyses, 100 µl of solution was added to deionised water (40 µL) prior to being centrifuged. The upper aqueous phase of the resulting biphasic solution was transferred and dried in vacuo at 35°C. Removal of water using methanol (100 µl) and drying steps occurred before samples being reconstituted (methanol, 50 μL) and placed in the GC-MS. Statistical analyses included ANOVA, principle component analyses (PCA) and partial least squares-differential analyses (PLS-DA), T-tests, boxplots and correlation charts. Tentative identification of different compounds involved comparing mass and retention time data with online and in-house databases. PAGE 129 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Results and Discussions Cultivation lowered soil water DRP (P = 0.01), TDP (P = 0.048) and DRP/TDP (P = 0.02) while DUP (organic P) was unaffected suggesting that the differences in DRP and TDP between cultivation treatments is attributable to orthophosphate. The only other cultivation effect was an increase in in the TDN/TN ratio (P = 0.006). Presumably this was due to cultivation mixing the nutrient and organic rich surface soil (0-20 mm) with soil from lower in the profile (i.e. 0-150 mm) resulting in the cultivated surface soil having lower P and N concentrations and an increased capacity to adsorb orthophosphate (Nash 2013). For the fertiliser treatments, DRP (P < 0.001) and DRP/TDP (P = 0.002) increased. These results presumably reflect the time-dependency of orthophosphate reactions in soil (Barrow 1989) and lowered soil P buffering at higher application rates and consequently higher soil water concentrations of orthophosphate. The ratios of DUP/DRP (P = 0.03) and DUP/TDP (P = 0.002) decreased with increasing fertiliser application rates but were likely due to DRP and TDP rather than the concentrations of organic P. PCA and PLS-DA of the physicochemical parameters displayed similar trends to the univariate statistics. For the PCA, the first two principal components explained 54% of the variation. The largest contributors to this variation were all P measurements as well as TN and TDN. C versus NC groups were more clearly separated when PLS-DA was used, with the first two factors predicting 40% of the separation between C and NC and accounting for 57% of the variation. TN, TDN and NO3 clustered around the NC predictor which is consistent with N being derived from the residual soil organic matter that remained on the soil surface for the NC treatments. Compared to the NC treatments, mouldboard ploughing had lowered surface soil organic matter (Nash 2013). Surprisingly, despite DRP and DRP/TDP being highly correlated with each other, along with them being inversely correlated with DUP/TDP and DUP/DRP, they were approximately at right angles to both the C and NC predictors, indicating there was little correlation between them and either predictor. At least in part this may reflect the differing fertiliser application rate treatments. LC-MS analyses of the samples produced over 2000 peaks among all samples. When PLS-DA was used, separate groups could be discerned when using cultivation and pasture treatments as predictors (Fig. 2). Using cultivation treatments C and NC as predictors, it could be seen that C and NC were separate groups with little overlap of points. The first two factors predicted 15% of the separation and accounted for 82% of the variation. This suggests that only a few compounds were responsible for the majority of the difference between C and NC treatments and probably reflects the overall lower soil organic matter and associated decomposition products for the C treatments. In comparison, using pasture and cultivation as predictors (i.e. MNC, RNC, MC and RC) factor 1 and 2 predicted 22% of the separation but only accounted for 36% of the variation. The RNC and MNC treatments were almost completely separate from each other and the C treatments. This could be because ryegrass has been shown to have a larger root system and rhizophere than clover that promotes beneficial fungal growth, improving aggregate stability (Tisdall and Oades 1982). PAGE 130 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Fig. 2: PLS-DA of LC-MS data. Left: Cultivation predictors, C (green circles) versus NC (red squares). Right: Pasture predictors, RNC, RC, MNC and MC (MNC = green squares; RNC = red circles; MC = blue cross; RC = black open circle). C = Cultivated, NC = Non- cultivated, R = Ryegrass only, M = Mixed ryegrass and clover.
T-tests further confirmed the PLS-DA results with 18 compounds that were significantly different between treatments (Fig. 3). Tentative identities of the different compounds were determined, with the majority of compounds containing carbonyl (81%), amine (65%), hydroxyl (56%), ether (44%), chloro (12%), phosphate (11%) and/or sulphate (6%) functional groups. A correlation chart between physicochemical data and the significantly different compounds showed a grouping of six compounds that were negatively correlated (-0.4 to -0.2) with the physicochemical parameters DRP, TDP, DRP/TDP, TN and TDN in the C-NC section (Fig. 3). The compounds possibly include organic acids, dicarbonyl compounds, the herbicide Chlorfenprop-methyl, and a short chain organophosphate. Two compounds positively correlated with DRP/TDP, DRP and TDP were possibly a long chain phosphate molecule or a phospholipid. These two metabolites were the only compounds in the C-NC section that were more concentrated in the NC than the C treatments which when coupled with their variable concentrations suggests that they may be associated with microbial activity. These results are interesting because it may have been expected that the higher organic matter for the NC treatments, which also had higher TDP, DRP and TN, would have yielded more significantly different compounds (i.e. there would have been a positive correlation) than the C treatments. Compounds different due to pasture treatments were negatively correlated to EC, DUP/ TDP, DUP/DRP, TDN/TN and DUP and positively correlated with DRP/TDP, DRP and TDP. These compounds possibly include amino acid-nucleotide complexes, aromatic sulphur compounds and a nucleobase. Increased microbial activity or differences in the microbial community structure in R versus M treatments could explain why the compounds are negatively correlated. The strongest positive correlation (0.65) was between N-(1-Deoxy-1-fructosyl)valine, a metabolite of the fungi Phaeosphaeria nodorum found associated with crops (Kim, Ferri et al. 2010) (m430.695t5.29, Fig. 2) and EC. Mycorrhizal association with fungi with similar characteristics to Phaeosphaeria nodorum have been known to help plants survive in saline soils, perhaps explaining why the concentration of this compound increases with increasing EC (Al-Karaki 2006). PAGE 131 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Fig. 3: Correlation chart of physicochemical data to versus LC-MS data. C = Cultivated, NC = Non-cultivated, R = Ryegrass only, M = Mixed ryegrass and clover; m = m/z, t = retention time (min).
GC-MS analyses found 93 compounds. Compounds were tentatively identified as sugars (46%), alcohols (18%), organic acids (14%) and amino acids (7%). PCA of the GC-MS data set was similar to the LC-MS data set in that there was no separation into groups by treatment. A PLS-DA of the data found that it could be separated by cultivation treatments with predictors C versus NC predicting 36% of the separation and 57% of the variation. T-tests found four significantly different compounds, all of which were greater in concentration in the C treatments versus NC treatments. The compounds included mannitol, floridoside, glucose and scyllo-inositol. Whether scyllo-inositol is formed via microbial metabolism as a scyllo-inositol phosphate, or is epimerized within plant and/or bacterial enzymes from myo-inositol, is unclear (Barrow 1989). Despite the thousands of peaks that were identified using LC-MS and GC-MS, metabolomic techniques were able to identify 22 compounds that take into account differences in cultivation and pasture. It is envisaged that monitoring compounds like these may help to assess the suitability of cultivation techniques in relation to the health of the soil, plants and animals. Moreover, further investigations using metabolomic techniques may pinpoint individual compounds that are indicative of soil health, function and processes. Limitations include correlating the chemical data with the microbes within the soil matrix. However, metabolomic techniques allow for the combining of analyses (e.g. quantitative polymerase chain reaction), with chemical techniques to correlate biomarkers with microbes. A site could then be monitored using the identified biomarkers, and a more tailored approach (e.g. adding beneficial microbes to the rhizosphere) could be used to optimise the condition of the soil for agricultural production. PAGE 132 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
References Al-Karaki GN (2006) Nursery inoculation of tomato with arbuscular mycorrhizal fungi and subsequent performance under irrigation with saline water. Scientia Horticulturae 109(1), 1-7. Barrow NJ (1989) Surface reactions of phosphate in soil. Agricultural Science 2(5), 33-38. Kim S, Ferri S, Tsugawa W, Mori K, Sode K (2010) Motif-based search for a novel fructosyl peptide oxidase from genome databases. Biotechnology and Bioengineering 106(3), 358-366. Nash D (2013) Profitable pasture production that protects our water quality. Final Project Report. Department of Environment and Primary Industries, Ellinbank, Victoria, Australia. Tisdall JM, Oades JM (1982) Organic matter and water-stable aggregates in soils. Journal of Soil Science 33(2), 141-163. PAGE 133 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
WEDNESDAY 26 MARCH 2014 - TECHNICAL SESSION 4
Soil carbon under perennial pastures; benchmarking the influence of pasture age and management Susan E. Orgill1, Nancy Spoljaric2 and Georgina Kelly3
1 NSW Department of Primary Industries, Wagga Wagga Agricultural Institute, NSW 2650, Australia, Email [email protected] 2 Monaro Farming Systems, PO Box 27, Bombala NSW 2632 3 NSW Department of Primary Industries, Level 2, 470 Church St, North Parramatta, 2151, Australia, Email [email protected]
Abstract This paper reports baseline soil carbon stocks from a field survey of 19 sites; 8 pairs/ triplet in the Monaro region of New South Wales. Site comparisons were selected by the Monaro Farming Systems group to demonstrate the influence of land management on soil carbon, and included: nutrient management, liming, pasture age and cropping history. Soil carbon stocks varied with parent material and with land management. The fertilised (phosphorus) native perennial pasture had a greater stock of soil carbon compared with the unfertilised site; 46.8 vs 40.4 Mg.C.ha to 0.50 m. However, the introduced perennial pasture which had been limed had a lower stock of soil carbon compared with the unlimed site; 62.8 vs 66.7 Mg.C.ha to 0.50 m. There was a greater stock of soil carbon under two of the three younger (<10 yr old) perennial pastures compared with older (>35 yr old) pastures. Cropped sites did not have lower soil carbon stocks at all sites; however, this survey was conducted after three years of above average annual rainfall and most sites had been cropped for less than three years. At all sites more than 20% of the total carbon stock to 0.50 m was in the 0.30 to 0.50 m soil layer highlighting the importance of considering this soil layer when investigating the implications of land management on soil carbon. Our baseline data indicates that nutrient management may increase soil carbon under perennial pastures and highlights the importance of perennial pastures for soil carbon sequestration regardless of age.
Introduction Well-managed, perennial pastures may increase soil carbon (C) agricultural soils compared with cropping systems (Chan et al. 2010). This is due to the extensive fibrous root systems of perennial pastures that may contribute more organic matter (OM) through root biomass and create drier soil conditions, and the minimal soil disturbance compared with most agricultural crops and cropping practices. In this study we used a paired-site approach to benchmark and compare; a) the age and management of perennial pastures on soil C stocks and b) the impact of changing land use from perennial pasture to annual cropping on soil C stocks. This paper presents baseline soil C (Mg.C.ha to 0.30 and 0.50 m) data for the Monaro region, New South Wales (NSW).
Methods Site location and sampling Nineteen sites were sampled in the Monaro region, southern NSW. The Monaro region is located 800 m above sea level with an average annual rainfall of 645 mm and is PAGE 134 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
classified as Cfa - Köppen-Geiger climate classification (Peel et al. 2007). Paired or triplet study sites were selected where the desired comparison was within 500 m. To be included in the paired comparison, sites were required to be on the same parent material (granite, sedimentary or basalt) and have similar soil and landscape attributes. Site comparisons were selected by the Monaro Farming Systems land-holder group to demonstrate the influence of management practices on soil C (Table 1). Sites with native perennial pastures (NPP) had never been cultivated and were typically wallaby grasses (Rytidosperma spp.), speargrasses (Austrostipa spp.) and snowgrass (Poa sieberiana). Introduced perennial pastures (IPP) were typically phalaris (Phalaris aquatica L.) and cocksfoot (Dactylis glomerata L.). Both NPP and IPP included exotic annual species such as subterranean clover (Trifolium subterraneum). Sites were sampled on a 25 x 25 m sampling grid according to SCRP protocols (Sanderman et al. 2011) using a hydraulic soil corer in late spring 2012. Ten cores were collected at each site to 0.50 m using a 40 mm diameter core and combined to form one composite sample for each soil layer (0.10 m increments) for each site. Four cores were collected to 0.50 m using a 75 mm diameter core for bulk density (BD).
Table 1 Comparison details and site history for each parent material class. Vegetation types include: native perennial pastures (NPP), introduced perennial pastures (IPP), crop and pine plantation. Soil treatments include: liming and phosphorus (P) application.
Parent material Comparison Treatment 2.5t/ha lime broadcast in 2002. IPP Granite IPP: Limed vs unlimed sown 1970. Crop (wheat): cropped since 2011 NPP vs <5 yr old IPP vs (previously IPP). New IPP sown Crop 2011 (previously cropped since 2009 from IPP). High P site: P management plan IPP: High P vs Low P since 2005. Low P; nil P.High P and low P application on native pasture IPP: Aspect North vs IPP sown 1989 and pasture Sedimentary South improved in 2010 IPP: <10 yr old vs >35 yr Old IPP sown in 1974. New IPP old sown in 2003. IPP sown in 1960. Crop (oats): >35 yr old IPP vs Crop vs cropped since 1998. Pine Pine plantation plantation established 2002. Crop (wheat): cropped since 2009 NPP vs <5 yr old IPP vs (previously NPP). New IPP sown Crop 2010 (previously cropped since 2004 from IPP). Crop (barley): first year crop Basalt NPP vs Crop (previously NPP). PAGE 135 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Analytical methods Total Carbon (TC g/100g) was determined on all samples using a LECO (CNS 2000) combustion furnace (Rayment and Higginson 1992; Method 6B3). Bulk density (BD) was determined for each core and each depth interval as described by Dane and Topp (2002). Results were calculated as BD in g/cm³ on an oven-dry basis. Results for this paper are reported as C stock in Mg.C.ha calculated by fixed depth C stock (FDCS); FDCS (Mg C ha) = TC (g/100g) x BD (g/cm3) x depth (cm) x gravel correction factor.
Results and Discussion Parent material significantly influenced the stock of soil C; with basalt- and sedimentary- derived soil having a significantly greater stock of C (P <0.05) in the 0 to 0.50 m compared with granite-derived soil; 77.5 (7.1 sd) vs 72.1 (17.2 sd) vs 54.9 (9.3 sd) Mg.C.ha respectively. The 0.30 to 0.50 m soil layer contained a considerable proportion of soil C with on average 31 %, 21 % and 24 % measured in the 0.30 to 0.50 m layer for basalt-, sedimentary-and granite-derived soil respectively. This highlights the importance of considering this soil layer when investigating the implications of land management on soil C stocks. The high P treatment under NPP on granite-derived soil had a greater stock of soil C compared with the low P treatment; 46.8 vs 40.4 Mg.C.ha to 0.50 m (Table 2). Granite- derived soil in the Monaro region are inherently low in P. Addressing the P requirements of pastures can increase above and below ground biomass and therefore OM supply to the soil (Chan et al. 2010) and can increase the stability of soil C (Kirkby et al. 2011). The limed treatment under IPP on granite-derived soil had a lower stock of soil C compared with the unlimed site; 62.8 vs 66.7 Mg.C.ha to 0.50 m (Table 2). This is consistent with Chan et al. (2011) and may indicate increased microbial decomposition of native soil C when acidic soil constraints are removed by liming. Interestingly, there was a greater stock of soil C under two of the three younger (<10 and <5 yr old) perennial pastures compared with older pastures (>35 yr old IPP or NPP). The <5 yr old pasture on sedimentary-derived soil which had more soil C (+29.5 Mg.C.ha to 0.50 m) compared with the older pasture was sown in 2010 and had been cropped for the previous 6 years. Therefore this greater stock of soil C may be explained by continued fertiliser (N,P,S) application during both the cropping and pasture phase and the three years of above average rainfall since pasture establishment in 2010. Two of the cropping sites; one on sedimentary-derived soil and one on basalt-derived soil, had a lower soil C stock compared with their perennial pasture pair; 54.6 vs 60.2 and 72.4 vs 82.5 Mg.C.ha to 0.50 m respectively (Table 2). However, there was no difference between the cropped and pasture site on granite-derived soil and there was a greater stock of soil C under the cropped site compared with the NPP on the remaining sedimentary-derived soil; 60.6 vs 68.0 Mg.C.ha to 0.50 m (Table 2). PAGE 136 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 2 Carbon stocks (Mg/C/ha) for the 0 to 0.30 and 0 to 0.50 m soil layers. Underlined treatments were used to calculate differences in carbon stock (Mg/C/ha) for a given pair/triplet.
Comparison Parent material C stock (Mg.C.ha) class and Treatment (Mg.C.ha) comparison 0 - 0.30 m 0 - 0.50 m 0 - 0.30 m 0 - 0.50 m Granite derived soil NPP: Low P vs High NPP Low P 32.92 40.44 P NPP High P 36.29 46.08 (+) 3.4 (+) 5.6 IPP: Unlimed vs Unlimed 52.8 66.7 Limed Limed 46.9 62.8 ( -) 5.9 (-) 3.9 NPP 41.4 58.5 NPP vs <5 yr old IPP IPP <5 yr old 39.5 51.5 (-) 1.9 (-) 7.0 vs Crop (wheat) Crop 41.4 58.1 0.0 (-) 0.4 Sedimentary derived soil IPP: Northern vs Northern Aspect 47.3 56.4 Southern Aspect Southern Aspect 45.7 55.6 (-) 1.6 (-) 0.8 IPP: <10 yr old vs IPP <10 yr old 75.2 92.7 >35 yr old IPP >35 yr old 73.6 96.6 (+) 1.5 (-) 3.9
IPP >35 yr old vs IPP >35 yr old 45.2 60.2 Crop (oats) vs Pines Crop 42.7 54.6 (-) 2.5 (-) 5.6 (10 yr old) Pines (10 yr old) 67.2 85.7 (+) 22.0 (+) 25.4 NPP 46.5 60.6 NPP vs <5 yr old IPP IPP <5 yr old 70.3 90.1 (+) 23.8 (+) 29.5 vs Crop (wheat) Crop 52.1 68.0 (+) 5.7 (+) 7.4 Basalt derived soil NPP 61.3 82.5 NPP vs Crop (barley) Crop 46.8 72.4 (-) 14.5 (-) 10.1
Conclusion Soil C stocks in the Monaro region varied with parent material and with land use and management. Our baseline information from 19 sites; 8 pairs/triplets, in the Monaro region of southern NSW indicates that nutrient management may increase soil C stocks under perennial pastures. This field survey also highlights the importance of perennial pastures, regardless of pasture age, for the accumulation of C in soil. In this field survey cropping did not decrease soil C stocks at all sites and cropping may therefore offer an opportunity for landholders to diversify their enterprise without compromising soil PAGE 137 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
C in the long-term. However, three of the four cropped sites had only been cropped for less than three years, and these three years coincided with three years of above average annual rainfall in the Monaro region. Furthermore, the vulnerability of C to decomposition (that is, C fractions) was not investigated in this study and it suspected that under different previous rainfall circumstances these results may have been different.
Acknowledgements The co-operation of the land holders in the Monaro region whose properties were sampled and who provided extensive land management information and continuing interest is gratefully acknowledged. The authors would like to thank Robert Smith (NSW DPI) for field assistance. This project is funded through the Action on the Ground program (Round 1) as part of the Australian Governments Carbon Farming Futures initiative.
References Chan KY, Conyers MK, Li GD, Helyar KR, Poile G, Oates A, Barchia IM (2011) Soil carbon dynamics under different cropping and pasture management in temperate Australia: Results of three long-term experiments. Soil Research 49(4), 320-328. Chan KY, Oates A, Li GD, Conyers MK, Prangnell RJ, Poile G, Liu DL, Barchia IM (2010) Soil carbon stocks under different pastures and pasture management in the higher rainfall areas of south-eastern Australia. Soil Research 48(1), 7-15. Dane JH, Topp CG (Eds) (2002) ‘Methods of soil analysis: Part 4 Physical Methods.’ Agronomy ; no. 9 (Soil Science Society of America, Inc: Madison, Wis. USA) Kirkby CA, Kirkegaard JA, Richardson AE, Wade LJ, Blanchard C, Batten G (2011) Stable soil organic matter: A comparison of C:N:P:S ratios in Australian and other world soils. Geoderma 163(3–4), 197-208. Peel MC, Finlayson BL, McMahon TA (2007) Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth Systems Sciences 11, 1633-1644. Rayment GE, Higginson FR (1992) ‘Australian Laboratory Handbook of Soil and Water chemical methods.’ (Inkata Press: Melbourne) Sanderman J, Baldock J, Hawke B, MacDonald L, Massis-Puccini A, Szarvas S (2011) National Soil Carbon Research Programme: Field and Laboratory Methodologies. CSIRO, Adelaide. PAGE 138 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Influence of cropping practices on soil organic carbon: evidence from three long-term field experiments in Victoria, Australia Fiona Robertson1, Roger Armstrong2, Debra Partington1, Roger Perris2, Ivanah Oliver1, Colin Aumann3, Doug Crawford4, David Rees5,
1 Department of Environment and Primary Industries, Private Bag 105, Hamilton, Vic, 3300, Australia. [email protected] 2 Department of Environment and Primary Industries, 110 Natimuk Rd, Horsham, Vic, 3400, Australia. 3 Department of Environment and Primary Industries, 255 Ferguson Rd, Tatura, Vic, 3616, Australia. 4 Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Vic, 3821, Australia, 5 Department of Environment and Primary Industries, 32 Lincoln Square North, Carlton, Vic, 3053, Australia.
Abstract Increasing the storage of organic carbon (SOC) in agricultural soils by using improved management practices is a widely suggested way of helping to mitigate rising atmospheric carbon dioxide concentrations whilst sustaining agricultural productivity. For dryland cropping systems, the most commonly proposed practices for enhancing SOC are reduced tillage, retention of harvest residues and alternative crop rotations. However, predicting the response of SOC to these practices is difficult because of the slowness of change and the apparent site-specificity of the effects, so that scientists must rely on simulation models that can accommodate long time frames and complex interactions among climate, soil and management. One of the most useful functions of long-term field experiments is their ability to provide data against which simulation models can be tested. In order to compare SOC stocks under various tillage, residue management, and rotation treatments, we sampled three long-term (12, 28 and 94 year old) field experiments in Victoria’s main cropping regions, the Mallee and the Wimmera. Soils were sampled to 30 cm depth in 10 cm increments. Organic carbon was measured by dry combustion and bulk density estimated from sample mass and volume. Our hypotheses were that SOC stocks are increased by: (1) residue retention, relative to residue removal, (2) zero- and minimum-tillage rather than conventional tillage (3) continuous cropping, rather than crop-fallow rotations and (4) pasture phases in rotations, relative to continuous cropping with grains crops. We discuss our findings and their significance for mitigating greenhouse emissions, sustaining soil fertility, and testing simulation models. PAGE 139 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
From rags to riches: the story of carbon, nutrients and pasture with dairy compost application Jess Drake1,2, Tim Cavagnaro1,2, Tony Patti1, Kevin Wilkinson3, Declan McDonald3, Pree Johnston1, Katrina Wilson1, Mick Rose1, Roy Jackson1
1 School of Chemistry, Monash University, Clayton Campus, Melbourne VIC 3800, jessica.drake@ monash.edu 2 School of Biological Sciences, Monash University, Clayton Campus, Melbourne VIC 3800 3 Department of Environment and Primary Industries, Victoria, Melbourne VIC
Around the world, dairy farmers are transforming dairy waste to compost for land application. In southeastern Australia, farmers are using composted dairy waste to increase production and reduce costs. In addition, the farmers are considering the benefits of compost for increasing sequestration of soil carbon, and on-farm nutrient retention. The “Carbon Farming Initative” in Australia is exploring the option to allow farmers to trade Carbon Credits for carbon stored in the soil. Compost also retains vital nutrients, such as N, on farm rather than importing N in the form of mineral fertilisers. Composting also reduces greenhouse gas emissions, such as CH4, compared to when stored in effluent ponds. This project will investigate if dairy compost applied to pasture improves carbon sequestration, nutrient retention and pasture production. In this project dairy compost, made from dairy effluent, feedpad waste, spoilt sillage and wood mulch, was applied onto a 1Ha field and companion plots at a rate of 0, 3, 6 and 12 t/ha. The field plot is open to grazing and normal farm management practices. The companion plots are being subjected to simulated grazing (mowing). The trials, currently underway will run for 18 months. Along with preliminary soil carbon results, this work will also include preliminary data for total and plant available nutrients, and farm biomass production. The outcomes of this research, and benefits it finds for “Carbon Farming” and nutrient retention has practical, policy and economic applications for world wide markets.
Managing cattle for sustainable soil properties: interactions between stocking rates and rainfall Moran Segoli1, Steven Bray2, Diane Allen3, Ram Dalal3, Ian Watson1, Andrew Ash4, Peter O’Reagain5
1 CSIRO Ecosystem Services, Townsville, Qld 4814, [email protected] 2 DAFF Queensland, PO Box 6014, Redhill Rockhampton QLD 4702 3 Landscape Sciences (ESP), DSITIA, GPO Box 2454, Brisbane, QLD, 4001 4 CSIRO Climate Adaptation Flagship, Dutton Park, Qld 4102 5 DAFF, PO Box 976, Charters Towers, QLD, 4820 Introduction Understanding the drivers of the dynamics of plant available nutrients in soil is important for the productivity and sustainability of rangelands grazed for livestock production, as well as identifying potential for increasing carbon in soils. In different rangeland ecosystems, grazing can increase, decrease or not affect the plant available nutrients in soil (Pineiro, Paruelo et al. 2010). Although many studies have examined the effects of grazing on plant available nutrients worldwide, there is a lack of manipulated field trials addressing the effect of grazing in the dry tropics (but see (Allen, Pringle et al. 2013; PAGE 140 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Holt 1997; Pringle, Allen et al. 2011). One of the problems with studying the effect of grazing on plant available nutrients in soil is the slow response of the nutrients in soil to manipulations (Holt 1997). Therefore we used a 17 year grazing manipulation to examine the effect of grazing intensity on soil organic matter and soil mineral nitrogen.
Methods We used soil samples that were collected at a long-term cattle-grazing trial conducted at Wambiana station (20°34’ S, 146°07’ E), north Queensland, Australia. Mean annual rainfall of the area is 636 mm, with most of the rainfall (70%) falling between December and March (O’Reagain, Bushell et al. 2009). In 1997 ~100ha paddocks were established with different grazing pressure: 1) Moderate stocking rate (MSR) at ~8 ha per animal equivalent (450 kg steer) and 2) Heavy stocking rate (HSR) at ~4 ha per animal equivalent (O’Reagain, Bushell et al. 2009). There was considerable soil and vegetation heterogeneity within treatment paddocks, so to minimise this source of variability all soil samples were collected in a Eucalyptus brownii community on black sodosols and Yellow Kandosol in the MSR and HSR treatments sharing a common boundary (Pringle, Allen et al. 2011). The soil samples were collected in July 2008, March-April 2009 and July 2013. 25 cores of 0-10 cm depth were sampled from 1-ha square area. Soil samples were dried at 40˚C, large roots removed, soil crushed and passed through a 2 mm sieve. The dry soil was archived in sealed containers for further analysis. Available N (ammonium and nitrate-N) was extracted from the soil by shaking 8 grams of soil in 20 ml of potassium chloride (2 M KCl), shaken for 1 hr and filtered through No. 41 filter paper. Nitrate and Ammonium were determined in the KCl extracts by colorimetric methods (Best 1976; Keeney and Nelson 1982; Willis, Schwab et al. 1993). Organic matter (OM) content was determined by loss-on-ignition method (Segoli, Ungar et al. 2012). The effects of grazing intensity were subjected to a one-way ANOVA. When data violated ANOVA’s homogeneity of variances assumptions, reciprocal transformation was applied to the data (Zar 1999). When the distributions of the residuals were non-normal, a Mann- Whitney U-test was applied (Mann and Whitney 1947; Zar 1999). All statistical analyses were conducted with the STATISTICA 12.0 software (StatSoft Inc., Tulsa, OK, USA).
Results and discussion Soil organic matter was higher in the heavy grazing treatment in all years (Fig. 1), although this was only significant in 2009 (2008: one-way ANOVA, F(1,45) = 0.886, P = 0.07; 2009: one-way ANOVA, F(1,44) = 6.874, P = 0.01; 2013: one-way ANOVA, F(1,43) = 2.630, P = 0.11). The apparent positive effect of grazing on soil organic matter could be due to a preference of cattle to consume grasses and switch to shrubs and trees when grasses are limited. Therefore, in the heavy grazing treatment there will be a dominance of C3 woody trees and shrubs relative to C4 tropical grasses. This will cause higher sequestration of carbon into the soil (Pringle, Allen et al. 2011). Nitrate showed variation among the years (Fig. 2) and was significantly higher in the moderate grazing treatment of 2009 (2008: one-way ANOVA, F(1,41) = 0.603, P = 0.44; 2009: Mann-Whitney U-test, adjusted Z = -2.594, P < 0.01; 2013: Mann-Whitney U-test, adjusted Z = 0.054, P = 0.96). The variability among years in soil nitrate emphasises the relatively high intra- and inter- annual variability compared to differences in grazing management. For example, lower nitrate in 2009 compared to the other two years may reflect sampling season, since 2009 sampling was performed during the growing season PAGE 141 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
when nitrate uptake by plants would be high as opposed to the other years where the sampling was done in the dry season. Ammonium also showed variation among the years (Fig. 3), but without any significant difference between the grazing treatments (2008: reciprocal-transformed data, one- way ANOVA, F(1,41) = 0.002, P = 0.96; 2009: reciprocal-transformed data, one-way ANOVA, F(1,48) = 1.500, P = 0.23; 2013: reciprocal-transformed data, one-way ANOVA, F(1,31) = 1.494, P = 0.23). The variability among years in soil ammonium also emphasises the relatively high intra- and inter- annual variability compared to differences in grazing management. Our results suggest that heavy grazing does not have negative effects on the availability of soil nutrient in this system, and may even sequester higher amount of carbon in the soil. However, this should not be taken as a managerial recommendation since ecosystem function, productivity and sustainability are further controlled by additional factors (e.g. plant cover, leakage, plant community, erosion) all of which may be affected by heavy grazing. Furthermore, evidence exists that results are likely to vary according to soil type (Pringle, Allen et al. 2011), topographic unit (O’Reagain, Brodie et al. 2005), vegetation type (Richards, Brackin et al. 2012) and rainfall dynamics (O’Reagain, Bushell et al. 2009).
3
2 OM (%) 1
0 2008 2009 2013 Year
Figure 1. Soil organic matter under moderate (grey) and heavy (black) cattle stocking rates. Data are represented as means ± SE. PAGE 142 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
1.0
0.8 ) 1 - Median 0.6 25%-75% Non-Outlier Range
(mg N kg Outliers - 3 0.4 Extremes NO
0.2
0.0 2008 2009 2013
Year
Figure 2. Box plot of soil nitrate under moderate (grey) and heavy (black) cattle stocking rates.
12
10 ) 1
- 8
6 (mg N kg + 4 4 NH 2
0 2008 2009 2013
Year
Figure 3. Soil ammonium under moderate (grey) and heavy (black) cattle stocking rates. Data are represented as means ± SE. PAGE 143 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
References Allen DE, Pringle MJ, Bray S, Hall TJ, O’Reagain PO, Phelps D, Cobon DH, Bloesch PM, Dalal RC (2013) What determines soil organic carbon stocks in the grazing lands of north-eastern Australia? Soil Research 51(7-8), 695-706. [In English] Best EK (1976) An automated method for determining nitrate-nitrogen in soil extracts. Queensland Journal of Agricultural and Animal Sciences 33(2), 161. Holt JA (1997) Grazing pressure and soil carbon, microbial biomass and enzyme activities in semi-arid northeastern Australia. Applied Soil Ecology 5(2), 143-149. [In English] Keeney DR, Nelson DW (1982) Nitrogen - inorganic forms. Methods of soil analysis. Part 2. Chemical and microbiological properties, 643-698. [In English] Mann HB, Whitney DR (1947) On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. 50-60. [In en] O’Reagain P, Bushell J, Holloway C, Reid A (2009) Managing for rainfall variability: effect of grazing strategy on cattle production in a dry tropical savanna. Animal Production Science 49(2), 85-99. [In English] O’Reagain PJ, Brodie J, Fraser G, Bushell JJ, Holloway CH, Faithful JW, Haynes D (2005) Nutrient loss and water quality under extensive grazing in the upper Burdekin river catchment, North Queensland. Marine Pollution Bulletin 51(1-4), 37-50. [In English] Pineiro G, Paruelo JM, Oesterheld M, Jobbagy EG (2010) Pathways of Grazing Effects on Soil Organic Carbon and Nitrogen. Rangeland Ecology & Management 63(1), 109-119. [In English] Pringle MJ, Allen DE, Dalal RC, Payne JE, Mayer DG, O’Reagain P, Marchant BP (2011) Soil carbon stock in the tropical rangelands of Australia: Effects of soil type and grazing pressure, and determination of sampling requirement. Geoderma 167-68, 261-273. [In English] Richards AE, Brackin R, Lindsay DAJ, Schmidt S (2012) Effect of fire and tree-grass patches on soil nitrogen in Australian tropical savannas. Austral Ecology 37(6), 668-677. [In English] Segoli M, Ungar ED, Shachak M (2012) Fine-Scale Spatial Heterogeneity of Resource Modulation in Semi-Arid “Islands of Fertility”. Arid Land Research and Management 26(4), 344-354. [In English] Willis RB, Schwab GJ, Gentry CE (1993) Elimination of Interferences in the Colorimetric Analysis of Ammonium in Water and Soil Extracts. Communications in Soil Science and Plant Analysis 24(9-10), 1009-1019. [In English] Zar JH (1999) ‘Biostatistical Analysis.’ 4th edn. (Prentice Hall: Upper Saddle River) 929 PAGE 144 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Savanna soil changes: The effects of land use change on savanna soils of northern Australia Samantha Grover1, Stephen Livesley2, Lindsay Hutley3, Stefan Arndt4, Jason Beringer5
1 Department of Agricultural Sciences, AgriBio, La Trobe University, 5 Ring Road, Bundoora, VIC 3086, [email protected] 2 Department of Resource Management and Geography, The University of Melbourne, 500 Yarra Boulevard, Richmond, VIC 3121 3 Research Institute for the Environment and Livelihoods, Charles Darwin University , NT 0909 4 Department of Forest and Ecosystem Science, The University of Melbourne, 500 Yarra Boulevard, Richmond, VIC 3121, 5 School of Geography and Environmental Science, Monash University, Clayton, VIC 3800
Savanna ecosystems cover a quarter of the Australian continent. This area constitutes the largest intact savanna in the world, as little agricultural development has occurred in Australia’s north. There is a continued push at the national policy level to develop the perceived agricultural potential of northern Australia and it’s high rainfall. Little consideration has been given to the poor quality of soils, which tend to be high in sand and low in nutrients. To gain some insight into what agricultural development of north Australian savannas might mean for soil health, we sampled soils in the Douglas Daly region of the Northern Territory, where some of the earliest agricultural development occurred. We chose three replicate sites of each of three land uses: uncleared savanna, young pasture (7 years) and old pasture (25 years). We took soil samples from two replicate pits at each site at 0-10, 10-20, 20-30, 40-50, 70-80 and 120-130 cm depth. Bulk density, %C, %N, pH, EC, CEC and P were measured. Initial analyses indicate that the pasture soils have increased in bulk density, total C and total N compared with the uncleared savanna soils. The change is confined to the top 20 cm. Mid infra-red analyses are being applied to predict the fractionation of C into charcoal, particulate organic matter and calcium carbonate. This may lend further insight into the likely duration of the observed increase in soil C. The observed increase in soil C must be considered in the context of the whole greenhouse gas balance of this land use change, including gas fluxes from the soil and changes in above-ground biomass. PAGE 145 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
WEDNESDAY 26 MARCH 2014 - TECHNICAL SESSION 5
What can legacy datasets tell us about soil quality trends? B.P. Marchant1, D.M. Crawford2, N.J. Robinson3
1 British Geological Survey, Nottingham, UK. [email protected] 2 Department of Primary Industries, Ellinbank, Victoria, Australia 3 Department of Primary Industries, Bendigo, Victoria, Australia
Abstract Globally there is a need for broad-scale monitoring of the soil to identify where its functionality is threatened. Broad-scale monitoring networks have been recently established in many countries but the vast majority of these have only completed one phase of sampling which can only be used to estimate the status of soil properties rather than to identify temporal trends. There is potential to use legacy soil data to quantify temporal change but some caution must be applied because these data were not collected for this purpose and therefore their design might be unsuitable. We focus on two legacy datasets from Victoria, Australia and explore whether they can be used to quantify changes in soil quality over the past few decades. One of these datasets consists of the results of soil analyses requested by farmers on more than 75 000 occasions since the early 1970s. The other one is a research survey of soil fertility from the 1970s which has recently been resampled by the Victorian DPI. In each case we use spatio-temporal statistical techniques to identify where significant temporal trends are evident in the datasets. We then discuss the extent to which we can be confident that these trends correspond to real changes in soil quality and do not reflect changes in other factors such as farmers’ willingness to request soil tests or the laboratory techniques. Finally we discuss the implications of the results in terms of the continued functionality of Victorian soils and the design of future monitoring efforts.
A pilot survey of change in soil pH using legacy sites D M Crawford1, N Robinson2
1 Farming Systems Research Division, Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Victoria, 3821, [email protected] 2 Farming Systems Research Division, Department of Environment and Primary Industries, Taylor Street, Epsom, Victoria, 3551 Abstract Soil acidification is a threat to the prime agricultural lands of Victoria’s dryland farming systems. Few recent surveys are available that report on acidification under crops and pastures. Sites from the National Soil Fertility Project (NSFP), representing Victoria’s major soils used for grazing or cropping, were first sampled in 1968-1972, and most recently in 2011-13, as reported here. These sites have provided an initially unintended opportunity to detect changes in soil properties over the last 40 years. Legacy reports and initial records were used to relocate the sites and apply the same methods as when PAGE 146 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
first assessed. At each site, typically 30 m by 50 m for a pasture field trial or 60 m by 100 m for a wheat field trial, soil was sampled to 1 m depth by taking 8 or 9 cores, sectioning them into 10 cm increments, and combining the increments to form a bulk sample representing each 10 cm depth. Samples were analysed to assess differences in soil pH (1:2.5 soil:water) between 1968-72 and 2011-13. Data analysis is confined to depths at sites where differences in electrical conductivity were less than 0.16 dS/m to remove artefacts from differences in salinity on measurements of pH using aqueous suspensions. Preliminary results for 45 wheat field trial sites showed that soil pH was significantly (P<0.05) more acidic in the 0-40 cm depth, by between 0.1 and 0.3 pH units, than when they were first sampled where the average pH ranged from 7.9 to 8.6.
100 Years of Superphosphate Addition to Pasture – Where did it go? Cassandra R Schefe1, Nathan Robinson2, Doug Crawford3, George Croatto4, Ron Walsh4, Tim McLaren5, Ron Smernik5
1 Department of Environment and Primary Industries, 124 Chiltern Valley Rd, Rutherglen, 3685, [email protected] 2 Department of Environment and Primary Industries, Cnr Midland Hwy & Taylor St, Epsom 3551 3 Department of Environment and Primary Industries, 1301 Hazeldean Rd, Ellinbank 3821 4 Department of Environment and Primary Industries, Macleod Ernest Jones Drive, Macleod 3085
5 School of Agriculture Food and Wine and Waite Research Institute, The University of Adelaide, Waite Campus, PMB 1, Glen Osmond, SA 5064, Australia.
Abstract The ‘Permanent Top-Dressed’ pasture (PTD) long-term experiment at DEPI – Rutherglen has received two rates of superphosphate addition for the past 100 years. The experiment was established in 1914 to demonstrate the value of adding phosphorus (P) fertiliser to increase pasture productivity for lamb and wool production. Of the original treatments which were established, only three of these have been maintained since 1914. These are: i) native pasture (control); ii) 125 kg superphosphate ha-1 applied every second year, and, iii) 250 kg superphosphate ha-1 applied every second year. To commemorate the 100th anniversary of this experiment, a detailed investigation was conducted to determine firstly, how the applied treatments have modified the nutrient and pH status of these treatments, and secondly, the forms of P present in the soil, and where it is physically located. The experimental site was surveyed using EM38 to capture the inherent variability of each plot. Representative bulked soil samples were then analysed for a range of soil chemical properties, including soil pH and Olsen P, which can be related back to historical records to determine changes with time. Soil physical fractionation (> 50 µm, < 50 µm) was conducted to determine the physical location of P and C within the soil structure, and NMR was used to determine the chemical forms of P and C present in the soil, and its fractions. The results from this work will be presented, demonstrating how a multi-faceted approach can provide new knowledge to a 100 year old question in P research – where did it go … in time, space and function. PAGE 147 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Can soil change be assessed for the Victorian dairy industry? Sharon R. Aarons1, Douglas Crawford2, Mark Imhof3, Cameron Gourley4
1, 2, 4 Farming Systems Research Division, Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, 3821, Australia, [email protected], douglas.crawford@depi. vic.gov.au, [email protected] 3 Farming Systems Research Division, Department of Environment and Primary Industries, 32 Lincoln Square North, Carlton, 3053, Australia, [email protected]
Abstract Meeting the increased demand for dairy products will require careful management of soils to minimise land degradation and sustain increased production. Key to providing farmers with the tools to manage their soils sustainably is firstly understanding the soil types currently managed by dairy farmers and secondly quantifying changes in soil properties in response to management. The Victorian Land Use Information System was interrogated to identify dairy land parcels and this data overlaid on government soil survey information to identify the dominant soil orders managed by dairy farmers in all three dairy regions. Of the approximately 590,000 hectares of dairy land identified across the state; Sodosols (33%), Chromosols (20%), Dermosols (16%), and Vertosols (11%) are the major soil orders managed by dairy farmers. However, management considerations will vary in each region as the dominant orders differ. Legacy data from research and extension activities undertaken by Victorian government staff between 1995 and 2010 were collated to understand spatial differences in dairy soil properties. All soil properties were significantly and positively skewed with higher median pH, EC and available K greater in northern Victorian soils. Preliminary analysis compared the 1995 to 2010 data with data from samples analysed by the government analytical laboratory between 1973 and 1980 to understand changes over 38 years. The older soil chemical data were also positively skewed. These earlier soil samples had lower median soil pH, Olsen P and available K. This paper discusses the challenges associated with analysis of legacy data to inform policy and land management priorities.
Introduction The Australian dairy industry like other agricultural sectors is responding to increasing world demand (FAO 2009), especially as the economies of countries with traditionally lower gross domestic product improve and diets include more animal protein. While dairy farms have intensified production (Dairy Australia 2010) a major competitive advantage of the industry remains the low cost of producing permanent pasture; which requires farmers to carefully manage soils to maximise pasture and fodder production. A number of farm management practices have the potential to degrade soil condition and therefore restrict the degree to which dairy soils will support ongoing plant growth and animal health. These factors include nutrient management, the regular movement of the cows and forage management on dairy farms. For example, the dairy industry has increased its use of fertiliser, feed and effluent nutrients over the past 20 years, with average use of N, P, K and S of 102, 32, 83, and 26 kg / ha, while up to 565, 108, 409, and 85 kg / ha were applied on some farms respectively (Gourley et al. 2012). Moving cows can deform soils due to the size and weight of the animal (Bilotta et al. 2007), while the nutrient content of their excreta will also affect soil chemical and biological properties (Williams and Haynes 1995, Aarons et al. 2009). Sustainable land management is required to ensure meeting agricultural objectives while preventing or reversing PAGE 148 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
environmental degradation. To ensure sustainable dairy production systems into the future requires a) quantification of the soil types typically managed by dairy farmers and b) understanding of soil condition change in the industry. This paper describes the approach taken to meet these objectives and discusses the implication of the results.
Material and Methods Victorian dairy soil types The Victorian Land Use Information System (VLUIS) database, selected as it represents spatial coverage of land use across Victoria and is based on the state cadastral parcel layer (Morse-McNabb 2011), was used to identify parcels of dairy land cover within the ‘pasture/grassland’ land cover classification. The dairy parcels were overlaid onto soil cadastral layers based on soil survey information collected and held within the state government databases for each catchment management authority region in Victoria. The dairy soils identified were grouped into their dominant soil orders and their proportional representation. Victorian dairy soil change Legacy soil data from research undertaken by Victorian government research scientists, published in peer-reviewed journals or in the grey literature, were collated to quantify dairy soil condition. This data consisted of 4814 records collected from 16 experimental research and fieldwork activities undertaken on 297 dairy farms around the state between 1995 and 2010 and was used to assess the condition of Victorian dairy soils over the 16 years (Table1). Additional data sourced from the Victorian Government soil analytical laboratory was analysed to assess soil chemical changes with time. This dairy data (3738 out of 21,000 records) came from samples collected by government extension staff between 1973 and 1980 from farmer paddocks. These data sets were selected as samples had been collected using recommended procedures and accredited analytical methodologies had been used. PAGE 149 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 1. Experiments from which legacy soil data were sourced.
No of farms Data ID Dates Regions Reference sampled A4N 2008 17 Gi, swV, NV Gourley et al. 2011 Greenwood et al. HDLN 2006-2007 24 swV 2008 Gourley et al. pers HBI 2004-2006 2 Gi comm. Neilsen et al. pers DSAT 2003-2004 30 Gi, swV, NV comm GDRP-P 2002 1 Gi Aarons et al. 2006 Melland et al. pers TAR 2004 4 Gi comm. PfDF 1995-2001 DEPI Research Farm Gi Gourley et al. 2001 CSOP 1998 36 Gi, swV, NV Aarons et al. 2001 Crawford et al. SQ 1997-1998 1 Gi 1998 BBF 2004 2 Gi Gourley et al. 2001 Armstrong et al. MRF 2005 1 Gi pers comm. MDF 2007-2010 1 Gi Lane pers comm. Gourley et al. pers WFNP 1999, 2002 3 Gi comm NWH 1999 4 Gi White 1998 Standish pers AS 2007-2008 28 swV comm. NMP_ 2005-2008 142 Gi Kelsall pers comm. FSV
Sample collection and analysis The soil samples consisted of a minimum of 30 soil cores (2.5 x 10 cm) collected either randomly throughout a paddock or by walking along a transect representative of the paddock. Dried (40oC) and ground (<2 mm) samples were analysed for pH (in water or 0.01 CaCl2) and electrical conductivity (EC) at a 1:5 soil:solution ratio. Bicarbonate extractable Olsen P, available K (either by bicarbonate extraction - i.e. Colwell, or hydrochloric acid - Skene K), and S (KCl40-S) were analysed. For full details of the methods used see Rayment and Lyons (2010). Statistical analysis The data were analysed to quantify chemical properties of soils under dairy management and to investigate spatial and temporal differences in properties. Summary statistical analysis was initially carried out using Genstat 14.1 (VSN International) to calculate the mean, median, minimum, maximum, standard deviation and coefficient of variation. Legacy data were also analysed to assess spatial (Gi – Gippsland, NV – northern Victoria, swV – south west Victoria) soil differences using REML analysis, and to compare Past (1973 to 1980) and Recent (1995 to 2010) soil properties. PAGE 150 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Results and Discussion The Victorian dairy industry is approximately evenly distributed in three regions; in northern Victoria (Northern Irrigation Region - NIR and the north east), in Gippsland (including the Maffra Irrigation District) and the south west (Figure 1, Table 2, DEPI 2013), where the soil types in each region will influence soil properties. Victorian dairy soil types A total of 590,320 hectares of land is managed by dairy farmers in Victoria as estimated by the VLUIS. The major soil orders, Sodosols (33%), Chromosols (20%), Dermosols (16%) and Vertosols (11%) amounted to 80% of Victorian soils, with the remaining eight orders each comprising less than five percent (data not presented). However, the proportion of major soil orders vary in each region (Figure 1). For example, Sodosols, dominant soils in the NIR (67%) and Gippsland (19%), only comprise 9% and less than 1% of the dominant soils managed by dairy farmers in south west Victoria and the north east respectively. Vertosols, cracking clays, are not a dominant soil type managed by dairy farmers in Gippsland despite constituting the second and third dominant soil type on dairy farms in the north and in south west Victoria respectively. Chromosols comprise a similar proportion of the dairy soils in the north and Gippsland, but are the major soil type in south west Victoria. The characteristics of soils, such as sodicity, acidity and texture contrast will affect their response to typical farm management practices as well as influence how these soils should best be managed to minimise their chemical and physical constraints to pasture and fodder production and grazing management.
Figure 1. Locations of the Victorian dairy industry (number of dairy cattle; DEPI 2013) and identified soil orders in the northern irrigation region (NIR), north east (NE), Gippsland (Gi), and south west Victoria (swV) regions (utilising the VLUIS and Victorian soil survey data). PAGE 151 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 2. Numbers of Victorian dairy farms (2011/2012), legacy research and extension data and analytical records for Gippsland (Gi), northern Victoria (NV) and south west Victoria (swV) regions
Research and extension Government Region Dairy farms records analytical lab records Gi 1575 2490 1382 NV 1551 392 840 swV 1454 1932 1518 Victorian dairy soil change The legacy data sourced from Victorian research and extension activities had been collected in all regions with the majority (52%) from dairy farms in Gippsland and only 8% from northern Victoria (Table 2). This data is an indication of the location of previous research and extension groups and programs as well as the relative priority placed on irrigation rather than nutrient management in northern Victoria. Soil chemical properties were significantly skewed with higher median soil pH, EC and available K in northern Victoria; most likely influenced by the dominant soil orders in that region (Table 3, Figure 2a). Significant (P<0.04) region and farm differences and non-significant paddock differences were observed for soil properties.
Table 3. Summary statistics of 4814 records of legacy chemical data from soil samples collected on dairy farms in all three regions in Victoria between 1995 and 2010.
Soil No. of Minimum Mean Median Maximum Std Dev CV (%) Skew property samples
pH (H2O) 4156 4.5 5.7 5.6 8.4 0.50 9 0.95 pH (CaCl2) 3576 3.8 5.0 5.0 8.1 0.54 11 1.13 EC (dS/m) 3615 0.02 0.18 0.14 7.31 0.218 120 17.995 Olsen P 4814 3 38 33 711 27.6 74 8.5 m(g/kg) Avail K 4814 39 319 270 11000 334.5 105 17.8 (mg/kg) SKCL-40 4397 2 19 15 810 23.6 122 16.2 (mg/kg)
The legacy data from the government laboratory records (1973 to 1980) were more uniformly distributed in line with the distribution of dairy farms around the state (Table 2). As for the 1995 to 2010 data, soil chemical properties of these samples were also positively and significantly skewed (data not presented). However, median soil pH, K, and Olsen P from samples collected between 1973 and 1980 were lower than that from 1995 to 2010 mostly likely due to the greater use of inputs by the industry currently (Figure 2b,c, Gourley et al. 2012). PAGE 152 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Conclusions By identifying the dominant soil orders in each dairy region suitable regional soil and farm management priorities can developed for the industry. Analysis of legacy data has been used for understanding soil change more broadly in Victoria (Marchant et al. 2014), and in this paper industry specific legacy data was investigated to understand spatial and temporal change. While legacy data can be used to inform priority farm management and policy decisions and to strategically target soil monitoring resources, data analysis and interpretation needs to be mindful of sample collection and analytical methods and fit for purpose sampling.
10.0 10.0 775 9.5 9.5 725 9.0 9.0 675 a b 625 c 8.5 8.5 575 8.0 8.0 525 7.5 7.5 475 7.0 425 7.0 375 6.5 6.5 325 6.0 6.0 275 ) - 1995 to 2010 225
pH water in (1:5) 5.5 5.5 -1 5.0 pH water in (1:5) 5.0 100 4.5 4.5 4.0 4.0 3.5 3.5 7.85 7.05 2.80 50 6.25 2.50 5.45 2.20 Olsen (mg P kg )
) 4.65 1.90 -1
-1 3.85 1.60 3.05 1.30 2.25 1.00 0 EC (dS m EC (dS m 150
0.00 0.00 125 11300 3000
10100) 2800 -1 8900 ) ) - 1973 to 1980 7700 -1 2600 100 6500 -1 75 5300 2400 4100 2200 2900 2000 50 600 Olsen (mg P kg Available K (mg kg 400 25 Available K (mg kg 200 0 0 0 Gi NV swV Gi NV swV Gi NV swV Regions (1995 to 2010) Regions (1973 to 1980) Region
Figure 2. Soil pH in water, EC (dS m-1), and available K (mg kg-1) of legacy soil data collected on Victorian dairy farms between a) 1995 and 2010, b) 1973 and 1980 and c) Olsen P (mg kg-1) data from those two time periods for the Gippsland (Gi), northern Victoria (NV) and south west Victoria (swV) regions. Horizontal red lines indicate agronomic optimum levels.
References Aarons SR, O’Connor C, Hosseini HM, and Gourley CJP (2009) Dung pads increase pasture production, soil nutrients and microbial biomass carbon in grazed dairy systems. Nutrient Cycling in Agroecosystems 84:81–92. Bilotta GS, Brazier RE, Haygarth PM (2007) The impacts of grazing animals on the quality of soils, vegetation, and surface waters in intensively managed grasslands. Advances in Agronomy 94: 237-280. Dairy Australia (2010) Australian Dairy Industry In Focus 2010. DEPI (2013) http://www.depi.vic.gov.au/agriculture-and-food/dairy/dairy-industry-profile; last accessed November 2013 FAO (2009) How to feed the world in 2050. Gourley CJP, Dougherty WJ, Weaver DM, Aarons SR, Awty IM, Gibson DM, Hannah MC, Smith AP, Peverill KI (2012) Farm-scale nitrogen, phosphorus, potassium and sulfur PAGE 153 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
balances and use efficiencies on Australian dairy farms. Animal Production Science 52, 929-944. Marchant BP, Crawford DM, Robinson NJ (2014) What can legacy datasets tell us about soil quality? Soil Change Matters Symposium, Bendigo, Australia. Morse-McNabb E (2011) The Victorian Land Use Information System (VLUIS): A new method for creating land use data for Victoria, Australia. Proceedings of the Surveying and Spatial Sciences Biennial Conference New Zealand Rayment GE, Lyons D (2010) Soil Chemical Methods - Australasia Williams P, Haynes RJ (1995). Effect of sheep, deer and cattle dung on herbage production and soil nutrient content. Grass and Forage Science 50:263–271. PAGE 154 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
WEDNESDAY 26 MARCH 2014 - TECHNICAL SESSION 6
Digital mapping of Soil Change Budiman Minasny1, Alex. B. McBratney2, Brendan Malone3, Uta Stockmann4
Department of Environmental Sciences, Faculty of Agriculture and Environment, The University of Sydney, Biomedical Building C81, 1 Central Avenue, ATP, Eveleigh, NSW, 2015, Australia, 1 [email protected] 2 [email protected] 3 [email protected] 4 [email protected]
Digital soil mapping is a term used to describe approaches that seek to map soil properties or classes with the aid of modern mathematical techniques and environmental variables that aim to capture various aspects of the character of soil in a landscape. Digital soil mapping approaches have been successfully used to represent the spatial distribution of soil properties. However, they do not only produce a static map, they can also be utilized to map soil change. There are several approaches that seek to map soil change without the luxury of monitoring sites. One of these is utilizing historical soil survey data. For example, nationwide soil test databases which were collected for soil fertility assessment, could be used to detect decadal spatiotemporal changes in soil carbon and other properties. Digital maps also can be used to predict the likely soil change. It can be done through a mechanistic simulation model or through a partial dynamic empirical model. In the first approach, the map of soil carbon at a given time can be fed into a dynamic-mechanistic simulation model. The output can project change; however, most soil carbon simulation models do not consider a spatial aspect. The scorpan approach can also be used to infer the likely changes in soil properties over time and termed ‘partially dynamic soil scenario maps’. Finally, we can use pedological knowledge to quantitatively relate soil properties to the environmental drivers that formed soil through quantitative mechanistic modelling of pedogenesis. PAGE 155 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Assessing changes in carbon contents of Scottish soils: lessons learnt Allan Lilly1, Steve J. Chapman1
1 James Hutton Institute, Craigiebuckler, Aberdeen, Scotland, AB15 8QH, [email protected] 1 James Hutton Institute, Craigiebuckler, Aberdeen, Scotland, AB15 8QH, steve.chapman@hutton. ac.uk
Abstract Between 1978 and 1988, the soils at 721 locations throughout Scotland were sampled at 10 km intervals. They were described, characterised and samples taken from each of the main horizons. Material not used for analysis was stored in the National Soils Archive. Between 2007 and 2009, 179 of these locations were revisited (20 km intervals) and fresh samples taken to identify changes in nutrient status, pH and, in particular, soil carbon concentrations and stocks over a 19-31year period (Chapman et al., 2013). The archived soil samples from the original sampling period were reanalysed alongside those from the recent sampling to determine carbon concentrations. NIR spectroscopy was used to estimate their bulk density so that carbon stocks could be calculated as this was not measured at the original sampling. Results showed no detectable change in soil carbon stocks to 1 m depth for the main, broad land use types in Scotland apart from a small but significant increase (P=0.035) in soils under woodland. There was an approximate 11.5 % difference in carbon concentration between the reanalysed, archived soil and values originally obtained which was attributed to differences in analytical methods. Losses of 3.3 g kg−1 carbon in cultivated soils were detected (P=0.013). However, a significant increases of 2.7 cm (P=0.019) in topsoil thickness was sufficient to compensate for these losses in arable soils such that there was no detectable change in carbon stocks. The work shows the value of soil archives and of measuring horizon thickness.
Introduction The Scottish government have set ambitious targets to reduce greenhouse gas (GHG) emission to 58% of 1990 levels by 2020 and to 20% of 1990 levels by 2050 (Climate Change {Scotland} Act 2009). It is estimated that Scotland’s soils hold around 3000 Mt C and clearly losses of carbon due to climate or land use change could impact on the Government’s ability to hit these target reductions as well as losing a valuable asset in terms of ensuring food security and adequate soil functionality. In order to establish if Scottish soils were indeed losing soil carbon, the Scottish Government commissioned a research project to evaluate changes in carbon stocks and other soil properties such as pH, concentration of exchangeable cations and anions. This project not only assessed changes but investigated how sampling methods could affect the overall conclusions.
Material and Methods Between 1978 and 1988 an inventory of soils in Scotland was undertaken by locating and describing the soil profile at each 5km intersection between the easting and northing grid lines. The major soil horizons in each profile were also sampled at each 10km grid PAGE 156 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
intersection giving a dataset of 721 sampled profiles. These samples were analysed to determine the concentration of the major exchangeable anions and cations, carbon, nitrogen and phosphorus as well as pH and particle size distribution. The results of these analyses were stored in the Scottish Soil Database and material not used for analysis was stored in the National Soils Archive. Between 2007 and 2009, 183 of these locations were revisited on a 20km sample frame, that is, 25% of the original sites. At each location, the site characteristics such as slope, rockiness and aspect were matched with the previous description and a soil profile similar to the original was identified and each of the major soil horizons were sampled. Composite samples 0-15cm were also taken over a 20x20 m area using a soil auger and single cores 0-15cm and 0-5cm were taken from the profile pit at each of the sites. The subsequent analyses included those done previously as well as bulk density, a key property required in determining carbon stocks. In order to minimise instrument bias, the original archived soil were located and analysed alongside the newly sampled soils in randomised but paired batches. Samples for all analyses were air-dried at 30oC and then sieved (<2 mm). Samples for carbon content were further dried at 50oC and those for loss on ignition dried at 105oC. At each stage of subsampling, the soil was riffled to minimise sampling bias. The original, archive samples were also air-dried at 30oC prior to analysis. Elemental carbon (%) for the original analysis was performed on a Hewlett-Packard CHN 185 analyzer but for the analysis of the new samples and for the reanalysis of archived samples, a Flash EA 1112 Series Elemental Analyser connected via a Conflo III to a DeltaPlus XPisotope ratio mass spectrometer (all Thermo Finnigan, Bremen, Germany) was used. The C contents were calculated from the area output of the mass spectrometer calibrated against a standard reference material which was analysed with every batch of ten samples (Chapman et al. 2013). The bulk density was determined for each major horizon in the soil profile for the 2007-9 soil samples, using, wherever possible, three replicates per horizon. Stainless steel cylinders, internal diameter 7.3 cm and height 5 cm, were pushed into the soil and then carefully removed, trimmed to remove excess soil and then extruded into a polythene sample bag, giving a soil volume rounded to 210 cm3. These samples were then dried at 105oC, sieved (<2 mm) to remove stones and any roots and weighed. The bulk density of the fine fraction was obtained after making a correction for the volume of stones and roots. Bulk density was not available for the 1978-88 soil samples so there was a need to estimate the bulk density for these soils. As changes in carbon were being assessed, and the C content of a soil influences the bulk density, we could not simply assume that bulk density values from the new samples could be applied to the original samples. Also, many pedotransfer functions used to predict bulk density use C content as a predictor leading to a double reliance on the carbon content figure for carbon stock estimation and subsequent error propagation. To overcome these issues, Near Infrared (NIR) spectroscopy was used to predict soil bulk density. The measured bulk densities from the 2007-9 samples were used for calibration and subsequent prediction of bulk density values for both datasets to avoid the potential bias of having measured values for one set and predicted values for the other. NIR spectra of the <2mm soil fraction (that is, the dried and sieved soil from the cores taken to determine bulk density and the archived soil) were recorded from 1100 – 2500 nm at 2 nm intervals on a FOSS NIRS Systems 5000 spectrophotometer (FOSS PAGE 157 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
NIRSystems, Silver Springs, MD, USA), using a transport module sampling attachment and a quarter cup sample holder. Reflectance mode spectra were collected using Infrasoft International ISIscan Software, Version 2.85.3 (FOSS Analytical AB, Hoganas, Sweden). Infrasoft InternationalWINISI III Software, Version 1.50 E (FOSS Tecator AB, Hoganas, Sweden), was used for the development of calibration equations to predict the unknown bulk densities for the 1978-88 samples as well as predicting bulk densities for the 2007-9 soil samples. A total of 663 samples had measured bulk densities which were correlated with the full spectrum of the NIR spectral data to develop an initial equation. A range of pre- processing methods, regression methods and different mathematical treatments were tried in order to obtain the best possible equation, which was validated by cross- validation. As the derived equation did not perform as well for highly organic samples (≥ 37% C) as for more mineral soil horizons, two separate regression equations were developed for soils with ≥ 37% C and < 37% C, respectively. The total profile carbon stock at each site and at each of the two sampling times was calculated as the sum of the component soil horizon carbon stocks. The stock was calculated to 100 cm unless the profile was restricted by the presence of rock. Soils with flooding could not be sampled to depth. However, if the profile was flooded on one occasion but could be sampled on the other, the C content was also calculated to 100 cm. Each horizon stock was the product of horizon depth, bulk density and carbon content with a correction made for stone content. As both carbon content and bulk density were measured on samples taken from the horizon midpoint, there was an assumption that they were representative of the whole horizon. In some cases, two (or three) samples were taken from down the horizon, particularly for thick horizons; this was accounted for in the calculation by dividing the horizon into two (or three) equal parts and applying the corresponding sample values. The comparison of the mean values for %C, bulk density, soil depth, stone content and C stock between soils sampled previously and the newer samples was done using a paired t-test using Genstat 13th edition (VSN International, Hemel Hempstead).
Results and Discussion The average time between samples was 25 years (19-31years) and we found no detectable change in soil carbon stocks to 1 m depth over this time period for the main, broad land uses types in Scotland (arable, improved grassland, semi-natural grassland, moorland & bog) apart from small but significant increases (P=0.05) in soils under woodland (Table 1). However, this significant change was only apparent when the data for deep (>1m) peat soils (Histosols) were removed from the analyses. PAGE 158 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 1. Change in mean carbon stocks in Scottish soils over an approximate 25 year period by broad land use type.
Change Vegetation type No. sites P value (C t ha-1) All 8.0 149# 0.175
Arable 0.0 16 0.998
Improved grassland -4.5 32 0.343
Semi-natural grassland 5.4 26 0.709
Woodland 23.5 21 0.035*
Moorland 6.2 41 0.686
Bog 34.7 13 0.286
*significant at P<0.05; #deep peats (Histosols) and soils derived from shelly sands removed from analysis. However, interpreting the results was often far from straight forward and occasionally apparently contradictory. Although there was no change in C stocks of soils under arable cultivation, the C concentration in these soils did decrease (13.0 to 12.3 g kg-1) however, there was a corresponding increase in topsoil thickness which meant that stocks remained the same. Had the soils only been sampled between 0-15 cm as is common in other UK sampling schemes, we would have reported a C loss. We also found an approximate 11.5 % difference in carbon concentration between the reanalysed, archived soil from the 1978-88 sampling campaign and values obtained previously. Initial concerns were that C had been lost during storage but a similar magnitude of loss was not found when we examined Loss on Ignition data suggesting that the apparent difference was due to differences in the analytical instrument used in the 1970s and 1980s compared to the modern instrument. Having access to the archived material to re-run the analyses was a crucial element of the work especially given the 11.5% difference between the instruments used. Had this material not been available and we simply compared the results of the analyses over the two time periods then we would have drawn an erroneous conclusion that the C stocks of Scottish soils had declined by a considerable amount. It was not possible, in this study, to accurately determine peat stocks for some peat soils where they were considerably greater than 1m thick as the depth to the mineral layer or rock hadn’t been recorded hence the reason these were removed from the analyses. As peat soils have a narrow range of carbon contents, the only way to determine changes in C stocks for these soils is to determine if the bog is growing or declining. While we were unable to do this for all peat bogs encountered at the sample locations, it is even more difficult to determine changes if the sample depth is 0-15cm. Indeed, a bog could be actively growing but as the bulk density of the upper layers is less than that of amorphous peat, the stock calculation could suggest a decline. PAGE 159 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Another complicating factor in the analyses was the confounding factor of arable/grass rotations. While it was relatively easy to determine if the site was under continuous arable or permanent grassland, sites that had a mixed arable/grassland system could have be allocated to either arable or permanent grassland, often depending on the age of the grass. If it is assumed that carbon accumulates under grassland and is lost under arable systems, the age of the grassland and its position in the rotation will influence the result. Also its position in the rotation relative to the repeat sample could influence whether a decline or increase in C content was detected. Better recording of land use history would help improve detection of change. Finally, from a practical point of view, while pairing sites gave increased statistical power to the analyses, since it removed variability due to site differences, it was sometimes difficult to relocate and replicate the exact soil conditions. While external conditions such as slope, aspect and rockiness could be reasonably well matched, the variability inherent in soils meant that it could be difficult matching the sequence, thickness and presence of horizons exactly and could lead to accusations of bias. Alternative approaches such as sampling on an offset grid (for example 1km east and north of the original, would have avoided these problems but would have meant many more sites would have needed to be visited in order to obtain a sufficient sample size.
References Chapman SJ, Bell JS, Campbell CD, Hudson G, Lilly A, Nolan, AJ, Robertson AHJ, Potts JM, Towers W (2013) Comparison of soil carbon stocks in Scottish soils between 1978 and 2009. European Journal of Soil Science 64, 455-465.
Soil carbon in dryland farming systems of Victoria, Australia
Doug Crawford1, Fiona Robertson2, Debra Partington2, Ivanah Oliver2, David Rees3, Roger Armstrong4, Colin Aumann5, Roger Perris4, Michelle Davey3
1 Farming Systems Research Division, Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Vic, 3821, Australia, [email protected] 2 Farming Systems Research Division, Department of Environment and Primary Industries, Private Bag 105, Hamilton, Vic, 3300, Australia. 3 Farming Systems Research Division, Department of Environment and Primary Industries, 32 Lincoln Square North, Carlton, Vic, 3053, Australia. 4 Farming Systems Research Division, Department of Environment and Primary Industries, 110 Natimuk Rd, Horsham, Vic, 3400, Australia. 5 Farming Systems Research Division, Department of Environment and Primary Industries, 255 Ferguson Rd, Tatura, Vic, 3616, Australia.
Abstract As part of the Australia-wide Soil Carbon Research Program, we surveyed soil C stocks under different dryland farming systems on the main soil types in the main climatic zones of southern and western Victoria. Site selection criteria demanded that the farming system existed for at least 10 years. Therefore, the survey was effectively a space-for- time survey of relative changes in soil carbon stocks under different farming systems. Soil samples were collected from 626 sites on farms. On each site, ten cores were PAGE 160 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
removed from randomly selected positions on a 25 m by 25 m grid. Each 30 cm deep core was sectioned into three 10 cm depth increments and bulked. Organic carbon was measured by dry combustion, and bulk density was estimated from core number, dimensions and weight. Sites were classified as continuous cropping, crop-pasture rotation, permanent pasture grazed by sheep and/or beef cattle, permanent pasture grazed by dairy cattle, and by soil type using the Australian Soil Classification, with about 25 representatives for each farming system by soil type combination. Data on climate, paddock history and site characteristics were also collected. Random forest and multiple regression analyses were used to identify the impact of different farming systems and their management practises. Differences in soil organic C stocks between classes of sites were best described by differences in climate parameters such as annual rainfall and vapour pressure deficit. The relative unimportance of farming systems and management practices was noteworthy and will be discussed.
Using legacy sites to assess changes in soil organic carbon D M Crawford1, N Robinson2
1 Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Victoria, 3821, [email protected] 2 Department of Environment and Primary Industries, Taylor Street, Epsom, Victoria, 3551 Abstract ›› Changes in soil organic carbon affect the eco-system services provided by soil. The use of legacy sites to assess changes was explored. ›› Over the last three years, sites in western Victoria used by the National Soil Fertility Project (NFSP) from 1968 to 1972, were surveyed to assess the average difference in soil organic carbon content between these two periods. Local features and the records from the NSFP were used to locate the sites and apply the same methods as when they were first tested, e.g. the Walkley and Black method of measuring soil organic carbon. The sites are mainly on commercial farms. They were typically 30 m by 50 m for a pasture field trial or 60 m by 100 m for a wheat field trial. Each site was sampled by taking 8 or 9 cores of 1 m depth, cutting them into 10 cm sections, and bulking the segments to form composite samples representing each 10 cm depth increment. ›› The difficulties and rewards in using legacy sites and historical data will be discussed in terms of the apparent changes or lack thereof, in soil organic carbon. PAGE 161 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
THURSDAY 27 MARCH 2014 – SOIL CHANGE MATTERS WORKSHOP PLENARY Sampling requirements for soil monitoring: or on choosing the right length for a piece of string. Murray Lark, Kate Knights1
1 British Geological Survey, United Kingdom.
Abstract At some stage in the planning of a soil monitoring scheme decisions have to be made as to how much effort will be invested in field work. This decision affects both the costs of monitoring and the precision of the results. This is a key stage in the establishment of monitoring at which the soil scientist, statisticians and policy or management sponsor must be able to communicate effectively and come to agreement. In this presentation we shall discuss some of the challenges entailed in making decisions about sampling effort and protocols, using statistical results and other information about the variables and processes of interest. In particular we shall focus on the following questions: How can we make informed decisions about sampling to estimate change when no pre- existing multitemporal data are available? How can information on short-range variability of soil properties be used to plan efficient site-level protocols for soil sampling? What environmental and other factors influence the rate of survey across a region and so the best partition of effort between reducing variability of measurements at a site and increasing the total coverage of the survey? Here we will report recent results from a logistical study of a regional geochemical survey in Ireland. How can the relationship between the uncertainty of estimates made from a sample and the sample effort be effectively communicated to non-specialists?
Detection and measurement of change : sources of error and spatial uncertainty K.K. Benke1 and N.J. Robinson2
1 Department of Environment and Primary Industries – Parkville Centre, PO Box 4166, Parkville, Victoria, 3052, Australia. Email: [email protected] 2 University of Ballarat, University Drive, Mt Helen Victoria 3350, Australia.
Abstract Sources of error in predictive models of soil properties and digital soil mapping are often due to spatial and temporal uncertainty. Uncertainty in geometry, position, and polygon attributes are amenable to statistical descriptions. Integration of error sources is reviewed in the context of the SCORPAN model (McBratney et al. (2003), including covariate error, model error, laboratory analytical error, and positional error. A dynamic approach to modelling soil evolution and temporal uncertainty can be addressed by the stochastic simulation of the STEP-AWBH framework proposed by Grunwald et al. (2011). PAGE 162 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Introduction Digital soil maps provide information for policy makers, managers, analysts and scientists. In the case of modelling and simulation, uncertainty relates to both point source data and spatial data, and error propagation in predictive computer models. Once a large grid of sampled data is used to represent a landscape there are also additional issues relating to spatial uncertainty based on the fact that digital mapping refers to essentially a signal input of a 2-D array of point sources that has been subject to geometrical transformations. Digital soil mapping, using environmental correlation, are subject to various errors and uncertainties including error introduced when moving from point source data to spatial data and the effects of temporal correlation. Other potential sources of error (not discussed further) include implementation error (numerical approximations), incorrect framing or applied context and software implementation problems.
Uncertainty in object position and attributes A probabilistic framework can be developed for environmental variables represented as objects in the landscape subject to positional uncertainty and attribute uncertainty (Heuvelink et al., 2007). Examples for positional uncertainty include lakes, boundaries, roads, buildings. Examples of attribute uncertainty include soil types and chemical concentrations, such as soil organic carbon content. The framework proposed by Heuvelink et al. (2007) can be summarised as follows. Positional Uncertainty is described by a probability density function (PDF) and relates to objects comprising multiple points with structure that may or may not change under uncertainty (e.g. rigid objects and deformable objects). Positional uncertainty of a point object leads to a shift in its 4-D status (x,y,z,t) subject to enumeration by the PDF. A rigid body is subject to geometric transformations, such as translation and rotation about an axis, with specific enumerations also subject to a PDF. Deformable objects may be altered by positional uncertainty due to independence of the primitive points. Attribute Uncertainty may also be subject to a description by a PDF and covers (a) the nature of the measurement scale used, and (b) time-space variation. Heuvelink et al. (2007) list four classes for the measurement scale, i.e. ›› continuous numerical scale (e.g. chemical concentration in soil) ›› discrete numerical scale (number of plant species) ›› categorical scale (e.g. soil type) ›› descriptive text (e.g. history of soil type) In a similar manner, space-time variability is subject to four classes, with attributes that are ›› constant in space and time (e.g. universal gas constant) ›› constant in space but vary in time (e.g. national interest rate) ›› constant in time but vary in space (e.g. some geographic/geological features) ›› vary in both time and space (e.g. temperature) Heuvelink et al. (2007) links spatial uncertainty to variability using PDFs – which can be used to represent uncertainty in soil type boundaries and polygon representations. Benke and Pelizaro (2010) showed how uncertainty in maps of regional classifications and polygons can be modelled by a probabilistic framework based on Monte Carlo PAGE 163 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
simulation and visualisation. In the latter case, the numerical attribute of each polygon was associated with a metric of uncertainty, the standard deviation. Moreover, it can be shown that these two approaches, e.g. multiple enumerations of each object or polygon (by simulation), can be overlaid to produce fuzzy edges or boundaries that visually reflect the degree of uncertainty in these boundaries and can be linked to the uncertainty metric, such as standard deviation (see examples in Heuvelink et al. 2007, Benke et al., 2011). In the case of the standard deviation at a fuzzy edge, the spatial uncertainty in the boundary profile can be represented visually by the cumulative distribution function (CDF) as depicted in Figure 1.
Uncertainty in geometry There are a range of possible PDFs for modelling variability or uncertainty in either position or attributes of objects or areas. For example, the normal distribution (continuous variable) or the binomial distribution (discrete variable) may be used. The forms of the distribution may be estimated by a sample or calibration data, or by expert opinion (modelled by a probability distribution). If a parametric PDF is used for an attribute, there may be spatial-temporal dependence. There may also be correlation or dependencies between variables. Moreover, positional uncertainties may also be statistically dependent in space and time and between coordinate dimensions. More importantly, if there is statistical independence in space and time, the joint PDF from the model output is the product of the marginal PDFs -- and can be produced by estimating the separate marginal PDFs (Aerts et al., 2003; Heuvelink et al., 2007). If dependencies exist between variables, these must be determined together with the marginal PDFs. Heuvelink et al. (2007) note that if dependency exists, the joint PDF is often assumed to be the multivariate normal distribution where the covariance matrix is used for correlated variables. For positional and attribute uncertainty, under some conditions, the covariance depends on the distance between locations and is computed from the variogram (Heuvelink et al., 2007).
For an extended region or a polygon in a digital map, spatial uncertainty in the boundary can be represented as a fuzzy edge (Figure 1). The uncertainty metric is standard deviation and the edge profile or cross-section may be represented visually by the cumulative distribution function (CDF). A visualisation strategy may be a useful addition to future digital soil maps . Uncertainty in the the attribute, A, with mean, (from μA measurements or model predictions) is represented by the metric for the region, or σA a polygon, as shown in Figure 1 (see Benke et al., 2011). For the purpose of visualisation, the uncertainty metric for A may also be represented by a pseudo-colour encoding scheme in addition to an assigned numerical value for (e.g. a spectrum from blue to σA red, representing low to high uncertainty, respectively). Uncertainty at the fuzzy edge (boundary) enclosing attribute A is represented by , which can also be derived from σB sampling measurements or multiple model realisations using Monte Carlo simulation. Positional uncertainty in the homogeneous region or polygon would be represented by , representing the variability in the centre-of-mass (C.M.), assuming no shape σC deformation, which again would be elicited from multiple realisations from Monte Carlo simulation. Finally, rotational uncertainty about an angle would be represented by . θ σD PAGE 164 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
ó A
C.M.
ó C ó B ó D
Figure 1. Standard deviations for uncertainties in a soil attribute, A, its position, C, its boundary and rotation, D.
Integration of error sources Using the example of the SCORPAN model (McBratney et al., 2003), there are four main steps involved in digital soil mapping with uncertainty (Minasny et al., 2010). First, data input for the region of interest requires production of the digital map, using covariates, which may include terrain attributes, multispectral satellite imagery, land use data, geological information and possibly legacy soil maps. Second, estimates of soil properties, including uncertainties, are produced from relationships between point soil measurements and spatially covered covariates, i.e.
S = f (s,c,o, r, p, a, n) + e (1) where S is the soil property, attribute or class of interest, f is the model incorporating covariates s(other soil properties), c(climate properties), o(organisms), r(topography), p(parent material), a(age or time factor), n(spatial position absolute and relative), and ε is the error. Third, “using spatially inferred soil properties” to predict other soil functions, such as soil water content, carbon density, and phosphorus (see Minasny et al., 2010). Thus, the prediction uncertainty of the SCORPAN model combines uncertainties or errors in input data, spatial inferences and soil properties and functions. The fourth step includes completion of a digital soil assessment for use by policy makers and land use managers, including evaluation of soil functions such as biomass production and buffering capabilities (Carré et al., 2007). A framework for an integration of error sources in digital soil mapping in the SCORPAN approach has been suggested recently by Nelson et al. (2011). Here the implemented approach combined a geostatistical model and Monte Carlo simulation to produce many trials in order to estimate underlying errors. A Linear Mixed Model (LMM) was used to produce a digital soil map of clay content and prediction error. Nelson et al. (2011) considered four major sources of error including, ›› covariate error ›› model error ›› analytical error (of soil properties) ›› positional error PAGE 165 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
In the first category of covariate error, environmental covariates such as mean annual rainfall are aggregated into a common grid in the model. The principal source of error relates to measurement error, except where sensor data is introduced, or low sample rates are involved, or both, leading to the further inclusion of interpolation error. In the second category of model error, sources of error include incorrect model, parameter error, redundant covariates, and interpolation error. All of these can inflate the error variance in model prediction, given incorrect assumptions on statistical parameters such as stationarity in first and second order moments. For example, in the LMM, errors in fixed effects coefficients are assumed normal. Variation not explained by the model is quantified by the nugget and sill variances in geostatistics. Effectively digital soil mapping is a process distinguished by interpolation of low density soil observations into a dense grid of prediction locations. Model error is subsequently quantified by the error variance for these predictions. This is often executed by the process of bootstrapping, i.e. a model fitting exercise involving multiple realisations of the dataset, which may be obtained from probabilistic simulations of the original whilst retaining its statistical properties (such as first and second order moments). In the third category of analytic error, the primary consideration is the quantitative error in measurement of soil properties. In the case of soil properties, such as organic carbon content, laboratory methods are expensive but more accurate with lower dispersion than remote sensing methods. In the fourth category of positional error, samples taken near class boundaries produced greater errors than samples from class interiors (Figures 1). Historical samples, referred to as legacy data, are associated with larger errors than current data from accurate GPS technology (Grimm and Behrens, 2009; Carré et al., 2007). In the case of model-based geostatistics, LMM can be used with parameters estimated using the residual maximum likelihood (REML), which was recommended over the standard regression-kriging approach (Lark and Cullis 2004), with interpolation by the empirical best linear unbiased prediction (E-BLUP). Regression-kriging produces estimates of parameters and spatial correlation separately, which may lead to bias and errors in variable selection (Nelson et al., 2011). In the case study conducted by Nelson et al. (2011), the general form of the LMM was fitted to clay data using REML for estimation. Error sources were ranked by variance, i.e. contribution to mean square error (MSE) for four data quality scenarios (Table 1). Model error (parameter error, interpolation error, etc) accounted for two thirds of the total variance in prediction for all four scenarios. Position error accounted for less than 1% of variance and is related to the grid size of interpolation relative to the covariates. Sources of error are often analytic and covariate in nature, whilst the least error occurs with positional uncertainty and measurement error (Heuvelink and Brown, 2007; Nelson et al., 2011). The advantages of the so-called error budget approach is that it resolves the total error in the digital map of clay content into separate proportional contributions from different error sources. Further elaboration on these categories is provided by Refsgaard et al. (2007) and Benke et al. (2010, 2011). PAGE 166 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Table 1. Comparison of error sources for error-budget model for data quality of clay (indicative data from Nelson et al., 2011). Table shows proportion of variance contribution to MSE of predictions.
SCENARIO Error Source Good Spectroscopic Legacy Poor
Model 69% 72% 69% 72% Analytic < 0.5% 3.5% < 0.5% 3.55% Positional <-0.5% <-0.5% <-0.5% < -0.5% Covariate 1.7% 3.0% 1.4% 2.7%
The SCORPAN and error budget approaches are essentially static models that can be implemented at different times. A framework incorporating time dependence explicitly is the STEP-AWBH conceptual model for soil evolution, as proposed by Grunwald et al. (2011). It accounts for anthropogenic and natural forcings which determine and modulate soils and space-time interactions. The model addresses temporal factors correlating with soil change, including land use change and climate change in temperature and precipitation, and can be implemented by stochastic simulation methods or deterministic approaches, such as regression trees. Monte Carlo simulation can be used to address uncertainty.
References Aerts JCJH, Heuvelink, GBM., Goodchild, MF (2003) Accounting for Spatial Uncertainty in Optimization with Spatial Decision Support Systems. Transactions in GIS 7,211-230. Benke KK, Pelizaro C (2010) A spatial-statistical approach to the visualisation of uncertainty in land suitability analysis. Journal of Spatial Science 55,257-272. Benke KK, Pettit CJ, Lowell, KE (2011) Visualisation of spatial uncertainty in hydrological modelling. Journal of Spatial Science 56,73-88. Bishop TFA, McBratney AB, Whelan BM (2001) Measuring the quality of digital soil maps using information criteria. Geoderma 103,95-111. Carré F, McBratney AB, Mayr T, Montanarella L (2007) Digital Soil assessments: beyond DSM. Geoderma 142, 69-79 Grimm F, Behrens T (2009) Uncertainty analysis of sample locations within digital soil mapping approaches. Geoderma 155, 154-163. Grunwald, S, Thompson, JA, Boettinger, JL (2011) Digital soil mapping and modelling at continental scales – finding solutions for global issues. Soil Sci. Soc. Am. J. 75,1201-1213 Heuvelink GBM, Brown JD, van Loon EE (2007) A probabilistic framework for representing and simulating uncertain environmental variables. International J. Geog. Inf. Sci. 21, 497-513. Lark RM, Cullis BR (2004) Model-based analysis using REML for inference on systematically sampled data on soil. Eur. J. Soil Sci. 55, 799-813. PAGE 167 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
McBratney AB, Santos MIM, Minasny B (2003) On digital soil mapping. Geoderma 11,3-52. Minasny B, McBratney A, Malone B, Sulaeman Y (2010) Digital mapping of soil carbon. 19th World Congress of Soil Science, Soil Solutions for a Changing World 1 – 6 August 2010, Brisbane, Australia. Published on DVD. Nelson MA, Bishop TFA, Odeh IOA, Triantafilis J (2011) An error budget for deferent sources of error in digital soil mapping. Eur. J. of Soil Science, 62, 417-430. Refsgaard JC, van der Sluijs JP, Højberg AL, Vanrolleghem PA, (2007) Uncertainty in the environmental modelling process – A framework and guidance 22, 1543-1556.
A soil test database for French agricultural soils. State of progress, advantages and limits for monitoring soil changes Dominique Arrouays1, Nicolas P.A. Saby1, Blandine lemercier2, Christian Walter2
1 INRA, Infosol, US 1106, 45075 Orléans cedex 2, France, [email protected] 2 UMR SAS –INRA-Agrocampus-ouest, Rennes, France, [email protected]
Abstract Political awareness that soil is threatened by increasing pressures has been rising for several years. Indeed, the demand for soil information is increasing continuously. Although rates of soil degradation are often slow and only detectable over long timescales, they are often irreversible. Therefore, monitoring soil quality is essential in order to detect adverse changes in their status at an early stage. Analysing the results from existing soil measurement exercises, such as operational soil testing by farmers or fertiliser suppliers is one potential option for detecting large temporal trends in soil characteristics. However, the conclusions drawn using these kinds of data may be subject to several sources of bias that are inherent in a non-controlled sampling strategy. In this paper we review the state of progress of a database on soil tests since the 1990’s. We show some examples of spatial and temporal trend detection and we stress the advantages and the limits of such a monitoring tool. PAGE 168 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Accessing Queensland’s soil information – an open data revolution! Kelly Bryant, Lauren O’Brien, Daniel Brough1
1 Queensland Department of Science, Information Technology, Innovation and the Arts, Dutton Park Qld.
Abstract The Queensland government is the custodian of soil and land resource information with an estimated value of $75 million. The Soil and Land Information (SALI) system houses this data from over 600 distinct studies with some 96,000 soil observations dating back to the 1940s. This data is now not only used by government but by universities, councils, landowners, consultants and schools. Providing this information to the public in an easy and accessible way, with a focus towards online delivery is crucial. Previous issues with distribution of online soils information in Queensland have stemmed not only from limits to technology but also, changing departmental structures and multiple websites. The department which manages soils information in Queensland has undergone nine name changes in the last 12 years due to Machinery of Government (MoG) restructures. This constantly changing web presence and branding is as confusing for people sourcing soils information as it is for those providing it. The Queensland government has now moved to a whole of government online environment. This is a single website with no reference to the convoluted structures within government or department names. The aim is to prevent impacts from future MoG changes on the provision of data and information to the public. Queensland government soils now has a single dedicated website (qld.gov.au/environment/land/soil) which has allowed us to start to build a repository for soils information and is a single portal for people to access soils data. It has been demonstrated that this consistent approach to websites improves trust and confidence of users (Queensland government, 2011) and from this, confidence in using Queensland soils information and data and ultimately better land management decisions.
Historical issues with accessing soils information in Queensland Where do I get soils information in Queensland? This question would have been an easy answer up until the early 2000s. The Department of Primary Industries (DPI) and the Department of Natural Resources (DNR) were around collectively for 38 years (33 years for DPI and 5 years for DNR). These departments became common brands for the place to go to get soil and land information in Queensland. In the past 12 years the Queensland state department which delivers soil and land information has had seven names (Figure 1). This makes it difficult to become familiar with who to contact for information. Currently to access soils information the contact is through the Department of Science, Information Technology, Innovation and the Arts. To most people this would not be a first port of call.
Figure 1. Queensland government’s departmental name changes 1963 to 2012 PAGE 169 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Whole of government sites and stable URLs for delivering information The Queensland government has moved to a whole of government or “one stop shop” website structure. All Queensland government websites will have the same look and feel and the entry to the site is through a single URL www.qld.gov.au. There were two main aims to achieve by moving to this structure: To be client focused—ensuring that information online is provided based on what the public want. This information comes from focus group research to determine the type of information that people want to access on a state government website site relating to the environment. Previously information provided was governed by each department. To be content focused—focusing on the information provided rather than the department that provides it. Information may be produced by several departments i.e. soils information is currently provided by two different departments, but this is not apparent on the website. This ensures that future restructures of departments do not impact on information delivered through state government websites.
Who uses information and what do we provide? Now there is a stable environment for us to build a repository of soils information the focus has shifted to determining who are our audience is, what information they need and how can we provide this. The users vary considerably - from people wanting to know what soil is good for growing veggies in their backyard to statewide energy providers planning major infrastructure developments. Over the past three years we have been monitoring the requests that come via phone and email to capture the type of information people are requesting. The majority of requests are for soils mapping and data. In 2012 the Queensland government introduced an open data program (data.qld.gov.au) with the aim of making all data collected by the Queensland government available to the public. Since December 2012 the majority of Queensland soils mapping has been available through this download service – now over 120 soil surveys are available. Currently about nine data sets are downloaded per day through this service, compared to previous requests (either by phone or email) of about nine data sets per month. PAGE 170 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Figure 2. Proportion of soil data downloaded by user groups
Future ways to target Generating statistics on users and information accessed is the best way to target the areas that we need to improve. We have tracked the broad groups of people who use this data to date (Figure 2). The use of more advanced services (such as Google Analytics) in the future will provide a greater level of detail on user groups and information accessed. These types of statistics are essential to enhance information delivery services as they improve our understanding and targeting of user groups.
References Queensland Government (2011), Consistent User Experience 3.0, viewed 25/10/2013, http://www.qld.gov.au/web/cue/standard/ PAGE 171 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Soil Archives: supporting Research into Soil Changes Linda Karssies, Peter Wilson 1
1 CSIRO Land and Water, GPO Box 1666, ACT, 2601, [email protected]
Abstract Soil archives provide valuable support to soil research by making soil specimens and associated data available, reducing the need for labour intensive and expensive fieldwork.
International Soil Archives The usefulness of soil archives has been well demonstrated. Work based on the Rothamsted collection in the United Kingdom, provided evidence of steady rates of increase in concentrations of dioxins during the last century (Rothamsted Research, 2006). The sample archive of the equally well known Hubbard Brook Experimental Forest in the U.S. led to the discovery of the link between the use of fossil fuels and increased acidification of rain and snow in North America (Likens and Bormann, 1995). Soil archives are also the Rosetta stone of terrestrial carbon cycling because atmospheric CO2 is spiked by radiocarbon from aboveground nuclear weapons testing and offers a tracer for carbon in soil.
Anstralia’s National Soil Archive The CSIRO National Soil Archive was established in 2003 to consolidate several smaller CSIRO collections (e.g. Merry, 1984). The soil specimens from these original collections had been collected for a wide range of research projects and differ in the amount of soil sampled, sampling depth and data collected in the field and in the chemistry laboratories. Since 2003, many state agencies archives and individual projects have contributed soil specimens and data, helping to build a truly National Soil Archive facility. Currently the collection holds 70,000 soil specimens from 9,000 profiles (see Figure 1). The National Soil Archive has protocols both for sample submission and sample use. The National Soil Archive’s mission is to provide facilities and protocols for conserving the long-term scientific value of soil specimens and associated soil data. Archived specimens and their data al also made available for public research both now and in the future, when new analytical techniques may be brought to bear on the specimens. Specimens submitted are required to meet a number of criteria such as having adequate location data and other documentation and be non-toxic and in a dry condition. Sample submissions vary in size from several hundred to several thousand specimens and may originate from national monitoring programs or local studies, soil science research or ecological studies and may be recently collected or several decades old. Sample use requests are even more varied and range from a single, specific sample to hundreds of samples that are reanalysed with modern laboratory methods. Co-located spectroscopy research by CSIRO has already re-analysed thirty-thousand specimens using NIR and several thousand using MIR (Viscarra Rossel et al. 2006). Users of the CSIRO National Soil Archive sometimes only request the soil data, in other cases only the soil specimens. Usually the two go hand in hand. The National Soil Archive aims to balance access to the soil specimens to support current research with the need for preservation of specimens for future generations. PAGE 172 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
To archive soil specimens, air-dried soil material is transferred into standard long-life containers with bar coded labels. Associated soil site description, morphology and chemistry data is transcribed and loaded into a standardised relational database (NatSoil) (Jacquier et al. 2012). The site locations and data on soil morphology and chemistry can be accessed via the Australian Soil Resource Information System (ASRIS) (McKenzie et al. 2012) and through the SoilMapp for iPad app.
Figure 1. Soil profile locations of archived soil specimens in the National Soil Archive
Who contributes to the Collection? The National collection started in 2003 by combining all roughly 24,000 specimens from the former CSIRO Divisions of Soils campuses. These specimens are very rich in data, since they were collected specifically for soil research. This part of the collection reflects the changes in research focus over the last 50 years. Older specimens stored by the CSIRO National Soil Archive (from the 1920’s and the 1930’s) do not have location data of the currently required accuracy and are therefore not archivable, that part of the collection can still be used, if one is familiar with the publications. These specimens were collected prior to widespread application of pesticides and herbicides and before the start of nuclear testing in 1945, so provide an important baseline of soil condition. The very old specimens and other yet unarchived specimens are referred to as ‘stored’ specimens (Figure 1). Large projects such as the Terrestrial Ecosystem Research Network (TERN) which connects ecosystem scientists and enables them to collect, contribute, store, share and integrate data across disciplines have contributed their AusPlots specimens for safe storage and future use (White et al 2012). Other national projects such as the Soil Carbon Research Program (SCaRP), who provide a nationally consistent assessment of soil carbon condition across the major land-use/soil type combinations used for agricultural production, have also contributed specimens (Sanderman et al. 2011). Early discussions between the Soil Archive and these large projects, which may involve tens of thousands of soil specimens, have led to streamlining of the archiving process. The National Soil Archive holds the data and soil samples that have been captured since 2004 for the Soil Condition Evaluation and Monitoring (SCEAM) Project which provides benchmark data for a range of soil condition indicators. For this study all major land-uses PAGE 173 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
in Tasmania - intensive cropping, grazing, native and plantation forestry – are targeted and re-sampling occurs at five year intervals (Cotching and Kidd 2010). Some examples of retrospective archiving of specimens from the 1990’s are provided by the submission of 1200 well-documented soil specimens from the Desert Uplands Strategic Land Resource Assessment (DUSLARA) in north and central western Queensland and a set of over 4,000 soil specimens with complete data from the Western Australian Department of Environment and Conservation whose dozens of ecological studies focus on remote, non-agricultural, remote areas.
How is the Collection used? The National Soil Archive supports many soil research projects, for example – A rapid assessment of the distribution of soil with toxic levels of boron, which was completed by analysing existing specimens at half the estimated expense of new field sampling Thousands of archived specimens have been scanned and the MIR and vis -NIR spectra obtained. Using existing limited laboratory data and these spectra, models have been constructed and important soil properties predicted The Australian Federal Police have used archived soil specimens to examine the potential of new forensic methods A study on acid sulfate soils re-analysed archived specimens from the 1920s The Victorian DEPI has re-analysed specimens from the 1970s National Soil Fertility program to study changes in soil carbon content since that time
An opportunistic Archive The fact that the specimens in the National Soil Archive derive from a host of project, agencies and researchers over an 80 year period adds to the breadth of the collection, but it can also cause some difficulties for users. In terms of data, older collections may be geo-referenced post-collection, which may introduce location errors. The data associated with specimens are often highly variable as far as the level of detail provided, for example, the precision of the geographic coordinates of the collection location, morphology description, collection method and analytical methods used. During archiving, legacy soil data needs to be transcribed from paper copies of internal reports, ASCII text, Fortran based non-relational flat file database files or spreadsheets to fit the dedicated relational database. This involves a substantial effort which requires professional staff because the re-interpretation of descriptive codes, recalculation of measurement units, methods and nomenclature requires familiarity with current and previous classification systems. For example, when Victorian researchers used archived soil data from the 1970’s, some serious implications for using legacy data when dealing with Soil Organic Carbon came to light. A shift in Organic Carbon results appeared due to lab-based modifications over a 40 year period of the Walkley-Black method used to determine OC (Robinson, pers. comm.). The researchers decided to re-analyse archived specimens to develop a correction factor, something that has been done before by Skjemsted et al. (2000) in the Carbon conversion factors for historical data project. PAGE 174 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Partners The maintenance of soil archives is not trivial and requires continued financial support and institutional commitment (Boone et al., 1999). The National Soil Archive is one of the key deliverables of the Australian Collaborative Land Evaluation Program (ACLEP). ACLEP is a partnership between CSIRO, the Australian Government’s Department of Agriculture and the state and territory agencies responsible for land resource assessment.
Conclusion The collection of soil specimens in the CSIRO National Soil Archive is an irreplaceable and valuable resource for current future research due to the number of specimens, the quality of the information available for them and the broad range of sampling locations. Archived soil specimens can support problem-solving soil research by allowing calibration of new analytical methods, enabling the development of baseline national soil property maps and by providing soil material for carbon cycle research. Care and commitment are needed to establish a world-quality soil archive and the long- term value ought never to be compromised by short-term gains. Decisions of policy makers of the future must rest on the best possible data (Leigh et al. 1994). To secure the future of the National Soil Archive it is essential that its importance is recognized by the whole scientific community and, importantly, by those who fund the work. This requires foresight, patience and an awareness of the great potential value of the information that can emerge from archived samples. It also involves collections owners taking responsibility for their collections and being held accountable for their long-term care. Besides being an unrivalled resource for addressing many established issues, the National Soil Archive has the potential to provide answers to agricultural and environmental problems not yet recognised.
References Boone RD, Grigal DF, Sollins O, Ahrens RJ, Armstrong DE (1999). Soil sampling, preparation, archiving and quality control. In Standard soil methods for long-term ecological research. In ‘Long-term ecological research network series No. 2’ (Eds GP Robertson, DC Coleman, CS Bledsoe, P Sollins) pp. 3-28 (Oxford University Press: New York). Cotching W, Kidd D (2010) Evaluation of surface soil condition in Tasmania, Australia. 19th World Congress of Soil Science, Soil Solutions for a Changing World 1-6 August 2010, Brisbane, Australia. Published on DVD. Jacquier DW, Wilson P, Griffin T and Brough D (2012) Soil Information Transfer and Evaluation System (SITES) – Database design and exchange protocols Version 2.0. CSIRO Land and Water report, Canberra. Karssies L, Jacquier D, Wilson P, Ringrose-Voase, A (2011) CSIRO National Soil Archive Manual. CSIRO Land and Water report, Canberra. Leigh RA, Prew RD, Johnston AE (1994). The management of long-term agricultural field experiments: procedures and policies evolved from the Rothamsted classical experiments. In’ Long-term experiments in agricultural and ecological sciences’. (Eds RA Leigh, AE Johnston) pp. 253-268 (CAB International: Wallingford). Likens GE, Bormann FH (1995). ‘Biogeochemistry of a Forested Ecosystem’. Second Edition, New York: Springer-Verlag. PAGE 175 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
McKenzie NJ, Jacquier DW, Maschmedt DJ, Griffin EA, Brough DM (2012) The Australian Soil Resource Information System (ASRIS) Technical Specifications. Revised Version 1.6, June 2012. The Australian Collaborative Land Evaluation Program. CSIRO Land and Water report, Canberra. Merry, RH (1984) A summary listing of the archival soils collection of the Division of Soils. CSIRO Division of Soil Divisional Report 80, Adelaide. Rothamsted Research (2006) Rothamsted Research: Guide to the Classical and other Long-term Experiments, Datasets and Sample Archive. Sanderman J, Baldock J, Hawke B, Macdonald L, Massis-Puccini A., Szarvas S (2011) National Soil Carbon Research Programme: Field and Laboratory Methodologies. CSIRO Land and Water Report, Adelaide. Skjemstad, JO, Spouncer, LR and Beech, TA (2000). Carbon conversion factors for historical soil carbon data. National Carbon Accounting System Technical Report No. 15, Australian Greenhouse Office, Canberra. Viscarra Rossel RA, Walvoort DJJ, McBratney AB, Janik LJ, Skjemstad JO (2006) Visible, near infrared, mid infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties. Geoderma 1131, 59-75. White A, Sparrow B, Leitch E, Foulkes J, Flitton R, Caddy-Retalic S (2012) AusPlots Rangelands Survey Protocols Manual Version 1.2.9. University of Adelaide Press. Adelaide.
The Rise of Information Technologies: a Changing Landscape for Soil Science Pierre Roudier1, Alistair Ritchie2, Carolyn Hedley1, David Medyckyj-Scott1
1 Landcare Research - Manaaki Whenua, Private Bag 11052, Manawatu Mail Centre, Palmerston North 4442, New Zealand 2 Landcare Research - Manaaki Whenua, PO Box 69040, Lincoln 7640, New Zealand
Abstract The explosion of information technologies has been the most important development in science over the last 15 years. We live in times where more data are being generated than ever before – a phenomenon described as the “data deluge”. It has had a massive impact on science, and whole disciplines have moved from data-poor to data-rich. Soil science is no exception, and this paper will discuss the opportunities, but also the risks, that are associated by the growing influence of information technologies on soil science.
The Data Deluge in Soil Science The data deluge has been described using the three “Vs”: volume (the amount of data to process is very important), variety (the data sources and the form they take is increasingly variable and diverse), and velocity (data is changing through time and does so much quicker than before). All of those three aspects represent a significant challenge to soil science. Traditionally, soil data are generated from lab analysis and/or from field observations. Because these measurements are very demanding, both in time and PAGE 176 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
associated costs, the number of locations sampled is usually low. But the development of sensing methods proposes another type of soil data. With the help of information technologies, soil properties are observed indirectly, but at a much higher rate, in time and/or space, and for a fraction of the price per sample. The trade-off to this is that the form that the flow of incoming data takes is different from the traditional soil datasets. An example of such developments is proximal soil sensing (PSS). Proximal soil sensors enable field analysis of soil attributes, such as soil carbon and moisture, and on-the-go sensor surveys can define the spatial variability of these soil attributes. Further, wireless sensor networks can be deployed to monitor the temporal change of the spatially defined soil attributes. These tools make it possible to map and monitor changes of soil attributes in the soil profile at a resolution previously impractical. A good review of the ever-growing ecosystem of sensors available for soil applications has been proposed by Adamchuk and Viscarra-Rossel (2010).
Development of Analysis Capabilities to address Data Growth The rapid and parallel growth of computing capabilities with the data availability has allowed scientists to investigate complex systems using quantitative methods. This is a new paradigm for science: using data mining techniques, scientists can try to discover new patterns from the data itself, rather than formulating a hypothesis and testing it empirically (Hey et al., 2009). Such approaches have been applied in soil science, and pedotransfer functions (PTF) are generated by interrogating important soil profile databases (Bouma, 1989).
Environmental data, with satellite missions such as Landsat or SRTM, are providing soil scientists with environmental data at a very good spatial and temporal resolution. As a consequence, these data are providing scientists with prior information on the areas they may wish to study, and such environmental covariates can be used to inform their sampling design (de Guijter et al., 2006): for example, the sampling locations at which soil spectra have been taken, or at which the soil moisture sensors have been installed, could be determined by interpretation of a stack of environmental data layers. Digital soil mapping (McBratney et al., 2003), is another application of data mining methods in soil science: soil profile data, whether generated using traditional lab methods or using sensing methods such as visible near-infrared spectroscopy, can be combined with environmental covariates to map soil attributes over the landscape. Of course, the application of quantitative methods to soil science, either to predict new soil properties from measured ones (PTF) or to map given soil properties in space (DSM), is not new and has been occurring in our field for the last two decades at least. However, the development of the computing power necessary to process the ever-increasing amounts of available environmental data is occurring in new platforms for soil scientists: while computing power is developing, the amount of data to be processed has well kept up with the Moore’s law. As a consequence of this, while most of the computing tasks have traditionally been done on single core, desktop computers, computing efficiently now requires to (i) move to high-performance, multi-core cluster computers, and (ii) to do the processing task close to the data to limit the overheads associated with moving big chunks of data around. To enjoy the benefits of high-performance computing, soil scientists need to adapt their analysis and tools, and change the way their data are stored and distributed. PAGE 177 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Reproducibility and Data Re-use in Soil Science The ability for scientific results to be reproduced by peers is at the very heart of the scientific method. The computational developments in science mean that to be fully reproducible, science needs a shift towards the way it manages its data and analysis methods - and makes these available publicly. On top of the guaranteeing the necessary reproducibility of the claimed results, data and model availability are also working hand in hand to accelerate scientific progress and upscale (Wolkovich et al., 2012). By their nature, data-oriented approaches are often weaving various sources of data together, with data availability and structure guiding the pace at which such science can progress. This is thus raising the value of datasets than have been collected in the past. Unfortunately, scientific data has traditionally be managed on a per-project basis, making data re-use a very challenging task. A consequence of this is “dark data” - data that has been collected but has become invisible to scientists because it is inaccessible (Heidorn, 2008). Shining a light on “dark data” by making it available will ensure that it can be used for further applications, improving the cost-efficiency of such datasets. It’s an opportunity for soil science to give a second youth to legacy data, and this has been one of the major drivers of the Global Soil Map project (Hempel et al., 2014). But as the development of information technologies requires researchers to make their data publicly available, those technologies also make it easier to share and distribute data. The data modelling methods, which were first introduced in the domain of software development, are helping formalise the scientific concepts and objects that scientists are manipulating, along with their relationships. These data models explicitly define the structure of data. In soil science, the ANZSoilML initiative (Simons et al., 2013) is aiming at modelling soil science data from the pedologist’s perspective. The ANZSoilML model describes and organises soil data, but it also provides for the capture of necessary metadata, ensuring that the data generation process is documented, and reproducible. Data models like ANZSoilML, if adopted by the community, are also ensuring the interoperability of the various data sources, which lowers the cost of the data collation for the scientist. Such data models can then be implemented in an operational solution using web services. These are standardised protocols to stream data over the Internet rather than to store it in downloadable files. The advantages of web services over static files are that they allow data to be more dynamic, providing users with updates as data change. Web services also allow the use of new tools, beyond the traditional spreadsheet on a desktop computer, facilitating the weaving of various data sources on various computing platforms (desktop, cloud, but also mobile).
Discussion The data revolution is impacting most aspects of science, and soil science is no exception. With the contribution of technologies such as remote and proximal soil sensing, the data used to improve our understanding of soil is becoming more diverse, more bulky and more dynamic. High-performance computing hardware and data mining are supporting these important mutations of data, and enable numerical approaches in soil science. However, those upcoming changes require scientists to adapt their tools to fully benefit from these developments. In particular, a shift towards data-driven approaches is requiring data to be openly available, so to maintain reproducibility of the scientific results, and to foster the upscale of those results. Empirical data also suggests that PAGE 178 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
there is a significant trend from scientific journals to make not only data, but also code, available alongside papers published in their columns (Stodden et al., 2013). More than just collecting sensor data, the “data revolution” is an opportunity for soil science to capture expert knowledge and to acknowledge legacy data. Ensuring interoperability and availability of soil data is not only a requirement for the emerging sensing technologies, but can also improve the visibility of the staggering amount of soil information that is not accessible anymore.
References Adamchuk V, Viscarra-Rossel R (2010) Development of on-the-go proximal soil sensor systems. In ‘Proximal Soil Sensing’. (Eds R Viscarra-Rossel, AB McBratney, B Minasny) pp 15-28. (Springer:Berlin) Bouma J (1989) Using soil survey data for quantitative land evaluation. Advances in Soil Science 9, 177–213. de Gruijter J, Brus D, Bierkens M, Knotters M (2006) ‘Sampling for Natural Resource Monitoring.’ (Springer:Berlin) Heidorn PB (2008) Shedding light on the dark data in the long tail of science. Library Trends 57(2), 280-299. Hempel JW, McBratney AB, Arrouays D, McKenzie NJ, Hartemink AE (2014) GlobalSoilMap project history. In ‘GlobalSoilMap: Basis of the global spatial soil information system’. (Eds Arrouays D, McKenzie NJ, Hempel JW, Richer de Forges AC, McBratney AB) pp 3-8. (Taylor & Francis:London) Hey T, Tansley S, Tolle KM (2009) ‘The fourth paradigm: data-intensive scientific discovery.’ (Microsoft Research:Redmond) McBratney AB, Mendonça Santos ML, Minasny B (2003) On digital soil mapping. Geoderma 117(1), 3-52. Simons B, Wilson P, Ritchie A, Cox S (2013) ANZSoilML: An Australian-New Zealand standard for exchange of soil data. In ‘EGU General Assembly Conference Abstracts’ (Vol. 15, p. 6802). Stodden V, Guo P, Ma Z (2013) Toward reproducible computational research: an empirical analysis of data and code policy adoption by journals. PloS one, 8(6), e67111. Wolkovich EM, Regetz J, O’Connor MI (2012) Advances in global change research require open science by individual researchers. Global Change Biology, 18(7), 2102-2110. PAGE 179 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
SOIL CHANGE MATTERS WORKSHOP POSTERS – STATIC Soil changes after fire across an Atlantic rainforest – boundary vegetation transition in Lavras, Brazil DMG Apgaua1, RM Santos1, MAL Fontes1, AT Oliveira-Filho2, DYP Tng3, DMJS Bowman3
1 Universidade Federal de Lavras, Departamento de Ciências Florestais, Câmpus Universitário 37200000, Lavras, Minas Gerais, Brasil, [email protected] 2 Universidade Federal de Minas Gerais, Departamento de Botânica, Av. Antônio Carlos 6627, Belo Horizonte, Minas Gerais 31270-901, Brasil 3 University of Tasmania, School of Plant Science, Private Bag 55, Hobart, Tasmania 7001, Australia
Abstract Studying fire effects on soil across rainforest – woodland boundaries is paramount for understanding fire – vegetation – soil interactions. From thirty-eight 300m2 permanent plots along a transect from an Atlantic rainforest out into Eremanthus-dominated boundary-zone vegetation in Lavras (21° 19.664’S 44° 58.546’ W), Brazil, we collected surface soils (0-20cm depth) in 2000 and after a fire in 2011. From these data we compare the soil chemistry of (a) unburnt rainforest (12), (b) burnt rainforest (15), and (c) boundary-zone burnt plots (11). Soils were analysed for potential acidity, pH, nutrient concentrations of P, K, Ca, Mg, Al, and organic matter. Two-Way ANOVA showed that boundary-zone soils has typically higher potential acidity (15.6 cmol/dm3) and higher level of Al (2.9 cmol/dm3) than in the rainforest (12.9 and 2.3 cmol/dm3, respectively). pH was significantly lower in 2011 (F = 3.7 p = 0.03) regardless of fire. Although P was higher in 2011 (F = 3.5 p = 0.03), it is counteracted by the also higher Al (F = 3.6 p = 0.03). P (F = 3.4 p < 0.001) increased especially in the boundary-zone. In the burnt plots, K increased (F = 19.6 p = 0.03) while organic matter (F = 19.6 p < 0.0001) and Ca (F = 10.7 p < 0.0001) decreased. Mg was not significantly changed in factors interaction. Our results show that soil properties changes after fire, but also without fire due to other processes. These may in turn have significant implications for the control of rainforest boundaries. PAGE 180 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Options for improving soil health in the Tweed Valley Peter E. Bacon 1, Sebastien Garcia-Cuenca2, Hamish Brace3 and Eli Szandala4
1 Woodlots and Wetlands Pty Ltd, 220 Purchase Road, Cherrybrook, NSW 2126, [email protected] 2 Sebastien Garcia-Cuenca, formerly Tweed Shire .Council PO Box 816 Murwillumbah NSW 2484 3 Hamish Brace, Tweed Shire Council PO Box 816 Murwillumbah NSW 2484, [email protected] 4 Eli Szandala, Tweed Shire Council PO Box 816 Murwillumbah NSW 2484, [email protected]
Abstract The warm, humid climate, relatively low slopes and high apparent initial fertility in the Tweed Valley resulted in intensive pasture, vegetable and field crop production. Apparent loss of soil fertility has led to interest in sustainable productivity. This project examines the impact of adding 0, 10 or 20 T/ha/y of compost to 3, 1 ha plots at 30 sites. Landuses included 5 sugar cane, 5 sweet potato, 4 other vegetable, 5 dairy, 5 beef, 2 nut trees, 2 banana and 2 avocado sites. Bulk density and soil organic carbon (SOC) were sampled as per the SCARP protocol (Sanderman et al, 2011). Additionally the surface samples were analysed for a wide range of physical and chemical attributes including trace elements. Pre-clearing SOC was estimated using the method of Gray et al. Mass of SOC in the surface 100 mm ranged from 20 t/ha under a sugar cane and a sweet potato site, to more than 80 t/ha under two permanent pasture sites. TOC in the surface 300 mm ranged from less than 60t/ha to more than 160 t/ha. There were statistically significant relationships among plots within the same landuses. On average, vegetable sites have 72% of predicted 0-100 mm SOC. Conversely, the average dairy site had 1.67 times the predicted SOC. The ability to significantly increase SOC is limited in already fertile pastures. Chemical analysis showed Mo, Co and Se were below detection in many soils. Use of biosolids in the compost mix may alleviate trace element deficiencies in these soils. PAGE 181 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Farm management survey reporting according to land and soil capability. Are New Soil Wales soils being managed sustainably? Chapman GA1 , Gray JM2 and Young JA
1 NSW Office of Environment and Heritage PO Box 3720 Parramatta NSW 2124 greg.chapman@ environment.nsw.gov.au 2 NSW Office of Environment and Heritage PO Box 3720 Parramatta NSW 2124 jonathan.gray@ environment.nsw.gov.au 3 NSW Office of Environment and Heritage PO Box 3720 Parramatta NSW 2124 john.young@ environment.nsw.gov.au
Abstract We spatially compared theoretical upper sustainable land capability limits for land management practices against recorded land capability across NSW. A rule based, 8 class, land and soil capability (LSC) classification has been developed and mapped for acidification, water erosion, structure decline, wind erosion, shallow rocky soils, salinity, mass movement and waterlogging. Across 90,000 NSW map polygons there are 3060 separate LSC signature types out of possible 5040 combinations. The Land Management and Farming in Australia 2010 survey from the Australian Bureau of Statistics (ABS) was reported against LSC. To meet ABS privacy and spatial reporting requirements we generalised LSC mapping using multivariate cluster analysis to summarise LSC signatures into eight separate clusters and commissioned ABS to report according to the LSC clusters. The areal extent of particular practices was compared with the amount of capable land within each of the clusters. The degree to which the areal extent of the practice exceeds the area available for the practice indicates the level of unsustainability. Acidity and sheet erosion relevant practices (N and lime applications, soil testing, ground cover monitoring, traffic control and stubble management practices) were mapped and assessed. Whilst there are numerous spatial generalisations and data limitations, the method shows promise for regional planning. From the present preliminary study it is very apparent that sheet erosion is better managed than soil acidity across NSW. Lack of soil testing and paucity of lime application on vulnerable (poorly buffered) soils over large tracts of agricultural NSW is widespread and unsustainable. PAGE 182 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Effect of organic material input on the densities of granulated organo-mineral fertilizer Lotfi Khiari1, Antoine Karam1, Alfred Jaouich2
1 Département des sols et de génie agroalimentaire, Université Laval, Sainte-Foy, Québec. Canada G1V 0A6. 2 Département des sciences de la terre et de l’atmosphère, Université du Québec à Montréal, Montréal, Québec, Canada H3C 3P8.
Abstract Organo-mineral fertilizers (OMF) based on a judicious mix of both organic materials and mineral nitrogen (N) and phosphorus (P) plant nutrients play an important role in improving soil fertility and crop yield. However, the efficiency and reliability of OMF made from mixtures of separated solid fraction of pig slurry (SPS) and ammonium phosphate as alternative sources of N, P and organic amendments depends, among several factors, on the density of OMF granules. The main objective of this laboratory experiment is to develop OMF with target bulk and granule densities close to granular mineral fertilizer. In a set of experiments, different amounts of nine SPS treated (biodrying and drying) and/ or composted with organic materials (peat moss, wood chips, litter or bark) were mixed with monoammonium phosphate and diammonium phosphate in order to obtain 20, 40, 60, and 80% SPS in the final OMF. As a result, SPS produced lower OMF densities. The bulk and granule densities of OMF decreased with increasing of organic material input from respectively 0.9 g mL-1 and 2 mg mL-1 to less than 0.3 g mL-1 and 0.9 mg mL-1 depending of the type of SPS. The highest density values were obtained with three OMF types: i) SPS composted with peat moss, ii) SPS composted with bark, and iii) SPS obtained by means of flocculation-sedimentation. The target bulk and granule densities values of 0.6 mg mL-1 and 1.2 mg mL-1, respectively, can be obtained with more than 60% of SPS in OMF. The density of OMF depends on the density and rate of the organic component of the OMF.
PAGE 183 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Flowering plants for pollinators vs. biofuels: What to do with forest harvest residue? M.J.McCavour1,2
1 Loyola College for Diversity and Sustainability, Concordia University, Montreal, Québec, Canada 2 Le Centre d’étude de la forêt, Département des sciences biologiques, Université du Québec à Montréal (UQAM), Québec, Canada [email protected]
Abstract Post- harvest residue (“slash”) is increasingly viewed by many countries as a source of biofuel. At the same time, maintenance of ecosystem services is increasingly viewed as integral to sustainable forest and agricultural management. In order to maintain soil fertility and biodiversity, indicators have been developed to define how much slash should remain on site, but few studies have considered the potential ecological impact of the spatial distribution of forest harvest residue. We show that soil N and P is higher, and flowering species achieve far greater abundance, growth and reproductive output, on or near slash piles than elsewhere in 7-year-old hybrid poplar (Populus) plantations. These aggregations of fine wood debris are islands of high soil fertility and light, leading to a greater abundance, growth, and reproduction of flowering plants and therefore serve as hotspots for pollinators and frugivores. Pollination by animals is a crucial step in the production of much of the world’s food supply, and the great majority of our most nutritious foods. However, domesticated honeybees are in decline. Though there is a diverse set of proximate reasons, all are ultimately linked to overdependence of industrial agriculture on this single species. One solution is to augment crop pollination by increasing wild pollinator habitat and connectivity in agricultural landscapes, including those featuring biofuel crops of fast-growing hybrid trees. Retention of piled or windrowed slash in hybrid plantations can serve as habitat for wild pollinators, thus increasing the security of our food supply, and concomitantly contributing to biodiversity maintenance through the provision of wild food resources to frugivores.
Estimating the expected rates of change in soil carbon following changes in land management practices _ a proposed normalising equation Brian Murphy1, Andrew Rawson2, Warwick Badgery3
1 Honorary Scientific Fellow, NSW Office of Environment and Heritage, Cowra, NSW. brian.murphy@ environment.nsw.gov.au 2 Associate Professor School of Agriculture and Wine Sciences, Charles Sturt University, Orange, NSW. 3 NSW Department of Primary Industries, Orange Agricultural Institute, 1447 Forest Rd, Orange, NSW, Australia 2800 Abstract Significant efforts have been made in recent years to establish soil carbon levels in agricultural landscapes across Australia (eg SCaRP). For farmers to receive payment for changes in soil C (derived from land management improvement), knowledge of SOC PAGE 184 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
sequestration rates are imperative, especially for short term contracts. However specific knowledge of sequestration rates is lacking for many environments in Australia, and surrogate measures or indices may provide the best option in the short term. Existing data on soil C indicates that soil C sequestration potential differs across the landscape, for example, SOC levels under basalt derived soils have been shown to be much higher than those under granite derived soils across varying climatic environments. Similarly, rates of change in SOC differ across climates zones and soils. Therefore there is a requirement for simple equations or indices to derive or estimate these rates.
The proposed equation predicts the rate of change in soil carbon for the first 5 years of the change in land management and uses the initial level of the soil carbon store and the long term equilibrium level of the soil carbon store under the expected or contracted land management system. The maximum change in soil carbon stores that can be expected under changes in agricultural land management and the maximum rate of change for a soil/climate zone are also inputs. This equation or index has the potential to make maximum use of the existing data from SCaRP, FullCAM, existing publications, legacy data and local knowledge.
Impact of soil copper levels on vineyard soil bacteria, fungi and nematodes Whitelaw-Weckert MA1, Rahman L1,
1 National Wine & Grape Industry Centre, Charles Sturt University, New South Wales Department of Primary Industries, Locked bag 588, Wagga Wagga, NSW 2678, Australia.
Abstract The long term effect of moderate soil Cu accumulation on soil biology was the focus of this investigation. Culturable oligotrophic bacteria, copiotrophic bacteria, total fungi, cellulolytic fungi and pseudomonads plus nematode trophic groups (total plant parasitic nematodes, bacteria-feeding nematodes, omnivorous nematodes, predatory nematodes and total free-living nematodes) were monitored in two NSW vineyards (Tumbarumba and Wagga Wagga) over a 2 year period; and later, two Griffith vineyards (Yenda and Hanwood) over a 3 year period. Linear regression analysis showed that DTPA extractable soil Cu (Cu-DTPA) at levels below 19 mg kg-1 was associated with increased plant parasitic nematode populations, and reductions in soil populations of oligotrophic bacteria, pseudomonads, actinomycetes, cellulolytic fungi and free living (beneficial) nematodes. These results are novel because they involve relatively low soil Cu concentrations in comparison to most similar studies. PAGE 185 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Changing epistemic uncertainties in soil classification and digital mapping D. B. Rees1 , K.K. Benke1 and J. Hopley2
1 Farming Systems Research Division, Department of Environment and Primary Industries – Parkville Centre, PO Box 4166, Parkville, Victoria, 3052, Australia. Email: [email protected], kurt. [email protected] 2 Farming Systems Research Division, Department of Environment and Primary Industries – Epsom Centre, Cnr Midland Hwy and Taylor St, Epsom, Victoria, 3551, Australia. Email: jonathan.hopley@ depi.vic.gov.au
Abstract Changing epistemic uncertainties are discussed in the context of soil classification leading to the production of a digital soil map. Uncertainties are associated with a diverse range of data inputs with each input subject to possible errors in data acquisition and transmission. Epistemic uncertainties relate primarily to the information content, or lack thereof, rather than to variability in covariates or inputs – which are referred to as aleatory uncertainties. The discussion covers topics related to changing uncertainties in data and its collection over time, and the possible effects on classification of soils and their representation at different scales and the surrogates used.
Introduction There is a hierarchy of information used in soil classification.Soils are generally nested initially under landform and become more easily understood and identified at the appropriate scale. At all levels of a proposed hierarchy, there is spatial uncertainty in the delineating boundaries and also categorical uncertainty in the classification. The current hierarchy for national level information on soils, the Australian Soil Resource Information System (ASRIS 2011) has 7 levels. Soil differentiation is not viable till at least level 4. Thus landscape information such a geomorphology provides the framework for soil. At levels 6 and 7 soil descriptions are much more specific with level 7 being the site description.
Role of classification Geomorphology (landform) mapping provides the landscape framework for soil forming processes and inferred soil types. Landscape-soil relationships assume that particular soils are associated with particular landform/topographical positions and there is a pattern of occurrence within the map unit. The degree of survey (sampling) depends upon land characteristic-variability (for analogue classification) for accepted uncertainty. For example, consider landform patterns and their relative relief classes, such as Plain ( 0-9 m), Rise (9-30 m), Low hills (30-90 m), Hills (90-300 m) and Mountains (>300 m) (NCST 2009). This classification of landform may not be fine enough to cope with low relief landscapes of Victoria such as the Riverine Plains or Mallee regions. There is uncertainty of measurement (spatially) as well as variability with classes that may reflect variability of other properties, such as soil type. Uncertainty in classification of soil type depends on factors differentiating soil types. Local soil variation (i.e. within 2-3 m) such as gilgaied terrain means reoccurring soil patterns at a fine scale.
Changing epistemic uncertainties Epistemic uncertainties reflect lack of information, or ignorance, and are distinct from statistical variability, which is known as aleatory uncertainty; although epistemic PAGE 186 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
uncertainties may also be enumerated (Benke et al. 2007). Epistemic uncertainties in soil classification, which may change over time with improving technology, are involved in the following ›› Data gathering quality and types of data – legacy data, missing data, expert opinion ›› Categorical error - what constitutes a soil type ? Changes in definition? ›› Measurement error – especially with boundary issues, e.g. pH maps ›› Digital vs analogue data - truncation errors in conversion Analogue classification allows for quick attribution based on assumed relationships. Analogue is more subjective and legacy-based, whilst the modern trend is moving to a quantitative digital presentation. Issues of error propagation in the transfer process now become apparent, with lower uncertainty associated with more recent data collection.
Texture The change of texture with depth is an issue in soil classification. The degree and abruptness of texture change employed in classification has relevance to practical outcomes viz. assumed water movement and drainage. Texture contrast soils have lighter textured horizons abruptly or clearly overlying heavier textured subsoils. The gradient and magnitude of the change in texture represents a fuzzy boundary encapsulating the uncertainty in the transition. With changes in time and methodology of measurement, these transitions are likely to be more accurately assessed and therefore associated with less uncertainty.
Boundary conditions Uncertainty of classification is affected by horizon boundary features. For example, in the case of colour, the difference between grey and brown might be very minor (one colour chip in standard Munsell colour chart). The effects of changing illumination in the laboratory or field, or changes in time with either method add new uncertainties in judgement compared with historical data.
Enumeration of epistemic uncertainties An example of epistemic uncertainty and its enumeration graphically illustrates its nature (see Figure 1 and Phoon and Kulhawy 1999; Benke et al. 2007; Davidović et al. 2010). Spatial variation of a soil property as a function of depth may be expressed as follows (Davidović et al. 2010):
ξ(z) = t(z) + w(z) + e(z) (1) where z is depth, ξ(z) is a soil property, t(z) is deterministic trend function, w(z) is the deviation from trend (i.e. a random component) and e(z) is measurement error. Note that the trend (t) is fitted by a smooth deterministic function. Measurement error (e) is an epistemic uncertainty that can introduce additional variability to the soil property measurement. It is reducible by experimental means. Another epistemic uncertainty is due to soil properties in unsampled regions following the well-known spatial PAGE 187 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
interpolation technique called kriging. For example, if soil property data are available at specific locations, it is possible to estimate the value at other locations by predicting the
average value of X at specified points. If X1,X2,..., Xn are known values of parameter at the
points x1,x2,..., xn defined by coordinates, then the estimated value of parameter at point x is given by:
(2)
where βi are weights applied to the respective values Xi , which sum to unity (Davidović et al. 2010). The value of the property at other locations represents lack of information (epistemic uncertainty) and this approach is an attempt at enumeration of this information. The approach may be regression-based and future uncertainty may be decreased by new non-linear approaches. Another area for enumeration of epistemic uncertainty relates to category classification error. Repeated category selection against a set of criteria by a number of observers may lead to an estimate of error probability in classification using statistical sampling techniques.
Figure 1. Vertical variation of a soil property, e.g. organic carbon content (Davidović et al 2010). The assignment of soil type to separate land units of the same designation might be compromised by the limited data (soil sites available) such that large areas may be assigned an inappropriate soil type given the range of intra unit variability. The assignment of soil type might also be uncertain if the possible choices are perceived with limited information. For example, large tracts of public land with variable soil PAGE 188 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
types might be assigned on the basis of very few data points which are not necessarily representative, such as Tenosols instead of Kandosols or Dermosols. The example presented in Figure 2 shows a land unit with a similar proportional distribution of landform elements within a dune and swale landform pattern. In this example, the assignment of a dominant soil class to the land unit will have a high degree of uncertainty.
Figure 2. A land unit in the Victorian Mallee comprising of sandy rises (dark) and heavier swales (light). Soil sites are sparse and located on different landform elements.
Uncertainties in the development of the Victorian Digital Soils Map The developers of the Victorian digital land unit map have acknowledged the uncertainties inherent in such a production (Hopley et al. 2013). It was noted that the individual surveys collated were collected at different times, by different surveyors, and according to different objectives. The scale of mapping, and volume, quality and interpretations of survey data is variable. These factors, which are epistemic in nature results in variable confidence in the assignment of a dominant soil and its Australian Soil Classification (ASC) to each land unit (Isbell 2002). It was observed that harmonising the scale of input surveys may result in improved consistency in the quality of the map outputs. This could be achieved through downscaling mapping to the smallest survey scale or alternatively by up-scaling mapping using spatial disaggregation approaches. Spatial disaggregation of broad scale map units has already been achieved in north west Victoria in the past but was not included in VicDSMv1 (Hopley et al. 2013). The authors suggest there is great difficulty in reducing to a single consistent scale of soil mapping, given the variable soil site density and landscape complexity across Victoria. Validation and refinement of the digital soil map outputs has been undertaken through expert opinion review. This step addresses uncertainties in the soil classification procedure for land units and site profiles. Uncertainties may be reduced in future with new experts and methodologies in future revisions of the map. PAGE 189 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Conclusion There are many changing epistemic uncertainties relating to soil classification and these include errors in category assignment in the taxonomy, changing expert opinion and judgement errors, reduced measurement error of soil properties due to improved techniques, scale-based mis-matches in the soil hierarchy, improvements in assessment of texture and boundary error. The challenge is to develop a framework that includes both changes in epistemic errors that allows for a more comprehensive treatment of uncertainty in soil classification and map production in future revisions.
References ASRIS (2011). ASRIS - Australian Soil Resource Information System. http://www.asris.csiro. au. Accessed November 1, 2013 Benke KK, Hamilton AJ, Lowell KE (2007) Uncertainty analysis and risk assessment in the management of environmental resources. Australasian Journal of Environmental Management, 14,243-249. Carré F, McBratney AB, Mayr T, Montanarella L (2007) Digital Soil assessments: beyond DSM. Geoderma 142, 69-79 Davidović N, Prolović, V and Stojić, D (2010) Modeling of soil parameters spatial uncertainty by geostatistics. Facta Universitatis Series: Architecture and Civil Engineering 8, 111-118. DOI: 10.2298/FUACE1001111D Grimm F, Behrens T (2009) Uncertainty analysis of sample locations within digital soil mapping approaches. Geoderma 155, 154-163. Hopley J, Robinson N, Rees DB, MacEwan RJ, Clarke R, Benke K, Imhof R, Bados D (2013) A digital soil map of Victoria – VicDSMv1. Global Soils Map Conference, Orleans, France. Isbell RF 2002. The Australian Soil Classification. Revised Edition. Collingwood: CSIRO Publishing National Committee on Soil and Terrain (2009) ‘ Australian soil and land survey field handbook (3 edn)’. (CSIRO Publishing: Melbourne) Phoon, KK, Kulhawy, FH (1999). Characterization of geotechnical variability. Canadian Geotechnical Journal 36, 612-624. PAGE 190 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Are soil archives the critical link to understanding changes in soil? NJ Robinson1, A Lilly2, L Karssies3
1 Department of Environment and Primary Industries, Taylor Street, Epsom, Victoria, 3551. Nathan. [email protected] 2 The James Hutton Institute, Craigiebuckler, Aberdeen, Aberdeen AB15 8QH, Scotland. 3 CSIRO Land and Water, GPO Box 1666, ACT, 2601.
Abstract A critical driver of soil science research today is the international focus to double food production to support the global community of the future. To achieve sustainable use of the soil resource, understanding current and historical changes in soil condition will enable us to predict with greater certainty future trends. There is global recognition of the importance of soil archive collections as these can provide baseline data for the assessment of the impacts of climatic and other environmental changes as well as land management on the soil resource. This intrinsic value of legacy soil samples in archives has been highlighted in approaches to understanding changes in long-term soil experiments (Richter et al., 2007). Effectively archived soil specimens are invaluable “time capsules” for assessing temporal changes in soil properties, particularly as new analytical tools become available (Boone et al., 1999). Key to the success of any long-term soil experiment or research trial is that archival needs should be a central component in the investment/establishment of such experiments to maximise its scientific value. Of additional importance is the critical quality control of data and soil samples to yield reliable and assured data for modelling and simulation purposes. Changes in data integrity due to evolution in analytical methods can also represent a significant source of potential error in stochastic analysis methods. Recent examples where these were key considerations include the National Soil Inventory of Scotland (1978 to 1988 and 2007 to 2009) and the National Soil Fertility Project in Victoria (1968 to 1972 and 2011 to 2013).
References Boone, R.D., D.F. Grigal, P. Sollins, R.J. Ahrens, and D.E. Armstrong (1999). Soil sampling, preparation, archiving, and quality controls. p. 3–28. In G.P. Robertson et al (ed.) Standard soil methods for long-term ecological research. Oxford Univ. Press., New York. Richter Jr, D., Hofmockel, M., Callaham Jr, M., Powlson, DS., Smith, P (2007) Long-Term Soil Experiments: Keys to Managing Earth’s Rapidly Changing Ecosystems. Soil Sci. Soc. Am. J. 71:266–279. PAGE 191 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Spatio-temporal changes in soil properties over 40-years in western Victoria – two case studies N. Robinson12, D. Crawford3, B. Marchant4, R. Clark 2, K. Sheffield5, R.MacEwan 2
1 Science, Information Technology and Engineering, University of Ballarat, Australia. Nathan. [email protected] 2 Victorian Department of Environment and Primary Industries, Epsom, 3551. 3 Department of Environment and Primary Industries, 1301 Hazeldean Road, Ellinbank, Victoria, 3821. 4 British Geological Survey, Nottingham, UK. 5 Department of Environment and Primary Industries – Parkville Centre, PO Box 4166, Parkville, Victoria, 3052.
Abstract Maps provide a snapshot of historic soil conditions as information used in the map creation process are often from legacy data reflecting historic conditions (Grunwald et al. 2012). The growing demand for contemporary information on soil condition and how, where and when changes occur to the soil resource are of primary interest in an age of significant land use change and human impact. Recent advances in spatial and temporal analysis of legacy soil site data has successfully estimated the variogram structures of both components using maximum likelihood and the Best Linear Unbiased Predictor (BLUP) using a farmer soil analyses dataset for Victoria (Marchant et al., 2012). In western Victoria, soil analyses are available from a range of sources including paddock samples, soil benchmarking sites, research trials and soil and land inventory sites. This analytical data has been collected over different time-intervals during the last 40-years. Recently satellite imagery corresponding with this time period has also become available. The role of time-series satellite imagery in predicting soil properties, and in spatial and temporal models to predict changes in DSM is largely unknown, however this represents a great opportunity to test and evaluate the potential of this imagery to support spatio- temporal assessments of soil properties over this time-period. Preliminary results of this analysis using spatio-temporal analytical techniques and time- series satellite imagery for the 1970-1980, 1980-1990, 1990-2000, 2000-2012 periods is presented for two dryland agricultural regions within western Victoria.
Soil measurement scale, resolution and cost matters Chris Waring1, Uta Stockmann2, Brett Whelan2
1 Institute for Environmental Research, Australian Nuclear Science & Technology Organisation, Locked bag 2001, Kirrawee DC NSW 2232, Australia [email protected] 2 Soil Security Laboratory, Faculty of Agriculture and Environment, The University of Sydney, Biomedical Building C81, Suite 401, 1 Central Avenue, Australian Technology Park, Eveleigh, NSW 2015, Australia
Abstract Discerning significant change to soil parameters as a consequence of environmental factors or management actions requires systematic measurement. However, measurement cost imposes severe constraints upon the scale, resolution, precision and PAGE 192 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
acceptable error for detecting significant change to soil parameters. A consequence of the cost restriction is less precise information on soil change when more precise information is required. An example of this dilemma is the Carbon Farming Initiative where farmers may receive payment for the creation of carbon credits by sequestration of carbon in soil after following an accepted methodology. If net change in the soil carbon pool after years of accumulation (or loss) is within measurement error farmers may not generate tradeable carbon credits yet would bear measurement costs. Similarly scientists seeking to quantify the efficacy of management practices or soil amendments are required to carefully design studies with sufficient resolution to answer the question posed within a restricted budget. One approach to alleviate the current situation is the use of automated soil sensors, or often described as proximal soil sensing (PSS). Ideally proximal soil sensing technology would significantly lower the cost of measurement as well as improving measurement certainty. A cost & performance comparison of PSS technologies visible - Near Infra-Red (vis-NIR) reflectance spectroscopy and Fast Neutron Activation Analysis (FNAA) with conventional soil sampling and laboratory analysis is presented.
Changes in soil carbon from using broiler litter as an alternative fertiliser for pastures Lisa Warn1
1 Mackinnon Project, University of Melbourne, 250 Princes Hwy Werribee, Vic, 3030, l.warn@ unimelb.edu.au
Abstract Rising costs of conventional fertilisers has led broad-acre livestock producers to seek alternative nutrient sources to fertilise pastures and crops. Spent litter from broiler sheds is an economic alternative, particularly when key inputs like phosphorus fertiliser prices are very high, or where a range of macro nutrients and trace elements are required. Manures and litters also add organic matter to the soil although this benefit has not been quantified for pasture soils. The wider implication of increasing organic matter in soil is the potential for carbon sequestration and carbon trading. Replicated field experiments were conducted at 2 sites in central Victoria, with different soil types, to study pasture and soil responses to 3 rates of broiler litter in comparison with inorganic fertilisers. One aim was to examine the rate at which broadcast litter could improve total soil organic carbon and what impact it had on the different carbon fractions. Soil organic carbon increased by 0.4 - 0.5% in the topsoil where high rates of litter were applied relative to the Control (nil fertiliser) over 4 years. There was a trend for total soil organic carbon in the topsoil to increase by around 5 t/ha at both sites where high rates of litter (20 t/ha over 4 years) were applied, compared with the Control. The increase in soil carbon did not translate into pasture yield responses over and above that of the conventional fertilisers. However, increased carbon stored in soils would have an economic value in the advent of carbon trading. PAGE 193 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
Grape marc char – effects on soil structure, water efficiency and fertilizer productivity Romy Zyngier1, Wendy Quayle2, Hamish Creswell3, Antonio F. Patti4 and Tim Cavagnaro5
1 School of Chemistry, Monash University, Wellington Rd, Clayton, Victoria, 3800, romy.zyngier@ monash.edu.au 2 CSIRO Land and Water/Sustainable Agriculture Flagship, Griffith Laboratory, Research Station Rd, Griffith, New South Whales, 2680, [email protected] 3 CSIRO Land and Water- Black Mountain, Christian Laboratory, Clunies Ross St, Black Mountain, Australian Capital Territory, 2601, [email protected] 4 School of Chemistry, Monash University, PO Box 23 Clayton, Victoria, 3800, tony.patti@monash. edu.au 5 School of Agriculture, Food and Wine, The University of Adelaide, PMB1 Glen Osmond, South Australia, 5064, [email protected]
Abstract: Understanding changes to soil physical characteristics when amended with biochar is vital for assessing the potential of biochar to reduce water and fertilizer crop requirements. To address this a field trial was undertaken to assess the potential for grape marc BC to increase soil water retention, improve hydraulic conductivity, reduce bulk density and increase nutrient availability, with the goal of improving crop performance and yield. Grape marc char was applied to a red-brown sandy-loam planted with rockmelons in NSW, SE Australia, classified as a semi-arid climate. The soil has low carbon content and presents dispersive tendencies and surface crusting under irrigation. Grape marc char was produced via continuous-flow pyrolysis. A composite char was made using equal proportions of four final temperature pyrolysis conditions: 350°C; 400°C; 450°C; and 500°C (in brief: C 47.7%; N 2.58%; pH 8.8 and CEC 89 cmol+/kg). Char was applied at four rates (0, 0.25, 1, 20 t ha-1) with two rates of irrigation (~4 ML ha-1; half rate ~2 ML ha-1) in a split plot design. Initially soils were fertilized with 48kg ha-1 of N and 20.8 kg ha -1 of P, followed by 30 kg ha-1 of N according to “farmers practice”; a half rate was applied to all char treatments. Results from the assessment of crop yield, plant water use, hydraulic conductivity, water and fertilizer productivity, soil water retention and bulk density changes are presented. The potential of char incorporation into irrigation farming systems for increasing water and fertilizer productivity will be discussed. PAGE 194 SOIL CHANGE MATTERS PROCEEDINGS
NOTES PAGE 195 SOIL CHANGE MATTERS PROCEEDINGS
NOTES PAGE 196 SOIL CHANGE MATTERS PAPERS AND ABSTRACTS
THE DEPARTMENT OF ENVIRONMENT AND PRIMARY INDUSTRIES THANKS THE SPONSORS
Principal sponsor
Co-operative Research Programme on Biological Resource Management for Sustainable Agricultural Systems Gold sponsor
Silver sponsor
Bronze sponsors
Supporters Authorised and published by the Department of Environment and Primary Industries, 8 Nicholson Street, East Melbourne. Print managed by Finsbury Green, March 2014.
ISBN 978-1-74326-806-3 (Print) ISBN 978-1-74326-807-0 (pdf)
OECD disclaimer The opinions expressed and arguments employed in this publication are the sole responsibility of the authors and do not necessarily reflect those of the OECD or of the governments of its Member countries.
OECD CRP accreditat ion The Workshop was sponsored by the OECD Co-operative Research Programme on Biological Resource Management for Sustainable Agricultural Systems, whose financial support made it possible for some of the invited speakers to participate in the Workshop.
All details are correct at the time of publication. The Organising Committee reserves the right to change the programme and functions at any time without notice.
Opinions expressed by authors in their abstracts and during their presentations are personal and do not necessarily reflect the opinions of the Organising Committee members or Victorian Department of Environment and Primary Industries.
Accessibility If you would like to receive this publication in an accessible format, such as large print or audio, please telephone DEPI Customer Service on 136 186 or email [email protected]