Sample Paper for the 19Th World Congress of Soil Science, Brisbane 2010

Comparison of agricultural land-use change and soil capability A case study by using LC and MicroLEIS DSS in Sevilla Province (Spain)

María Anaya-RomeroA, Rafael PinoB, Braulio AsensioC, José M. MoreiraD and Diego de la RosaA

ASpanish National Research Council (CSIC), Institute of Natural Resources and Agrobiology (IRNAS), Reina Mercedes 10, 41012 Sevilla, Spain, E-mail [email protected], [email protected]. BDepartment of Statistics and Operational Research, University of Seville, Reina Mercedes s/n, Sevilla, Spain, E-mail [email protected] CDepartment Physical, Chemical and Natural Systems, University of Pablo de Olavide, Carretera de Utrera km. 1. 41013 Sevilla, Spain, E-mail [email protected], [email protected] DDG Sustainable Development and Environmental Information. Regional Deputy Minister of Environment. Regional Government of Andalucia. Ave. Manuel Siurot 50, 41071 Sevilla, Spain, E-mail [email protected]

Abstract

This article analyses the land use and land cover changes registered in agricultural lands between 1956 and 2003 in Seville province (SW Spain). Likewise, it compares the agricultural land use change with land capability with purpose to analyse dysfunctions relating to the agricultural capacity of the territory. As well, this paper examines the factors affecting agro-ecological land use in the province in the light of their implications for agro-forestry sustainability. We formulated a methodology to quantify and explain the change that can be used in similar cases elsewhere. The approach is tested in Seville province, considering the Land Cover (LC) maps of 1956 and 2003 (E 1:25 000) made by Andalusian Network of Environmental Information (REDIAM) belonging to Regional Government of Andalucia. The agricultural capability of the soil was determined using the MicroLEIS system and data from the soil data base of SEISnet-IRNAS. There are significant differences in the extent and rate of agricultural land use change at a regional level. The present circumstances in the province are favourable for a reversal of agro-forest uses; nevertheless urbanization process is the major pressure in the best agricultural regions in the province.

Key Words Land Cover (LC), land use change, land evaluation, soil capability, MicroLEIS.

Introduction

The role of the soil is an essential component to evaluate the optimal land use for each particular zone based on its own capability and vulnerability (FAO 1976; 1978). On the other hand, land use change, influenced by process like urbanization, industrialization, and intensive agriculture often result in rapid landscape changes, losses of ecological capacity, diversity, and scenic beauty, as well as damage to historically valuable cultural landscape (Bastian, Krönert, & Lipský, 2006, p. 359). Corine Land Cover 2000 (CLC2000) contributes to the knowledge of the land cover (LC) and its changes in 24 European countries. LC reflects the biophysical state of the real landscape (including the effects of human activity on the biophysical unit). This is the reason why LC data are increasingly used for the derivation of various landscape attributes such as its changes, diversity, forecasting, etc. and in the modelling of its different properties (Feranec, J., et al., 2009). The aim of this paper is: 1. To evaluate quantitatively and objectively the agricultural land-use change between 1956 and 2003. 2. To compare agricultural land-use change with the land capability prediction in order to analyze dysfunctions relating to the agroforestry capacity of the territory.

Methods

Study area Sevilla province is located in the Mediterranean region of Andalucia, SW Spain. The climate is semi-arid, with mild rainy winters, and hot dry summers of high solar radiation and a high rate of evaporation. This seasonal contrast is exacerbated by the erratic and unpredictable rainfall distribution from year to year, and crops can suffer from moisture deficits even during years receiving the mean precipitation.

© 2010 19th World Congress of Soil Science, Soil Solutions for a Changing World 1 1 – 6 August 2010, Brisbane, Australia. Published on CDROM. The approximate geographic coordinates of Sevilla Province are 36º51’ to 38º 12’ N and 5º04’ to 6º30’ W. Its slopes range from <2 to 30%, and the elevation is from 2 to 740m above sea level. The total province area is 1 425 726 ha.

Benchmark soils There were selected seven benchmark sites from the soil data base of SEISnet-IRNAS (Evenor-Tech, 2009). A general description of each site is summarized in Table 1.

Table 1. General description of the selected seven benchmark sites in the Mediterranean Province of Sevilla. Natural region Typical soil USDA-98 Average Elevation Approx. profilea Soil classification slope (%) (m) extension 103 ha Aljarafe SE0201 Typic Rhodoxeralf 2 - 8 100 59.2 Campiña SE0302 Typic Chromoxerert 8 - 16 60 558.8 Estepa SE0101 Entic Haploxeroll 16 - 30 480 59.1 Marismas SE0103 Salorthidic Fluvaquent <= 2 2 99.8 Sierra Norte SE0401 Palexerult 8 - 16 740 375.8 Sierra Sur SE0701 Vertic Xerorthent 16 - 30 250 115.3 Vega SE0501 Typic Xerofluvent <= 2 10 157.6 aFrom the SEISnet-IRNAS (Evenor-Tech, 2009).

The typical soils were selected because they occupy large proportions of the corresponding natural region (Fig.1.). For each of the benchmark site, a representative meteorological station was selected, based on monthly mean climate variables for the long period 1961–1990.

Figure 1. Natural regions of the Mediterranean Province of Sevilla, SW Spain.

The MicroLEIS DSS The agro-ecological decision support system MicroLEIS (De la Rosa, et al., 2004; 2009) is used to design the most sustainable land use and management. This DSS is based on the multifunctional evaluation of soil quality, using input data collected in standard soil surveys, and with particular reference to the peculiarities of the Mediterranean region. In the present study, there were used two models included in MicroLEIS DSS. The models are Terraza and Cervatana. While Terraza gives an empirical prediction of the bioclimatic deficiency of a site, Cervatana model forecasts the general land use capability or suitability for a broad series of possible agricultural uses. The models were applied for the selected seven benchmark soils.

© 2010 19th World Congress of Soil Science, Soil Solutions for a Changing World 2 1 – 6 August 2010, Brisbane, Australia. Published on CDROM. Presently, a Spin-off from the CSIC (named Evenor-Tech; www.evenor-tech.com) is being launched in basis to the MicroLEIS technology.

Land Cover and Contingency tables The LC nomenclature of Andalucia comprises three class levels characterized by quoted attributes. The present study was focused in the first level (artificial surface, wetlands and water bodies, agricultural areas, and forest and natural areas) which indicates the major categories of LC on the planet. To evaluate the coincidence between the LC of 1956 and 2003, we constructed contingency tables for each natural region, (Table 2 shows an example for Aljarafe natural region). These tables throw data for understanding the main pressure for biodiversity and natural resources sustainability. In this sense, the LCF in agricultural areas was analysed in order to stand up the principal processes related to the agro-forestry capacity of the territory.

Table 2. Contingency table: LC1 Change in Aljarafe natural region (Sevilla, Spain) for the period 1956-2003. Aljarafe Land Use (2003) 1 2 3 4 Surface (1956) Area (ha) 559 0 22 1 582 1 Artificial surface % 96,05 0,00 3,78 0,17 100 ) 6

5 Area (ha) 3 705 60 139 907

9 2 Wetlands and water bodies 1

( % 0,33 77,73 6,62 15,33 100

s Area (ha) 4.073 67 43.666 714 48.520

U 3 Agricultural areas

d % 8,39 0,14 90,00 1,47 100 n

a Area (ha) 297 83 4.015 4.803 9.198 L 4 Forest and natural areas % 3,23 0,90 43,65 52,22 100 Area (ha) 4.932 855 47.763 5.657 59.207 Surface (2003) % 8,33 1,44 80,67 9,55 100 1LC. Land Cover was used following Moreira (2007).

Results

Results of applying Terraza (bioclimatic deficiency) model and Cervatana (land capability) model in the selected seven benchmark sites are shown in Table 4, additionally this table indicates the agricultural actual land areas and LCF resulting from the respective contingency tables of each region.

Table 3. Comparison between bioclimatic deficiency-land capability results from Terraza and Cevatana models1 and agricultural actual land use and land cover flow during 1956-2003. Natural region Bioclimatic Land capability class3 Agricultural Land (2003) LCF4 (1956-2003) LCF5 (1956-2003) deficiency (Benchmark site) 103 ha [% Natural region] 103 ha [MLU] 6 103 ha [MLU] 6 (GPL, day) 2 Aljarafe 210 S2rb 47.8 [80.7] 4.1 [4] 4.8 [1] Campiña 250 S2tl 517.3 [92.6] 43.8 [4] 13.4 [1] Estepa 210 S3t 51.7 [87.5] 0.6 [4] 1.9 [1] Marismas 210 S2lb 56.5 [56.5] 32.4 [2] 18.7 [4] Sierra Norte 270 Ntl 47.1 [12,5] 4.8 [4] 5.3 [4] Sierra Sur 250 Nlr 72.8 [63.2] 5.9 [4] 4.3 [4] Vega 210 S1 104.4 [66.3] 14.1 [4] 17.2 [1] 1Development, inputs and validity of these models are described in De la Rosa et al. (2004). 2GPL, length of growing period. 3Land capability classes: S1 – Excellent; S2 – Good; S3 – Moderate; N – Not suitable. Limitation factors: t – topography: slope type and slope gradient; l – soil: useful depth, texture, stoniness/rockiness, drainage, and salinity; r – erosion risk: soil erodibility, slope, vegetation cover, and rainfall erosivity; b – bioclimatic deficiency (GPL). 4LCF. Land Cover Flow from other uses to agricultural areas. 5LCF. Land Cover Flow from agricultural areas to other uses. 6MLU. Main Land Use affected, 1-Artificial surface; 2-Wetlands and water bodies; 4-Forest and natural areas.

Five application sites are classified as arable or best agricultural lands, and another two as marginal or unsuitable lands. The Vega site (Typic Xerofluvent soil) presents the highest capability for most agricultural crops. In this region 11 610 ha were converted from forested and natural areas to agriculture during the last © 2010 19th World Congress of Soil Science, Soil Solutions for a Changing World 3 1 – 6 August 2010, Brisbane, Australia. Published on CDROM. half century. This was high linked with urban development in the zone, where 11 099 ha were transformed from agriculture to urban, industrial or transport uses. In contrast, the Sierra Norte site (SE06: Palexerult soil) and the Sierra Sur site (SE07: Vertic Xerorthent soil) show the most-unfavorable conditions. The length of the growing period, the slope, and the soil depth are the major limitation factors in this agro-ecological zoning classification of Sevilla sites. There was a low land cover flow in both areas during the period 1956- 2003. It should be resalted the high percentage of agricultural use in Sierra Sur despite these land capability results. In this sense, changes in land use from natural habitat to intensively tilled agricultural cultivation are one of the primary reasons for soil degradation. Deforestation for agricultural needs and overgrazing have led to severe erosion process in the region. In summary, a positive correlation between current land use and potential land capability would be necessary.

Conclusion

The main conclusion of this paper is that comparing land use with soil type information in decision-making is at the heart for sustainable use and management of agricultural land. This agro-ecological approach can be especially useful when formulating soil-specific agricultural practices to reverse environmental degradation, based on the spatial variability of soils and related resources. Any kind of agricultural management system will have a negative environmental impact when applied on land with very low suitability for agricultural uses. In the Mediterranean region, for example, marginal agricultural land under any kind of farming system is the ideal scenario for soil erosion. On the other hand, high capability land should be preserved in order to attend to a sustainability development.

References

Bastian O, Krönert R., Lipský Z (2006). Landscape diagnosis on different space and time scales – a challenge for landscape planning. Landscape Ecology, 21(3), 359–374. De la Rosa D, Anaya-Romero M, Diaz-Pereira E, Heredia N, Shahbazi F, (2009). Soil-specific agro- ecological strategies for sustainable land use – A case study by using MicroLEIS DSS in Sevilla Province (Spain). Land Use Policy 26, 1055–1065. De la Rosa D, Mayol F, Diaz-Pereira E, Fernandez M, De la Rosa Jr D (2004). A land evaluation decision support system (MicroLEIS DSS) for agricultural soil protection. Environmental Modelling & Software 19, 929–942. Evenor-Tech (2009). Soil database SEISnet-IRNAS. http://www.evenor-tech.com. Feranec, J., et al. (2009). Determining changes and flows in European landscapes 1990–2000 using CORINE land cover data, Applied Geography, doi:10.1016/j.apgeog.2009.07.003 FAO (1976). A framework for land evaluation. FAO Soil Bulletin 32, Rome. FAO (1978). Report on the agro-ecological zones project. World Soil Resources Report 48, Rome. Moreira JM (2007). Mapas de usos y coberturas vegetales del suelo de Andalucía.Escala 1/25000. Guía Técnica. Regional Deputy Minister of Environment. Regional Government of Andalucia. Weber JL (2007). Implementation of land and ecosystem accounts at the European environment agency. Ecological Economics, 61(4), 695–707.