EROSION AND WATER RESOURCES ASSESSMENT IN THE UPPER INABANGA WATERSHED, PHILIPPINES: APPLICATION OF WEPP AND GIS TOOLS
BY
IMELIDA C. GENSON (BSc Ag Eng)
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE (HONOURS)
WATER RESEARCH LABORATORY SCHOOL OF NATURAL SCIENCE, UNIVERSITY OF WESTERN SYDNEY HAWKESBURY CAMPUS, RICHMOND, NEW SOUTH WALES AUSTRALIA
JULY 2006
© IMELIDA C. GENSON 2006 ACKNOWLEDGEMENTS
I wish to express my deepest gratitude to individuals and groups for taking part in the success of this undertaking:
To the Australian Center for International Agricultural Research (ACIAR) through the John Allwright Fellowship for the financial support extended;
To my government through the Bureau of Soils and Water Management (BSWM) for the official leave of absence permitted;
To my supervisors, Associate Professor John Bavor, Dr. Anthony Haigh and Dr. Berthold Rembertus Hennecke for their support and guidance;
Special thanks and recognition to the efforts of Dr. Willie Joshua for his help and the fruitful discussions during the writing of this thesis;
Sincere appreciation to Bronwyn Davies for taking time to edit this report;
And to the field staff of the Watershed Project Bohol, Philippines for providing the data.
DEDICATION
This piece of work is dedicated to my family: Papa & Mama and to my eight brothers and three sisters.
Also, I wish to dedicate this work to my friend Prasan Sharp for her unselfish support and encouragement. And, to the Carty Family (Michael, Vicky and Phillip) who adopted me into their family. Thanks for providing a warmth environment of being homed away from home.
For all these blessings, to God be the glory.
STATEMENT OF AUTHENTICATION
This thesis contains no material which has been accepted for the award of any other degree or diploma in any university or institution and, to the best of the author’s knowledge and belief, contains no material previously written or published by another author except when due reference is made in the text.
______
Imelida C. Genson
Table of Contents
LIST OF TABLES ……………….. iv
LIST OF FIGURES ……………….. vi
LIST OF ACRONYMS AND ABBREVIATIONS ……………….. viii
ABSTRACT ……………….. ix
CHAPTER 1. INTRODUCTION ……………….. 1 1.1 Introduction ……………….. 1 1.2 The Philippine uplands ……………….. 1 1.3 The Bohol Watershed Project ……………….. 2 1.4 Aims ……………….. 5 CHAPTER 2. LITERATURE REVIEW 2.1 The integrated watershed approach of resources management ……………….. 6 2.2 Erosion processes ……………….. 8 2.3 Factors of erosion modeling ……………….. 9 2.3.1 Rainfall ……………….. 10 2.3.2 Soil properties ……………….. 11 2.3.3 Surface cover and cropping practices ……………….. 12 2.3.4 Topography ……………….. 13 2.4 Erosion modeling ……………….. 13 2.4.1 USLE model ……………….. 15 2.4.2 WEPP model ……………….. 16 2.4.2.1 Hillslope erosion component ……………….. 17 2.4.2.2 Hillslope surface hydrology ……………….. 19 2.4.2.3 Water balance and percolation ……………….. 19 2.4.2.4 Subsurface hydrology ……………….. 19 2.4.2.5 Soil component ……………….. 19 2.4.2.6 Plant growth component ……………….. 19 2.4.2.7 Climate component ……………….. 20 2.4.2.8 Residue decomposition ……………….. 20 2.5 Estimates of erosion at a plot and watershed scale ……………….. 20 2.6 Erosion studies in the Philippines ……………….. 21 2.6.1 Plot scale ……………….. 21 2.6.2 Watershed scale ……………….. 23 2.7 Erosion models and GIS ……………….. 24
i 2.7 GIS application in watershed management ……………….. 26
CHAPTER 3. MATERIALS AND METHODS 3.1 The study area ……………….. 28 3.2 On-site assessment of soil loss and runoff ……………….. 29 3.2.1 The runoff plots ……………….. 30 3.2.1.1 Agroforestry ……………….. 32 3.2.1.2 Cassava/corn ……………….. 32 3.2.1.3 Forest ……………….. 32 3.2.1.4 Grassland ……………….. 33 3.2.1.5 Oil palm ……………….. 33 3.2.2 Watershed measurement of flow ……………….. 33 3.2.3 Climate ……………….. 34 3.2.3.1 Rainfall erosivity ……………….. 35 3.3 Computer simulations ……………….. 35 3.3.1 Inputs for WEPP and GeoWEPP ……………….. 35 3.3.1.1 Climate file ……………….. 35 3.3.1.2 Slope file ……………….. 36 3.3.1.3 Soils input file and soil map ……………….. 37 3.3.1.4 Crop management file and land cover map ……………….. 37 3.3.2 Scenarios ……………….. 38 3.3.2.1 Hillslope application ……………….. 38 3.3.2.2 Watershed application ……………….. 40
CHAPTER 4. RESULTS AND DISCUSSION 4.1 On-site measurements ……………….. 42 4.1.1 Experimental runoff plots and rain gauges ……………….. 42 4.1.1.1 Weekly rainfall from the local rain gauges ……………….. 42 4.1.1.2 Weekly rainfall >60 mm ……………….. 44 Weekly rainfall-runoff from the experimental 4.1.1.3 runoff plots ……………….. 44 Weekly rainfall-soil loss from the experimental 4.1.1.4 runoff plots ……………….. 47 4.1.1.5 Totall rainfall, runoff and soil loss ……………….. 49 4.1.1.6 High erosion week ……………….. 51 4.1.2 Rainfall erosivity analysis ……………….. 52 4.1.2.1 Rainfall erosivity and soil loss ……………….. 54 4.1.3 Watershed measurement of flow ……………….. 56 4.1.3.1 Land cover and slope ……………….. 57 4.1.3.2 Rainfall and discharge curves ……………….. 58
ii 4.1.3.3 Suspended sediment concentration ……………….. 59 4.1.4 Summary of the on-site measurement of erosion ……………….. 61 4.2 Application of WEPP and GeoWEPP erosion models ……………….. 61 4.2.1 Erosion assessment at farm level ……………….. 62 4.2.1.1 Increasing slope ……………….. 63 4.2.1.2 Effects of terracing ……………….. 63 4.2.1.3 Additional terrace ……………….. 63 4.2.1.4 Grass strips ……………….. 64 4.2.1.5 Soil loss graph and deposition points ……………….. 64 4.2.1.6 Multiple flow simulations ……………….. 68 4.2.2 Erosion hazard assessment in the Bugsok Subwatershed ……………….. 70 4.2.2.1 Existing land use conditions ……………….. 72 4.2.2.2 Agriculture ≤ 18% slope> forest ……………….. 73 4.2.2.3 Application to land use planning ……………….. 75 CHAPTER 5. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 5.1 Summary ……………….. 76 5.2 Overall Conclusions ……………….. 81 5.3 Recommendations ……………….. 82
REFERENCES ……………….. 84
APPENDICES ……………….. 90
iii
LIST OF TABLES Some erosion rates at plot level extracted from Table 2.1 ………….. 22 different studies conducted in the Philippines
Selected watersheds and sediment yield in the island Table 2.2 ………….. 23 of Mindoro, 1984 (David 1988)
Landcover types and the area of the runoff plots used in the Table 3.1 field monitoring of soil loss and runoff. These land cover ………….. 30 types represent the major vegetation in the area.
Crop parameters and management files taken from WEPP Table 3.2 and GeoWEPP databases (USDA-ARS, 1995) and used in ………….. 38 simulations
Scenarios for the hillslope application of WEPP on steep Table 3.3 ………….. 39 slopes cropped with corn.
Land use scenarios for WEPP watershed application using Table 3.4 ………….. 41 the Bugsok AWaS catchment area.
Percentage contribution of soil loss above 60-mm weekly Table 4.1 rainfall to total soil loss from each experimental runoff ………….. 44 plot.
Week with highest soil loss contribution to the total soil Table 4.2 loss collected within 98 weeks including the respective ………….. 51 runoff and rainfall amounts.
Land cover distribution of the Bugsok and Pamacsalan Table 4.3 ………….. 57 Subwatersheds based on March 2002 Landsat-7 ETM+.
Slope distribution within Bugsok and Pamacsalan Table 4.4 Subwatersheds derived using ArcGIS 9 from a 30-m DEM ………….. 58 and classified according to BSWM cclassification criteria
WEPP simulation of soil, sediment and runoff for non- Table 4.5 ………….. 65 terraced and terraced conditions under different slopes
Percent decrease in soil loss, sediment yield and runoff Table 4.6 resulting from the use of one and two 1-m terraces relative ………….. 66 to a no-terrace condition
Percent decrease in soil loss, sediment yield and runoff Table 4.7 when terraces were replaced with grass strips relative to ………….. 67 no-terrace conditions.
Location of starting points of deposition and soil loss from Table 4.8 ………….. 69 the topmost part of the hillslope as determined by WEPP
iv simulation
On-site effects of land use change predicted by WEPP- Table 4.9 GeoWEPP and presented as percentage distribution of soil ………….. 70 loss under different land cover scenarios
Simulation results of Scenario A representing the area (in percent) occupied by each land cover type classified under Table 4.10 ………….. 72 tolerable and non-tolerable soil loss rates and further classified using the 18% slope criteria.
Off-site effects of land use changes predicted by WEPP- Table 4.11 ………….. 74 GeoWEPP model under each land use scenario
v
LIST OF FIGURES Location of the Bohol Watershed Project. Inset is the Figure 1.1 ………….. 3 map of the Philippines and the location of Bohol Island
The conceptual components of the Water Erosion Figure 2.1 ………….. 17 Prediction (WEPP) erosion model
Figure 3.1 Geographic location of the Upper Inabanga Watershed ………….. 29
Location of seven runoff plots, two Automatic Water Samplers (AWaS), two Automatic Weather Stations Figure 3.2 ………….. 30 (AWeS) and three rain gauges in the Upper Inabanga Watershed.
Field set up of runoff plots for monitoring soil loss and runoff under the forest land cover type. The galvanized Figure 3.3 ………….. 31 iron sheeting at the bottom part of the plots is secured on concrete lining.
Location of an Automatic Water Sampler (AWaS) along a stream. The pipeline provides a guide and an anchorage for the bubbler tube that was placed at the Figure 3.4 ………….. 34 bottom of the stream. The instrument was placed inside a metal casing and secured in one place on a concrete foundation.
Modification of hillslope topography to effect terracing: Figure 3.5 (a) no terrace (b) one terrace at the bottom of hillslope, ………….. 39 and (c) two terraces
Schematic diagram of overland flow elements for the Figure 3.6 ………….. 40 grass strips simulation
Weekly rainfall from the three localized rain gauges Figure 4.1 ………….. 43 during the 98-week data collection
Weekly rainfall and runoff from the agroforest, cassava/corn, forest, grassland and oil palm Figure 4.2 experimental runoff plots. Right-hand charts show data ………….. 45 points for rainfall below 60 mm and runoff below 0.5 mm.
Comparison of weekly rainfall and soil loss from the agroforest, cassava/corn, forest, grassland and oil palm plots. Right-hand chart shows data points at weekly Figure 4.3 ………….. 48 rainfall below 60 mm and weekly soil loss below 0.005 t·ha-1
Total rainfall, runoff and soil loss accumulated during the 98-week on-site monitoring. The values were Figure 4.4 ………….. 50 computed from weekly data.
vi Weekly rainfall and erosivity relationship computed Figure 4.5 from the 5-min data of Bugsok AWeS. ………….. 53
Weekly rainfall and erosivity relationship computed Figure 4.6 ………….. 53 from the 5-min rainfall data Pamacsalan AWeS
Weekly soil loss and rainfall erosivity for agroforest, Figure 4.7 ………….. 55 cassava/corn, forest, grassland and oil palm plots.
Catchment areas of Bugsok AWaS and Pamacsalan Figure 4.8 AWaS delineated using a DEM and their respective ………….. 57 coordinates.
Average daily discharge records from automatic water Figure 4.9 samplers and daily rainfall data from weather stations at ………….. 59 Bugsok and Pamacsalan.
Rainfall-discharge curves on December 2005 from Figure 4.10 ………….. 61 Bugsok and Pamacsalan monitoring sites.
Rainfall, discharge and sediment yield from Bugsok Figure 4.11 ………….. 62 AWaS recorded on July 22, 2004.
Rainfall, discharge rates and sediment yield from Figure 4.12 ………….. 63 Pamacsalan AWaS recorded on July 15-16, 2004.
Soil loss graph of a 50% slope (A) with 1-m width Figure 4.13 ………….. 67 grass strip at the bottom of the slope and (B) no terrace.
Trend of soil loss and deposition along a hillslope for three scenarios where the 2-grass strip condition has Figure 4.14 four OFEs; 1 grass strip condition has 2 OFEs and the ………….. 69 no grass strip condition has one OFE as determined in WEPP simulations.
Erosion maps of the Bugsok Subwatershed with six land use scenarios showing the on-site effects as predicted using GeoWEPP. White areas within the Figure 4.15 ………….. 71 subwatershed are the channels identified by GeoWEPP but were excluded in erosion simulation with the flowpath method.
vii LIST OF ACRONYMS AND ABBREVIATION
ACIAR Australian Center for International Agricultural Research
AWaS Automatic Water Sampler
AWeS Automatic Weather Station
BSWM Bureau of Soils and Water Management
DEM Digital Elevation Model
DENR Department of Environment and Natural Resources
FAO Food and Agriculture Organization
GIS Geographic Information System
ICRAF International Center for Research in Agroforestry
OFE Overland Flow Element
Philippine Council for Agriculture, Forestry and Natural Resources Research PCCARD and Development
SMU Soil Management Unit
TOPAZ Topographic Parameterization Program
UNCCD United Nations Convention to Combat Desertification
USDA- United States Department of Agriculture –Agricultural Research Service ARS
USLE Universal Soil Loss Equation
WEPP Water Erosion Prediction Project
viii ABSTRACT
To complement the Inabanga Watershed Project (BSWM, 2005), the study reported here was conducted to assess erosion and water resources degradation focused on the Upper Inabanga Watershed using the Water Erosion Prediction Project (WEPP) erosion model and geographic information system (GIS) tools. The study was divided into two sections. The first section was an assessment of the impact of land uses and farm management practices using five runoff experimental plots and two subwatersheds. A 98-week data set from the experimental plots was used to analyze runoff and soil loss linked to weekly rainfall data from localized rain gauges. Discharge rate data from water samplers and rain- intensity data from weather stations was used to characterize the subwatersheds in terms of runoff and sediment yield. The second section of the study was an application of the WEPP and GeoWEPP erosion models. Except for most of the crop management parameters, local climate, soil and topographic parameters were determined and used as inputs to run the model. The results of the study showed that cultivated areas for cassava/corn cropping generated high soil loss (42.5 t·ha-1·yr-1) compared to grassland, oil palm, agroforest and forest which were determined to have soil losses of 6.4 t·ha-1·yr-1, 2.6 t·ha-1·yr-1, 2.3 t·ha-1·yr-1 and 0.2 t·ha-1·yr-1, respectively. The major cause of high soil loss was attributed to farm soil management and cropping operations, which disturbed and exposed the soil surface to the impact of rainfall. The sediment concentrations from the Bugsok Subwatershed and Pamacsalan Subwatershed were as high as 201 mg·L-1 and 782 mg·L-1, respectively, during high rainfall events. These values indicated significant erosion taking place within the subwatersheds. Model simulations of the use of terraces on steep hillslopes (50-70% slopes) predicted reduced erosion by an average of 24% compared to a no-terrace hillslope. A further decrease in erosion was estimated when the terraces were replaced with grass strips. The WEPP-GeoWEPP watershed simulations predicted that any increase in agricultural areas increased on-site soil loss and sediment yield from the watershed.
ix Although constrained by limited input data especially with respect to crop parameters, the application of the model provided an initial step towards understanding erosion processes in the Upper Inabanga Watershed.
x
CHAPTER 1 INTRODUCTION
1.1 Introduction
Soil erosion is a most important environmental problem in the developing world (Ananda and Herath, 2003). Tropical soils are under particular threat as these soils are less stable than those in temperate climates because of their properties and climatic conditions (Steiner, 1996). Sloping uplands in Asia are particularly threatened by the serious problems of soil degradation (Lapar and Pandey, 1999) since much subsistence farming is carried out on these lands without soil conservation measures. At a global scale, land degradation may not pose a threat to food security but it does pose critical problems in areas where soils are fragile, property rights are insecure, and farmers have limited access to information and markets (Marcoux, 1996).
1.2 The Philippine uplands
Agricultural land degradation in the Philippines is a major environmental and development issue (Cramb, 2000a). FAO (2000), using the GLASOD (Global Assessment of Soil Degradation) database, estimate that 79 % of Philippine lands are threatened by severe degradation. Water erosion is a major driving force of land degradation in the Philippines. The process of land degradation is advancing at an alarming rate due to deforestation and inappropriate agricultural activities. In combination with fragile and highly sensitive mountainous environments, cyclones and frequent thunderstorms result in very high sediment yield rates throughout the country (White, 1995). Erosion rates in Philippines have been estimated at values exceeding tolerable soil loss rates. Soil loss rates of 10 tonnes per hectare per year was considered tolerable by Paningbatan (Paningbatan, 1987 as cited by PCARRD, 1991). In their
1 Philippine sites, the Management of Soil Erosion Consortium (MSEC) research program recorded soil loss of up to 54 tonnes per hectare per year (IWMI, 2002), David (1988) also reported that sediment discharges of Philippine rivers, in which catchments are subject to uncontrolled manipulation, exceed 30 tonnes per hectare per year. For instance, sheet erosion loadings to Magat Dam, in the Northern Philippines, was estimated to be in the order of 88 tonnes per hectare per year (Cruz et al., 1988). Adverse effects of soil erosion from upper watershed regions can be translated in terms of decreasing productivity and income of farmers in downstream areas. A specific case is presented by Lantican et al. (2003). In their study, soil erosion in the Upper Manupali Watershed in Northern Mindanao, Philippines, caused heavy siltation of irrigation canals consequently reducing rice yields of the affected areas by 27%. Excessive siltation decreases the volume of water for delivery to the service areas. As a result, water delivery schedules change, pressing affected farmers to change cropping patterns and/or shift from rice-based to vegetable farming. In order to effectively convey irrigation water to the service areas, silt deposits have to be removed. Removal of silt deposits incurs additional cost on top of regular irrigation management operation and maintenance costs. Several soil-water conservation technologies have been developed and tested (e.g. contour cropping, hedgerows) to address land degradation issues in the Philippines. Contour farming, for instance, has proven appropriate for Philippine uplands (Cramb, 2000b) based on a socio-economic evaluation of soil conservation technologies. Conservation-farming projects, such as ISFP (Integrated Social Forestry Program), have been implemented however with little success. Factors limiting adoption have been identified as the attributes of the technology itself and a range of social, economic and institutional environments where the technologies are promoted (Cramb, 2000b). Promising initiatives such as the Landcare Programs undertaken by ICRAF (International Center for Research in Agroforestry) and other groups are hoped to be more successful (ACIAR, 2004)
1.3 The Bohol Watershed Project
Bohol Island is the 10th largest island in the Philippines situated in the Central Visayas Region. Bohol Island is geographically located within 123o40’- 124o40’
2 longitudes and 9o30’ – 10o17’ latitudes, approximately 625 km southeast of Manila and 75 km east of Cebu (Figure 1.1).
Figure 1. 1. Location of the Bohol Watershed Project. Inset is the map of the Philippines and the location of Bohol Island.
The island has a land area of 411,700 hectares and is home to 1.14 million people. The population is growing at a high rate of 2.95% per annum compared to the national average of 2.38% based on 1995-2000 census data (NSO, 2002). Bohol is an agricultural province where 45% of the land area is cultivated in agricultural pursuits. The island is considered the leading food granary of the Central Visayas Region. Farming is the main source of income followed by fishing with seaweed farming (AusAid, 2001). PCARRD (1984), however, reported that more than half of the island land area is already eroded. Statistics also show that there is a higher incidence of poverty in the island reported at 47.3% compared to the national and regional poverty incidence of 28.4% and 32.3%, respectively (NSO, 2002). Similarly, an Australian
3 AID report (AusAID, 2001) identified widespread poverty especially in the upper catchments and small island coastal zones. Thus, assistance has been recommended to focus on these areas. In response to the needs identified by the Philippine Bureau of Soils and Water Management (BSWM), the University of Western Sydney (UWS) and the Australian Center for International Agricultural Research (ACIAR), a Watershed Project (ACIAR Project LWR1/2001/2003) was established to address the agricultural opportunities and natural resources problems in Bohol Island. The project’s major task was to inventory resources and develop strategies for protecting the environmentally and economically sensitive soil and water resources of the Inabanga River Watershed while maintaining agricultural productivity within the watershed. An integrated watershed approach of resource management was adopted in the project. The approach considered the elements within the watershed including social, economic, political, and environmental aspects in the sustainable management of land and water resources while producing goods and services for human consumption (Cruz, 1999). The approach is founded on the concept that watersheds are formed by natural landmasses and that water flows into a common point. In effect, pollutants and sediments carried by water end up in water bodies. Dealing with these problems as a whole is considered efficient in terms of data collection, monitoring and management rather than doing it in a piecemeal fashion (DeBarry, 2004). The ACIAR Watershed Project has taken advantage of GIS (Geographic Information Systems) techniques to develop a database of the current situation of natural resources and in identifying potential problem areas within the watershed. Field erosion plots provided an estimate of the extent of soil erosion under different land uses while actual surface water monitoring supplied an estimate of runoff generation and sediment yield from selected drainage areas. The project also involved the stakeholders and conducted socio-economic surveys to identify constraints and policy issues affecting soil and water resources use within the watershed. Based on the findings of the project, a number of management options have been recommended. In this project, a range of these options are evaluated using process based erosion model, WEPP (Water Erosion Prediction Project), and GIS
4 techniques. This report is on the process of assessing scenarios of land use and water resources management using an erosion model and GIS techniques. Soil erosion models and GIS are indispensable tools in erosion studies. Erosion models are predictive tools for evaluating the effectiveness of different management methods for conservation planning, project planning, and soil erosion inventories and for regulation (Nearing et al., 1994). Specifically, a process based erosion model predicts where and when erosion is occurring thus helping conservation planners target efforts to reduce erosion. GIS facilitates the collection, storage of data, and construction of model inputs as well as provision of display and analysis tools and presentation of the model outputs. Together with GIS techniques, erosion models provide better means of understanding erosion processes. The application of a process-based erosion model had not previously been carried out in the Upper Inabanga Watershed, and was considered a valuable planning tool for planners, thus the research was conducted.
1.4 Aims
The aims of the project are: 1. To describe the impact of land use management practices in terms of soil loss, runoff and sediment yield in both runoff plots and watershed basis 2. To apply the WEPP erosion model and its geo-spatial interface GeoWEPP a. To predict and simulate soil loss, runoff and sediment yield from agricultural hillslopes incorporating soil conservation measures b. To simulate and predict the effect of land cover change in a selected catchment in terms of runoff and sediment yield
The study is an initial step into understanding erosion processes within the Upper Inabanga Watershed by using an erosion model and GIS. The Upper Inabanga Watershed is an important catchment in Bohol Island since it is the drainage area of the major irrigation facility that supports rice-based farming of the downstream areas.
5
CHAPTER 2 LITERATURE REVIEW
The following key components of the research program are reviewed:
• Integrated watershed approach of soil and water resources management.
• Processes of soil erosion • Plot-scale and watershed-scale approaches to erosion assessment • Erosion models and GIS techniques • Application of erosion models and GIS A clear understanding of the factors affecting the erosion processes in the Upper Inabanga Watershed was considered as critical for development of relevant input into the modelling process in order to provide a viable tool for resource management and conservation planning.
2.1 The integrated watershed approach of resources management
Major causes of land degradation in Asia include deforestation, shortage of land due to increased populations, poor land use, insecure land tenure, inappropriate land management practices and poverty (FAO, 1995). The integrated watershed approach of managing resources is envisaged as an effective approach and is currently applied in Bohol Watershed Project. A review of the fundamental concepts of this approach is presented. The watershed approach of land and water resources management is based on the understanding that quality and quantity at a point on a stream reflects the characteristics of the upslope area (Davenport, 2003). A watershed is a landscape wherein rainwater is collected and drained at one point called the watershed outlet (Cruz, 1999). It is a dynamic system separated from other watersheds by a high ground perimeter that forms the boundary or watershed divide (Pereira, 1989).
6 The watershed is a major source of nutrients and pollutants (Davenport, 2003; Cruz et al., 1999). As water moves over the land surface, nutrients and sediments are carried with runoff and are deposited in lakes, coastal areas, lowland plains and rivers. These foreign materials can, in many circumstances, have more adverse effects than positive inputs to the new location. For instance, sedimentation of reservoirs is detrimental to the efficiency of the system while enrichment of a water body with nutrients causes eutrophication (Morgan, 2005;Davenport, 2003). Rates of land degradation are strongly influenced by the decisions of upland farmers (Coxhead and Shively, 2005). Disturbances in the upper areas of a watershed are translated into the downstream areas through the hydrologic process (Pereira, 1989). The adverse effects of the upper watershed disturbance directly affect the lives and property of the downstream community, as in the case presented by Lantican et al. (2003). The effect of floods, drought and sedimentation is a concern not just for the upper areas but also of the whole watershed (Pereira, 1989). The ecological linkage between the upstream land uses and the downstream water condition is a strong justification for an integrated watershed-based approach to resource management (Francisco, 2002) A policy note by Francisco (2002) pointed out three important points of undertaking a watershed as a planning unit for soil and water resources management in the Philippines. First, the watershed approach makes it possible to identify various sources of stressors at a point where the watershed drains. Second, stakeholders who have common concerns about the watershed are easily identified and organized. Lastly, various interventions are better implemented and monitored in a well-defined ecological unit such as the watershed. A range of on-site and off-site economic, social and environmental benefits can be derived from a sustainable watershed management as given in Cruz (1999). The economic benefits include production of agricultural crops, forest products for timber and water supply for domestic or industrial uses among others. Improved watershed management also provides social benefits such as reduction of risk to life and property brought about by natural disasters. The environmental benefits of sustainable watershed management include preservation of biodiversity, water and soil conservation and microclimate amelioration. In contrast, failure in watershed management could further exacerbate watershed degradation (Cruz, 1999), hence, reducing economic, social and environmental benefits.
7 Davenport (2003) recommended a watershed management model. This model consists of four phases: assessment phase, planning phase, implementation phase and evaluation phase. The assessment phase consists of a careful analysis of the land and water resources and a definition of issues, problem sources and critical areas within the study area. The assessment phase is the focus of this current review. There are a number of causes of watershed degradation (Cruz, 1999) and soil erosion process is one of the important contributors. To understand these processes, the following section reviews on the basic principles of soil erosion by water.
2.2 Erosion processes
Erosion is a natural process of soil formation. Erosion may either be geologically or human-induced, or a mix of both. Geologic erosion occurs without human interference while human-induced erosion is an accelerated erosion caused by land disturbance for crop production (Lal, 1994). Soil erosion is one form of land degradation characterized by the change in quality of soil, water and other characteristics that reduce the ability of land to produce goods and services that are valued by humans (UNCCD, 1994). The process of erosion by water can be described in three stages: detachment, transport and deposition (Hudson, 1995; Merritt et al., 2003; Morgan, 2005). In the first stage, soil particles are detached by the impact of raindrops or shear forces of flowing water. During the second stage or the sediment transport stage, sediments are moved downslope, which is caused by the splash action of raindrops and runoff. The third stage occurs when runoff velocity is reduced and load carrying capacity decreases causing some or all the sediments to be deposited. There are limiting conditions for erosion process (Hudson, 1995). The erosion process can be described as transport-limited erosion and detachment-limited erosion. Transport-limited erosion occurs when there is not enough runoff to carry away the soil particles detached by rainfall impact while detachment-limited erosion occurs when there is enough runoff to carry more soil than is actually being detached by rainfall impact (Hudson, 1995; Morgan, 2005). There are different types of erosion by water: sheet erosion, rill erosion, interill erosion, gully erosion and stream channel erosion (Schwab et al., 1992). Sheet erosion is the removal of a thin sheet of soil by overland flow (Schwab et al.,
8 1992) on hillsides. Overland flow occurs either when soil moisture capacity or infiltration rate of the soil is exceeded (Morgan, 2005). Rill erosion occurs when flows start to concentrate and create defined flow paths (Schwab et al., 1992). The detachment rate from the rills is a function of hydraulic shear stress of the flowing water, and the rill erodibility and critical shear of the soil, expressed as:
⎛ Qs ⎞ Dr = K r τ −τ c ⎜1− ⎟ ()⎜ T ⎟ ⎝ c ⎠ -2 -1 where: Dr is rill detachment rate in kg·m ·s -1 Kr is rill erodibility resulting from shear stress in s·m