A Geographically-Based Land Use Suitability Assessment andLand Capability Classification

Item type text; Proceedings

Authors Cruz, Rex Victor O.; Ffolliott, Peter F.

Publisher Arizona-Nevada Academy of Science

Journal Hydrology and Water Resources in Arizona and theSouthwest

Rights Copyright ©, where appropriate, is held by the author.

Downloaded 7-Feb-2018 06:26:22

Link to item http://hdl.handle.net/10150/296434

A GEOGRAPHICALLY -BASED LAND USE SUITABILITYASSESSMENT AND LAND CAPABILITY CLASSIFICATION

Rex Victor O. Cruz and Peter F. FfolliottSchool of Renewable Natural Resources

University of ArizonaTucson, Arizona 85721

Introduction

Land capability classification generally refers to thedescription and classification of lands relative to theirbiophysical features and ability to sustain various kinds ofuses. The USDA Soil Conservation Service land capabilityclassification guide is perhaps the most popular system everdeveloped so far (Klingebiel and Montgomery, 1961; andBrakensiek, et al., 1979). It has been modified and used inother countries including Israel, the Philippines, andZimbabwe (Hudson, 1981). In spite of its popularity, however,the USDA guide is based only on agronomic land uses and isqualitative in nature.

Classification systems developed in recent years aremore quantitative in nature as the understanding of therelationships among the different factors influencing soilproductivity and stability increases. For example, Larson etal. (1988) and Warren et al. (1989) developed classificationsystems based on estimated measures of productivity, soilresistivity, and soil erosion, respectively.

Land use suitability assessment is defined here as themeasurement and rating of the impacts of a land use on theproductivity and stability of an area. The impacts could bemeasured in terms of actual volume of production, decline insoil fertility, and the amount of soil erosion.

It is the purpose of this paper to describe amethodology for land use suitability assessment and landcapability classification based on estimated soil loss usingthe modified universal soil loss equation or MUSLE (Williams,1975), with the aid of a geographic information system (GIS).

The GIS

Soil erosion is influenced by many factors which aregeographic in nature, such as topography, soil properties,vegetation, and land use. As such, the use of a GIS wouldenhance the accuracy of soil erosion estimation. By dividingan area into smaller cells, the significance of geographic

51

characteristics which would otherwise be diminished as aresult of parameter lumping when the area is treated as onewho'e unit is preserved.

Through the years, GIS has been defined in several ways(Burrough, 1986; Berry, 1986; Parker, 1988). It has beendefined as a set of tools, a technology, and an automatedspatial information system capable of processing spatial orgeographically- referenced information. Most GIS are capableof data input, data storage and retrieval, data manipulationand analysis, and data reporting.

Essentially, GIS can be applied to any activity wherespatial considerations are important, and where largequantities of data need to be processed and reprocessed over anumber of times. It has found many uses in natural resourcesmanagement, economics and marketing, regional and urbanplanning, and engineering.

Works by Berry and Sailor (1987), Gilliland andPotter(1987), Vasconcelos (1988), and Warren et al. (1989) areexamples of the many uses of different kinds of GIS. In thisstudy, Map Analysis Package (MAP) developed by Tomlin (1986)was used.

Study Area

The study area is located in the Ibulao Watershed,Philippines. It is a 65,000 -ha subwatershed of Magat RiverBasin which supports vast areas of agricultural lands.Existing land uses in Ibulao includes, forestry, grazing, andagriculture. The climate generally is humid, with an averageannual rainfall of 2,200 mm. Soils are mostly clay loam, withtopography ranging from flat to very steep and rugged terrain.

Assessment Method

There are four basic components to the methodologydescribed in this paper, namely MUSLE, capabilityclassification, land use suitability assessment, and GIS(Figure 1). Other components, RAINGEN, CREAMS, and IRSX areincidental to MUSLE for the estimation of the runoff factor.

The study area generally was divided into 10 -ha cells.For the entire watershed, different data overlays, such asslope map, soils map, elevation map, and land use map, wereprepared using MAP. From the data overlays, the features andparameters needed in the estimation of soil erosion,capability classification, and land use suitability

52

RAINGEN- organize and analyzerainfall data

- simulate rainfallduration

CREAMS- perform water budgetanalysis

- compute for the soilmoisture

IRSX- simulate surface runoffvolume and the peakrunoff rate from eachcell

MUSLE- estimate the soil erosion

from each cell (for eachstorm event, annual)

CAPABILITY CLASSIFICATION- calculate the erosion

index for each cell- classify cells intocapability classesbased on soil erosionindex

iSUITABILITY ASSESSMENT- evaluate the suitabilityof each landuse based oncurrent legal land clas-sification, LCCG of thePhilippine Bureau ofSoils, and soil erosion

GIS Operations

MAP- organize and createsource data basemaps

- create derived maps(soils, slope, eleva-tion, landuse maps)classify the Ibulaowatershed into cellsof homogenous soil,slope, and landusefeatures

- estimate soil erosionfrom each cell

- create output over-lays (surface runoffand soil erosion maps)evaluate suitabilityof landuse for eachcell based on presentlegal land classifica-tion, LCCG, and soilerosioncreate suitabililtymaps based on presentlegal land classifica-tion, LCCG, and soilerosion

- create erodibility mapbased on soil erosionindex of each cellcreate capability mapbased on soil erodi-bility map

IDRISI- refine and organizeoutput overlays

- print out copies ofsource maps, derivedmaps, and outputmaps

Figure 1. Schematic Representation of the Generation and Flowof Information.

53

assessment for each cell were extracted. In turn, the resultsfrom these components were passed back to MAP for processingthe various output overlays.

MUSLE

The general form of the MUSLE is shown as:

E = 11.8 (Q *q)0.56 K L S C P

where E is soil loss (Mg /ha) estimated as the product ofrunoff factor consisting of surface runoff depth (Q) and peakflow (q), soil erodibility factor (K), topography factor (LS),soil cover factor (C), and conservation practice factor (P).

The factors KLSCP were estimated using the proceduresdescribed by Williams and Berndt (1972), Wischmeier and Smith(1978), Dissmeyer and Foster (1980), and David (1985). Adetailed description of the parameter estimation is presentedby Cruz (1990).

The surface runoff and peak flow for each cell weresimulated using IRSX, which is a modification of theinfiltration- kinematic routing program (IRS9) developed by theUSDA -ARS (Stone and Shirley, 1985). Most of the parametersused by IRSX, such as soil porosity, hydraulic conductivity,depth- discharge coefficient, and soil moisture, also aregeographic in nature.

The soil moisture for each cell was estimated by thehydrology component of CREAMS, a field -scale model forchemical, Runoff, and erosion from Agricultural Managementaystems model (Knise1,1980). Likewise, most of the soilrelated parameters used in CREAMS simulation are geographic innature.

Land Capability Classification

Land capability classification is based on soil erosionindex (I). For each cell, the erosion index was estimated by:

I = [11.8 (Q*g)056 KLS] / T

where T is the soil loss tolerance which represents the amountof annual soil erosion that can be sustained by an areawithout jeopardizing its long term productivity. (The valueof T usually ranges between 2.2 and 11.2 t /ha depending upon

54

the locally intrinsic rate of soil formation and soil depth;in this case, T was assumed to be 20 t /ha /year).

The capability class of a cell was identified using theestimated average annual erosion index (Table 1 and Figure 2).The different land use recommendations were determined bysolving for the maximum CP value (using the erosion indexequation) for a given capability that would yield an annualsoil erosion value not greater than the tolerance limit.

Land Use Suitability Assessment

Land use suitability in each cell was evaluated on thebasis of the estimated annual soil erosion. A land use in acell was rated suitable if the average annual soil erosion isless than or equal to the tolerance limit. Otherwise, landuse was rated unsuitable (Table 2 and Figure 3).

Applications

The land use suitability assessment and land capabilityclassification procedure described could be useful inidentifying the land use most suitable to a given area. Italso could provide a method of examining the status ofexisting land uses as far as impacts on soil productivity isconcerned. It could be used to identify which existing landuses need to be changed or if they cannot be changed, the toolcould help identify what kind of measures need to be taken tomitigate the adverse impacts of such land uses. Finally, themethod would be instrumental in determining the differentintensities of land uses that would best complement thecapability of an area.

Recommendations

An erosion -based land capability classification andland use suitability assessment is only as useful as the soilerosion estimation component. Therefore, improvements in theexisting tools for soil erosion estimation or development anduse of methods applicable to the area of concern always shouldbe considered.

The accuracy of results obtained from the methoddescribed in this paper also is largely a function of theaccuracy of the different input or source maps. In overlaying

55

Table 1. Erosion -Based Land Capability Classification ofIbulao Watershed.

Capabilityclass

Erosionindex

Area(10ha)

Maximumallowable

CP

1 0 -2 286 0.500

2 3 -5 146 0.200

3 6 -10 311 0.100

4 11 -20 1069 0.050

5 21 -30 408 0.030

6 31 -50 2176 0.020

7 > 50 2052 0.010

Recommendedlanduse

unrestricted

grassland,riceterrace /paddy,agroforestry

grassland,riceterrace /paddy,limited agro-forestry

rice terraces,limited agro-forestry withconservationmeasures, forestry

rice terrace,forestry

limited riceterrace, forestry

limited to forestry

,,,,`.,..,

class 1

class 2

class 3

INM

MIND

class 4 -

LEGEND

class 5

class 6

class 7 -

;,.'';..,,:,.ii i:

Figure 2, Erosion -Based Land Capability Classificationof Ibulao Watershed.

57

Table 2. Erosion Classes Coverage Under Different LandUses in Ibulao Watershed. (The values representthe number of 10 -ha cells).

Present Landuse

1 2

Erosion

3 4

Class

5 6 7 8 TOTAL

Forest

mossy 178 178

closed canopy 264 264

open canopy 2289 2289

Rice padd /terr 1177 320 1497

Open grass 68 234 62*

112* 116* 103* 583* 1278

Diverse crop 49 7 2 12* 126* 260* 251* 707

Nonvegetated 41 16 11* 14* 2* 11* 95

Residential 8 1* 4* 26* 39

Mainstream 45 5 4 2 10 35 101

TOTAL 4119 582 17 75 116 258 375 906 6448

* Rated unsuitable based on the average annual soil erosionloss.

LEGEND

suitable - unsuitable

Figure 3. Landuse Suitability Map of Ibulao WatershedBased on Soil Erosion.

59

several maps, format considerations, such as map resolutionand projection, should be appropriately defined to keep thejoint probability of coincidence the same at all locations.

The success of using geographically -based tools forland use suitability assessment and land capabilityclassification also would depend upon the adequacy ofdefinition and quantification of the different spatialrelationships involved. For example, the relationship betweenneighboring cells as far as surface runoff generated from eachcell should be clearly defined and represented to obtainsatisfactory prediction or simulation results.

References Cited

Berry, J.K. 1986. Learning computer assisted map analysis.J of Forestry 84(10):39 -43.

Berry, J.K., and J.K. Sailor. 1987. Use of geographicinformation system for storm runoff prediction fromsmall urban watersheds. Environ. Management11(1):21 -27.

Brakensiek, D.L.; H.B. Osborn; and W.J. Rawls. 1979. Fieldmanual for research in agricultural hydrology.USDA -SEA, Washington D.C.

Burrough, P.A. 1986. Principles of geographical systems forland resources assessment. Clarendon Press, Oxford.

Cruz, R.V.O. 1990. Landuse suitability assessment and landcapability classification in Ibulao watershed,Philippines. Ph D Dissertation, University of Arizona,Tucson, Arizona.

David, W.P. 1985. Erosion and sediment transport. UplandResource Policy Program. Technical Report Series No.87 -01. Philippine Institute for Development Studies.Quezon City, Philippines.

Dangermond, J., and C. Killpack. 1988. Mapping for thefuture. Landscape Architecture. July /August:46 -51.Dissmeyer, G.E., and G.R. Foster. 1984. A guide forpredicting sheet and rill erosion on forest land.Technical Publication R8 -TP 6. USDA -FS SouthernRegion. Atlanta, GA.

60

Foster, G.R., L.J. Lane, J.D. Nowlin, J.M. Laflen, andR.A. Young. 1980. A model to estimate sediment yieldfrom field -sized areas: Development of model. InCREAMS: A field scale model for chemicals, runoff,and erosion from agricultural management systems.Vol 1. Model Documentation, Ch 3. USDA Conser.Res. Report No.26.

Gilliland, M.W. and W. Potter. 1986. A geographic informationsystem to predict non -point source pollutionpotential. Water Resources Bulletin 23(2):281 -291.

Hudson, N. 1981. Soil conservation. Batsford Academic andEducation, Ltd., London.

Klingebiel, A.A., and P.H. Montgomery.1961. Land capabilityclassification. USDA Agricultural Handbook 210.

Knisel, W.G., Ed. 1980. CREAMS: a field -scale model forchemicals, runoff, and erosion from agriculturalmanagement systems. USDA Conservation Research ReportNo. 26.

Larson, G.A., G. Roloff, and W.E. Larson. 1988. A newapproach to marginal agricultural landclassification. J of Soil and Water Conservation.43:103 -106.

Marble, D.F., and S.E. Amundson. 1988. Microcomputer -basedgeographic information systems and their role inurban and regional planning. Environmental andPlanning B: Planning and Design. 15:305 -324.

Mein, R.G., and C.L. Larson. 1973. Modeling infiltrationduring a steady rain. Water Resources Res. 9:384 -394.

Parker, D.H. 1988. The unique qualities of geographicinformation system: A commentary. PhotogrammetricEng. and Remote Sensing. 54:1547 -1549.

Smith, D.D., and W.H. Wischmeier. 1957. Factors affectingsheet and rill erosion. Trans. Amer. Geophys. Union.38:889 -896.

Stone, J.J.; L.J. Lane; and E.D. Shirley. 1989. ProgramIRS, model documentation. Personal communication.

Stone, J.J., and E.D. Shirley. 1985. Unpublished modeldocumentation of IRS9. USDA -ARS. Tucson, AZ.

61

Tomlin, D.C. 1986. The IBM personal computer version ofthe Map Analysis Package. Laboratory for ComputerGraphics and Spatial Analysis. Harvard GraduateSchool of Design, Harvard University.

Van Roessel, J.W. 1986. Guidelines for forestry informationprocessing with particular reference to developingcountries. Food and Agricultural Organization- UnitedNations. Rome.

Vasconcelos, M. 1988. Simulation of fire behavior with ageographic information system. Masters Thesis.SRNR, University of Arizona, Tucson, AZ.

Warren, S.D., V.E. Diersing, P.J. Thompson, and W.D. Goran.1989. An erosion -based land classification system formilitary installations. Environ. Management. 13:251 -257.

Welch, R., and M.M. Remillard. 1988. Remote sensing andgeographic information system techniques for aquaticresource evaluation. Photogrammetric Eng. and RemoteSensing. 54:177 -185.

Williams, J.R. 1975. Sediment yield predicted with universalequation using runoff energy factor. In: Present andProspective Technology for Predicting Sediment andResources. USDA- ARS -S- 40:244 -252.

Williams, J.R. and H.D. Berndt. 1972. Sediment yieldcomputed with universal equation. Proc. of the Amer.Soc. of Civil Eng. Journal of the Hydraulics Division.HY 12.

Wischmeier, W.H. 1976. Use and misuse of the universal soilloss equation. Jo. of Soil and Water Conservation.31:5 -9.

Wischmeier, W.H.; and D.D. Smith. 1978. Predicting rainfallerosion losses. Agric. Handbook 537. USDA - SEA -ARS.

62