7
Application of a RUSLE-based soil erosion modelling on Mauritius Island Rody Nigel A,C and Soonil D. D. V. Rughooputh B A Stagiaire Postdoctoral, Bureau 5310, INRS Centre Eau Terre Environnement, 490 rue de la Couronne, Québec, QC G1K 9A9, Canada. B Faculty of Science, University of Mauritius, Réduit, Mauritius. C Corresponding author. Email: [email protected] Abstract. Soil erosion by water is one of the most important natural resources management problems in the world. The damages it causes on-site are soil loss, breakdown of soil structure, and decline in organic matter content, nutrient content, fertility, and inltration rate. Lands with the highest erosion risk on Mauritius Island are crop cultivations (sugarcane, tea, vegetables) on erosion-susceptible terrain (slopes >20% coupled with highly erodible soils). The locations of such lands on Mauritius were mapped during previous, qualitatively based regional-scale erosion studies. In order to propose soil conservation strategies, there is a need to apply a more quantitative approach to supplement the previous, qualitatively based studies. This paper reports an application of the Revised Universal Soil Loss Equation (RUSLE) within a geographical information system in order to estimate soil loss on the island, and particularly for the high-erosion areas. Results show that total soil loss on the island is estimated at 298 259 t year 1 , with soil loss from high-erosion areas summing 84 780 t year 1 (28% of total soil loss). If all of the high-erosion areas were afforested, their soil loss would be reduced to 10 264 t year 1 , i.e. a reduction of 88% for the high-erosion areas and a reduction of 25% for the island. This study thus calls for soil and water conservation programs directed to these erosion-prone areas before the land degradation and environmental damage they are causing become irreversible. The methodological approach used in this work to quantitatively estimate soil loss from erosion-prone areas can be adopted in other countries as the basis for a nationwide erosion assessment in order to better inform environmental policy needs for soil and water conservation. Additional keywords: erosion risk mapping, GIS, priority action areas. Received 4 July 2012, accepted 19 December 2012, published online 31 January 2013 Introduction Soil erosion is an environmental problem that exacerbates on-site land degradation, while at the same time being a source of sediment and pollutants that adversely affect off-site aquatic ecosystems (Lal 2001). Mauritius Island is potentially at risk to soil erosion because of its rugged topography, extensive sugarcane cultivation, and tropical climate. This makes the land particularly vulnerable to erosion. In addition, other ecosystems can be damaged due to erosion, namely, estuaries, marshes, sh habitats, sea grass, and coral reefs. Conservation measures are needed to reduce the effects of soil erosion, and successful conservation programs require the concentration of resources on priority action areas. Soil-erosion risk assessment at a regional scale is commonly used to identify such conservation areas (e.g. Vrieling et al. 2002). The rst published study to map the erosion risk across the whole island was conducted by Nigel and Rughooputh (2010b), developing a soil-erosion risk-mapping model termed MauSERM (Mauritius Soil Erosion Risk Mapping). Four erosion factors, namely rainfall, land cover, topography, and soil, were considered during the erosion risk mapping using the MauSERM model. Investigations were also made on the best way to combine the factors, and the decision rule methodology was chosen, as it can be easily manipulated to mimic the complex interactions of all four factors (as opposed to factorial scoring). A second study was later conducted by running the model with new datasets, where slope gradient was replaced with slope length (inclusive of land parcels and drainage structure effects). The results illustrate that erosion sites occur commonly in cultivated areas with steep slopes and erosion has a summer- dominant pattern caused by intensive summer rainfall (Nigel and Rughooputh 2010a). This study was qualitatively based and enabled identication of priority action areas. Now, a more quantitative study is needed, as this can provide assistance in river basin management and to indicate the order of magnitude of soil erosion and sediment yield (de Vente and Poesen 2005). The aim of the present study is to compute soil loss estimates for Mauritius Island and especially for the priority action areas. The specic objectives are: (1) apply the Revised Universal Soil Loss Equation (RUSLE) model for the whole island, (2) compute soil loss estimates for the high-erosion areas (HEA) initially mapped, and (3) dene a second approach for mapping of HEA and derive estimates of RUSLE soil loss for these HEA. In this second approach, the HEA can be mapped Journal compilation Ó CSIRO 2012 www.publish.csiro.au/journals/sr CSIRO PUBLISHING Soil Research, 2012, 50, 645651 http://dx.doi.org/10.1071/SR12175

Application of a RUSLE-based soil erosion modelling on Mauritius Island

Embed Size (px)

Citation preview

Page 1: Application of a RUSLE-based soil erosion modelling on Mauritius Island

Application of a RUSLE-based soil erosion modellingon Mauritius Island

Rody NigelA,C and Soonil D. D. V. RughooputhB

AStagiaire Postdoctoral, Bureau 5310, INRS Centre Eau Terre Environnement, 490 rue de la Couronne,Québec, QC G1K 9A9, Canada.

BFaculty of Science, University of Mauritius, Réduit, Mauritius.CCorresponding author. Email: [email protected]

Abstract. Soil erosion by water is one of the most important natural resources management problems in the world. Thedamages it causes on-site are soil loss, breakdown of soil structure, and decline in organic matter content, nutrient content,fertility, and infiltration rate. Lands with the highest erosion risk on Mauritius Island are crop cultivations (sugarcane, tea,vegetables) on erosion-susceptible terrain (slopes >20% coupled with highly erodible soils). The locations of such lands onMauritius were mapped during previous, qualitatively based regional-scale erosion studies. In order to propose soilconservation strategies, there is a need to apply a more quantitative approach to supplement the previous, qualitativelybased studies. This paper reports an application of the Revised Universal Soil Loss Equation (RUSLE) within ageographical information system in order to estimate soil loss on the island, and particularly for the high-erosionareas. Results show that total soil loss on the island is estimated at 298 259 t year–1, with soil loss from high-erosion areassumming 84 780 t year–1 (28% of total soil loss). If all of the high-erosion areas were afforested, their soil loss would bereduced to 10 264 t year–1, i.e. a reduction of 88% for the high-erosion areas and a reduction of 25% for the island. Thisstudy thus calls for soil and water conservation programs directed to these erosion-prone areas before the land degradationand environmental damage they are causing become irreversible. The methodological approach used in this work toquantitatively estimate soil loss from erosion-prone areas can be adopted in other countries as the basis for a nationwideerosion assessment in order to better inform environmental policy needs for soil and water conservation.

Additional keywords: erosion risk mapping, GIS, priority action areas.

Received 4 July 2012, accepted 19 December 2012, published online 31 January 2013

Introduction

Soil erosion is an environmental problem that exacerbates on-siteland degradation, while at the same time being a source ofsediment and pollutants that adversely affect off-site aquaticecosystems (Lal 2001). Mauritius Island is potentially at riskto soil erosion because of its rugged topography, extensivesugarcane cultivation, and tropical climate. This makes theland particularly vulnerable to erosion. In addition, otherecosystems can be damaged due to erosion, namely, estuaries,marshes, fish habitats, sea grass, and coral reefs.

Conservation measures are needed to reduce the effects ofsoil erosion, and successful conservation programs require theconcentration of resources on priority action areas. Soil-erosionrisk assessment at a regional scale is commonly used to identifysuch conservation areas (e.g. Vrieling et al. 2002). The firstpublished study to map the erosion risk across the whole islandwas conducted by Nigel and Rughooputh (2010b), developing asoil-erosion risk-mapping model termed MauSERM (MauritiusSoil Erosion Risk Mapping). Four erosion factors, namelyrainfall, land cover, topography, and soil, were consideredduring the erosion risk mapping using the MauSERM model.Investigations were also made on the best way to combine the

factors, and the decision rule methodology was chosen, as it canbe easily manipulated to mimic the complex interactions of allfour factors (as opposed to factorial scoring).

A second study was later conducted by running the modelwith new datasets, where slope gradient was replaced with slopelength (inclusive of land parcels and drainage structure effects).The results illustrate that erosion sites occur commonly incultivated areas with steep slopes and erosion has a summer-dominant pattern caused by intensive summer rainfall (Nigel andRughooputh 2010a). This study was qualitatively based andenabled identification of priority action areas. Now, a morequantitative study is needed, as this can provide assistance inriver basin management and to indicate the order of magnitudeof soil erosion and sediment yield (de Vente and Poesen 2005).

The aim of the present study is to compute soil loss estimatesfor Mauritius Island and especially for the priority action areas.The specific objectives are: (1) apply the Revised UniversalSoil Loss Equation (RUSLE) model for the whole island,(2) compute soil loss estimates for the high-erosion areas(HEA) initially mapped, and (3) define a second approach formapping of HEA and derive estimates of RUSLE soil loss forthese HEA. In this second approach, the HEA can be mapped

Journal compilation � CSIRO 2012 www.publish.csiro.au/journals/sr

CSIRO PUBLISHING

Soil Research, 2012, 50, 645–651http://dx.doi.org/10.1071/SR12175

Page 2: Application of a RUSLE-based soil erosion modelling on Mauritius Island

from only two factors such as land cover and slope gradient,which are readily available, instead of soil erodibility and lengthof slope, which are generally more difficult to map at large scaleas they require more geographical information system (GIS)datasets (see e.g. Vrieling 2006). The RUSLE soil map producedwill be the first for Mauritius and will enable regional erosionevaluation for the determination of areas where soil conservationshould be emphasised.

Materials and methods

Study area

Mauritius is in the south-west of the Indian Ocean (Fig. 1) atlatitude 208100S and longitude 578300E. The island has anelliptical shape with a major axis 63 km, minor axis 43 km,surface area 1859 km2

, and highest peak 828m altitude. Theisland is of volcanic origin with eruptions lasting from 10 to0.025million years Before Present. Two main soil groups exist:mature Latosols originating from highly weathered, basaltic lava

rock; and immature Latosolic soils with minerals still in theprocess of weathering. Most soils have high clay contents andhigh percentages of stone and organic matter content (organicmatter 3–11%). Soil depth is mainly 50–100 cm, similar tomaximum plant rooting depth (Parish and Feillafé 1965).

The climate is tropical maritime with two seasons: a rainysummer from November to April dominated by cyclonepassage, and a dry winter from May to October dominated bythe South-East Trade Wind and frontal systems. About 70% ofmean annual rainfall is received during summer. February is thewettest and hottest month, whereas October is the driest month.Mean annual rainfall is ~2000mm, equivalent to ~3700Mm3, ofwhich annual evaporation is 30%, surface runoff 60%, andgroundwater recharge 10% (WRU 2007). Surface runoff isconfined within 213 river basins, which occupy 83% of theisland (Fig. 1) (Nigel and Rughooputh 2010a). Torrential flowswith severe bank erosion and turbidity in the lagoon are commonduring intense rainfall events (Arlidge and Wong You Cheong1975).

Fig. 1. Study area, Mauritius Island.

646 Soil Research R. Nigel and S. D. D. V. Rughooputh

Page 3: Application of a RUSLE-based soil erosion modelling on Mauritius Island

There is extensive sugarcane cultivation, which occupies54.1% of the island, while forest covers 26.7%, sparsevegetation 7.1%, urban areas 5.5%, scrub 2.9%, tea 1.6%,water bodies 0.9%, barren lands 0.5%, wetlands 0.4%,vegetables 0.2%, and sand 0.1% (Nigel and Rughooputh2010b). Sugarcane is harvested each year from July toOctober, during which the cane ratoon is not removed butleft over for regrowth and replaced usually every 5–7 years,at which time the whole field undergoes tillage. Seeruttunet al. (2007) measured soil loss for five sites for fourconsecutive years, with each site having two plots: one bareand the other planted with sugarcane. The rate of soil loss frombare plots was in the range 0.5–37.6 t ha–1 year–1, whilesugarcane reduced soil loss by 80–99%. However, thesestudies were conducted only on plots having linear slopesand isolated from upslope contributing areas (soil type andrainfall erosivity did not vary over each plot size).

Estimating soil loss pattern with RUSLE

Like most qualitative models, one of the limitations ofMauSERM is that it provides qualitative outputs, which arenot linked to quantitative estimates of soil loss. These wouldhave been very useful for planning purposes, particularly for thehigh erosion areas. Converting erosion risk classes into estimatesof soil loss may be done by upscaling erosion measurementsfor selected sites, as proposed by Vrieling et al. (2002). In thepresent case, no erosion measurements for selected sites wereavailable to perform such an upscaling.

Another approach would be the use of a quantitative erosionmodel, such as the RUSLE model (Renard et al. 1997) or its firstversion, the USLE (Wischmeier and Smith 1978). These modelsare widely used for modelling soil loss quantitatively, includingfor large areas by using a GIS (e.g. Lu et al. 2003). It must benoted that the original USLE model is a quantitative model, butapplication of the model at a regional scale within a GIS requiressimplification of the model, and therefore, its use within a GISis commonly referred to as a semi-quantitative modelling oferosion. The soil loss equation for the RUSLE model is given asfollows:

A ¼ R� K � LS � C � P ð1Þwhere A is the spatial average soil loss (t ha–1 year–1), R is therainfall erosivity factor (MJ.mmha–1 h–1 year–1), K is the soilerodibility factor (t.hMJ–1mm–1), LS is the slope length andsteepness factors (unitless), C is the cover management factor(unitless, values ranging from 0 to 1), and P is the supportpractice factor (unitless, values ranging from 0 to 1).

In the present case, we computed rainfall erosivity R usingEqn 2 (Arnoldus 1980), and which was found appropriate for uselocally (Le Roux et al. 2005):

R ¼ 0:0302ðMFIÞ1:9 ð2Þwhere MFI is the Modified Fournier Index, and is given byEqn 3:

MFI ¼Xi¼12

i¼1

ðMiÞ2A

ð3Þ

where Mi is the rainfall amount received in month i and A is theannual rainfall amount.

Values of soil erodibility for soil types (Table 1) wereobtained from Le Roux (2005) and Seeruttun and Ah Koon(2006). A map of slope length, LS, was obtained from Nigel andRughooputh (2010a), computed by taking into account landparcel effects along with the methodology of Desmet and Govers(1996) as specifically implemented in the USLE2D software ofVan Oost and Govers (2000).

Further data needed for running the RUSLE model above(Eqn 1) are the C and P values for land cover types. These C andP values are given in Table 2, which were obtained mainly fromLe Roux (2005), except for sparse vegetation (which has beenset the same as for forest). These values of C and P were thenassigned to a 10-m cell-size land cover map, resulting in tworaster maps (of C and P) suitable for raster processing of Eqn 1to produce a soil loss map. The land cover map was producedmainly from published map series (updated year 2005) (Nigeland Rughooputh (2010a).

A second approach to identify high-erosion risk areas

Another approach that can be used to establish the high-erosionareas of the island is the use of only two factors that can bedirectly mapped. A land cover map is one such factor. The otheris a slope gradient map, which requires less processing andfewer datasets compared with a ‘land cultivation suitability’map, which in addition to topography requires information onsoil type. Initially, a land cultivation suitability map was used in

Table 2. RUSLE C and P values for the different land cover types

Land cover type C P

Sugarcane 0.011 0.975Sparse vegetation 0.001 1.000Scrub 0.001 1.000Barren lands 0.011 1.000Tea 0.003 1.000Vegetables 0.204 0.625Forest 0.001 1.000Urban areas, water bodies, wetlands and sand Nil 1.000

Table 1. RUSLE K soil erodibility values

Soil order Soil type RUSLE erodibility K(t.hMJ–1mm–1)

Zonal Low Humic Latosols, L 0.10Humic Latosols, H 0.10Humic Ferruginous Latosols, F 0.15

Intrazonal Latosolic Red Prairie Soils, P 0.20Latosolic Brown Forest Soils, B 0.20Dark Mag Clay, M 0.10Grey Hydromorphic Soils, D 0.10Low Humic Gleys, G 0.20Ground Water Laterite, W 0.30Mountain Slope Complexes, S 0.15

Azonal Lithosols, T 0.30Alluvial Soils, A 0.15Regosols, C 0.05

RUSLE-based soil erosion modelling Soil Research 647

Page 4: Application of a RUSLE-based soil erosion modelling on Mauritius Island

Nigel and Rughooputh (2010a), and it required (i) slopegradient, (ii) length of slope, (iii) soil erodibility, and (iv) apre-existing land suitability classification scheme [the onedefined for Mauritius by Arlidge and Wong You Cheong(1975) was used]. Additionally, land cultivation suitabilitywould be more difficult to directly visualise in the fieldcompared with steep slope gradients alone.

As such, based on land cover and slope gradient, a morestraightforward definition of HEA can be obtained. Thisalternative HEA will henceforth be termed HEA-B. Inessence, HEA-B can be mapped based on the occurrences ofcrop cultivation on slopes >20%. In a GIS, this is achieved byextracting land cover classes that show cultivation (sugarcane,tea, and vegetables) and intersecting this dataset with a slopegradient map that shows only values >20%.

Results and discussion

Here, results of the second approach are presented first, followedby results of the RUSLE modelling. Results of the mapping of

HEA-B show that 58 km2 of land has slopes >20% and is beingcultivated (with sugarcane 98.8%, vegetables 0.06%, and tea1.14%). The extent of HEA-B is thus about four times less thanthat of the original HEA (58 v. 251 km2).

The spatial distribution of the established HEA (originalHEA and HEA-B) is shown in Fig. 2. In the event that notall of the 251 km2 of the original HEA will be the focus forconservation efforts (due to financial constraints, for instance),then at least the areas defined as HEA-B, and particularly thesugarcane fields, must be targeted for conservation efforts.

The map of soil loss produced using the RUSLE model isshown in Fig. 3. The soil loss estimates were classified usingthe scheme of Le Roux et al. (2005) for Mauritius. The highestsoil-loss values are mostly distributed within the steep,cultivated hilly areas. The total soil loss on the island isestimated at 298 259 t year–1. Accordingly, based on RUSLErelative modelling, concerning the alternative definition of‘high-erosion areas’ (HEA-B), it is seen that total soil lossfrom the sugarcane fields of HEA-B is 63 673 t year–1, which

High Erosion Areas

HEA-B

HEA

N

0 5 10

km

Fig. 2. High erosion areas of Mauritius Island: HEA and HEA-B.

648 Soil Research R. Nigel and S. D. D. V. Rughooputh

Page 5: Application of a RUSLE-based soil erosion modelling on Mauritius Island

is 66% of total soil loss from all sugarcane fields on the island.Thus, while sugarcane on steep slopes occupies only 58 km2 ofthe 1006 km2 of land cultivated with sugarcane, it is nonethelesscontributing to 66% of total soil loss from all sugarcane fieldsand contributing to 21% of overall soil loss on the island. If atleast these sugarcane fields of HEA-B were afforested, their totalsoil loss would be reduced to 7641 t year–1, which would reducesoil loss from all sugarcane fields by 58% and reduce overall soilloss by 19% (Table 3).

Nonetheless, the first proposal for conservation efforts remainsthe conversion of 251 km2 of HEA into forest or other natural

vegetation, which would reduce the total soil loss from sugarcanefields by 74% and that of the island by 25%. Indeed, based onRUSLE modelling of alternative practices, if all of the HEA wereconverted to forest, their soil loss would be reduced from 84780to 10 264 t year–1, i.e. a reduction of 88% for the HEA and areduction of 25% for the island. Specifically, sugarcane in HEAcontributes 81 389 t year–1 of soil loss, which is 84% of the annualsoil loss from all sugarcane fields on the island (estimated at96865 t year–1).

According to Morgan (2005), predictive soil erosion modelssuch as the RUSLE do not, in general, yield true deterministic

N

0 5

km

RUSLE soil lossValues in t ha–1 year–1, and class

0–5, Very low

5–12, Low

12–25, Moderate

25–60, High

60–150, Very high

>150, Extremely high

10

Fig. 3. Classified soil loss map of Mauritius Island derived with RUSLE.

RUSLE-based soil erosion modelling Soil Research 649

Page 6: Application of a RUSLE-based soil erosion modelling on Mauritius Island

results of erosion rates. In the absence of measured soil loss orsedimentation data for model calibration and validation, sucherosion models are more wisely used for comparing differentland management systems, while their use for calculation ofabsolute soil loss and sediment yield remains a bad practice and,at best, speculative (Morgan 2005).

No sedimentation data have been available in the present caseto validate the RUSLE soil loss estimates. Field measurement ofsoil loss is time-consuming, as at least 2–3 years are neededbefore reliable measurements can be obtained (Lal 1990). Thus,in the present case, the use of the estimated soil loss on a relativebasis is favoured, and this would give an idea of the relativeimpact of afforestation or other revegetation scenarios beforethey are implemented.

The general recommendation is thus to convert the cultivationof the 58 or 251 km2 of HEA (HEA-B or original HEA,respectively) into planted vegetation or forest in order toreduce their soil loss and downstream sedimentation.Consequently, the measurement and modelling of sedimentyield for the HEA, as well as the establishment of suitableriparian vegetated filters, are also recommended for futureresearch.

Using the RUSLE map to build a ‘tolerable soil loss-rate’concept

The steeplands of tropical islands have long been recognised asbeing affected by soil erosion. For example, various USLEerosion modelling studies conducted for Hawaiian Islandshave reported accelerated erosion rates and high sedimentyield for agricultural cultivations (e.g. Calhoun and Fletcher1999). For the Virgin Islands, Ramos-Scharrón and Macdonald(2007) studied erosion for all land-use types with an empiricalGIS-based erosion model and found that the combination ofsteep slopes, small drainage areas, shallow soils, anthropogenicactivities, and high-intensity rainstorms resulted in highsediment yield ~3–9 times higher than under natural conditions.

In Mauritius, the current research is on studying erosionpatterns and estimating soil loss. The next step is to researchthe rates of soil loss that would be ‘tolerable’ for the island,i.e. not exceeding the rate of soil production. In Australia,Bui et al. (2011) reviewed three levels of ‘tolerable’ soil lossthat can be adopted for environmental conservation (T1 for soilconservation, T2 for agricultural productivity, and T3 for waterconservation). A comparison of the work of Bui et al. (2011)with the current study gives an insight into the way forwardfor soil erosion studies on Mauritius, where soil conservationservices, regulations, and policies do not even exist. Thus,drawing on a ‘tolerable soil loss-rate’ concept, it may be saidthat soil erosion rates of 5–12 t ha–1 year–1 for Mauritian soils

will not exceed a ‘tolerable T2’ soil reduction rate of0.4–1mmyear–1, assuming a bulk density of 1200 kgm–3

(based on USDA T2 estimates described by Montgomery2007). Accordingly, for ease of interpretation, the soilloss estimates in Fig. 3 are classified as very low(<5 t ha–1 year–1), low (5–12 t ha–1 year–1), and moderate–high(>12 t ha–1 year–1). In Australia, Bui et al. (2011) reported atolerable soil reduction rate of ~0.015mmyear–1, which issimilar to those in Europe but lower than those adopted bythe USA, China, and India (~0.4–1mmyear–1). In the currentwork for Mauritius, similar higher values are adopted, mainlybased on USDA values, and such values are being adopted in thetropical volcanic islands of Hawaii, albeit this range may havebeen set too high (Montgomery 2007; Bui et al. 2011).

Because accelerated soil erosion rates exceed soil productionrates, a point can be reached where there is no soil left, even ifthis takes 50–5000 years for a soil that had an original depth of1m (Montgomery 2007). In Mauritius, soils have relativelygood depths, being derived from highly weathered basalts(Parish and Feillafé 1965), but all soils, no matter how greattheir initial depth, are at risk of unsustainable, acceleratedsoil erosion rates driven by non-conservative agriculture(Montgomery 2007). Therefore, it is very important foragricultural countries where lands are sensitive to erosion,notably tropical islands with steep cultivated land and intenserainstorms, to at least assess soil erosion patterns and to estimatesoil loss, for instance, by using the RUSLE model as done inthis work. One of the key challenges to the production of anationwide RUSLE soil loss map is to be able to integrate allof the necessary information into a GIS in order to run themodel. As described here, this can be achieved by leveragingon methodological approaches that employ available data, suchas rainfall depths (instead of rainfall intensity measurementsas needed for the original empirical RUSLE). The RUSLEmap produced in this way can be used on a relative basis,for comparing proposed agricultural conservation practices.Producing and using the RUSLE map in this way willfacilitate further research, such as, as mentioned above, forassessing ‘tolerable’ soil reduction rate and discussion ofnational soil conservation strategies. Such an approach wasadopted by Bui et al. (2011) for Australia based on anearlier RUSLE soil loss map produced by Lu et al. (2003).This approach can be followed by other countries affectedby accelerated soil erosion processes and, for Mauritius, isdefinitely a recommended way forward.

Summary

A soil erosion risk mapping model (MauSERM) was developedpreviously for Mauritius. The model enabled an assessment of

Table 3. Results summary for soil loss in high-erosion areas (HEA)

Area Soil loss Conservation scenario (conversion to forest)(km2) (t year–1) Soil loss

(t year–1)% Soil loss reduction

for agriculture% Soil loss reduction

island-wide

Sugarcane fields 1006 96 865 – – –

HEA 251 84 780 10 264 88 25HEA-B 58 63 673 7641 58 19

650 Soil Research R. Nigel and S. D. D. V. Rughooputh

Page 7: Application of a RUSLE-based soil erosion modelling on Mauritius Island

the spatial and temporal variations in erosion patterns and theidentification of high-erosion areas of the island. One of thelimitations of the MauSERM model was its output qualitativeclasses, which could not be linked to quantitative classes of soilloss.

However, as shown in this work, by using the RUSLEmodel,it is possible to compute soil loss for the island and, mostimportantly, for the high-erosion areas. The soil losses fromthese areas were then analysed for different land conversionscenarios, principally sugarcane to forest in steep areas. Such aland conversion scenario (sugarcane to forest on steep slopes)is proposed as one means of reducing soil erosion by 19–25%overall on Mauritius.

As confirmed in this work and as reported for other tropicalislands with steep cultivated lands, there are areas of acceleratederosion that definitely need to be conserved. The methodologicalapproach used in this work to quantitatively and tocartographically estimate the existence of such erosion-proneareas can be adopted in other countries where there are needsto assess, map, and estimate soil loss on a nationwide basis inorder to better inform environmental policy needs for soil andwater conservation.

Acknowledgments

The authors acknowledge the Cartographic Section of the Ministry ofHousing and Lands and the Mauritius Meteorological Services for theprovision of data. Dr KH Mueller while at University of Marburgprovided other base data. This work was supported by a TertiaryEducation Commission (TEC) scholarship for the first author at theUniversity of Mauritius. We also thank the two anonymous reviewersfor their comments and suggestions made on earlier versions of thismanuscript.

References

Arlidge EZ, Wong You Cheong Y (1975) Notes on the land resources andsuitability map of Mauritius at 1 : 50 000 scale. Mauritius Sugar IndustryResearch Institute (MSIRI) Occasional Paper No. 29, pp. 1–137.

Arnoldus HMJ (1980) An approximation of the rainfall factor in theUniversal Soil Loss Equation. In ‘Assessment of erosion’. (Eds M deBoodt, D Gabriels) pp. 127–132. (John Wiley & Sons Ltd: Chichester,UK)

Bui EN, Hancock GJ,Wilkinson SN (2011) ‘Tolerable’ hillslope soil erosionrates in Australia: Linking science and policy. Agriculture, Ecosystems&Environment 144, 136–149. doi:10.1016/j.agee.2011.07.022

Calhoun RS, Fletcher CH (1999) Measured and predicted sediment yieldfrom a subtropical, heavy rainfall, steep-sided river basin: Hanalei,Kauai, Hawaiian Islands. Geomorphology 30, 213–226. doi:10.1016/S0169-555X(99)00030-6

de Vente J, Poesen J (2005) Predicting soil erosion and sediment yield at thebasin scale: Scale issues and semi-quantitative models. Earth-ScienceReviews 71, 95–125. doi:10.1016/j.earscirev.2005.02.002

Desmet PJJ, Govers G (1996) AGIS-procedure for the automated calculationof the USLE LS-factor on topographically complex landscape units.Journal of Soil and Water Conservation 51, 427–433.

Lal R (1990) ‘Soil erosion in the tropics—Principles and management.’(McGraw-Hill: Columbus, OH)

Lal R (2001) Soil degradation by erosion. LandDegradation &Development12, 519–539. doi:10.1002/ldr.472

Le Roux JJ (2005) Soil erosion prediction under changing land use onMauritius. MSc Thesis, University of Pretoria, South Africa.

Le Roux JJ, Sumner PD, Rughooputh SDDV (2005) Erosion modellingand soil loss prediction under changing land use for a catchment onMauritius. The South African Geographical Journal 87, 129–140.doi:10.1080/03736245.2005.9713836

Lu H, Prosser IP, Moran CJ, Gallant JC, Priestley G, Stevenson JG (2003)Predicting sheetwash and rill erosion over the Australian continent.Australian Journal of Soil Research 41, 1037–1062. doi:10.1071/SR02157

Montgomery DR (2007) Soil erosion and agricultural sustainability.Proceedings of the National Academy of Sciences of the UnitedStates of America 104, 13 268–13 272. doi:10.1073/pnas.0611508104

Morgan RPC (2005) ‘Soil erosion and conservation.’ (Blackwell Publishing:Oxford, UK)

Nigel R, Rughooputh SDDV (2010a) Soil erosion risk mapping with newdatasets: An improved identification and prioritisation of high erosionrisk areas. Catena 82, 191–205. doi:10.1016/j.catena.2010.06.005

Nigel R, Rughooputh SDDV (2010b) Mapping of monthly soil erosion riskof mainland Mauritius and its aggregation with delineated basins.Geomorphology 114, 101–114. doi:10.1016/j.geomorph.2009.06.013

Parish DH, Feillafé SM (1965) Notes on the 1 : 100 000 soil map ofMauritius. Mauritius Sugar Industry Research Institute (MSIRI)Occasional Paper No. 22, pp. 1–45.

Ramos-Scharrón CE, Macdonald LH (2007) Development and applicationof a GIS-based sediment budget model. Journal of EnvironmentalManagement 84, 157–172. doi:10.1016/j.jenvman.2006.05.019

Renard GI, Foster GR, Weesies GA, McCool DK, Yoder DC (1997)‘Predicting soil erosion by water: A guide to conservation planningwith the Revised Universal Soil Loss Equation (RUSLE)’. AgricultureHandbook No. 703. (USDA: Washington, DC)

Seeruttun S, Ah Koon D (2006) Soil erosion in Mauritius & its mitigation bysugar cane cultivation. MSIRI Talks, 5 December 2006. Mauritius SugarIndustry Research Institute, Reduit, Mauritius. Available at: www.msiri.mu/index.php?langue=eng&rub=215

Seeruttun S, Ah Koon D, Ismael FM, Ng Cheong R (2007) Sugarcanereduces soil erosion by more than 80% in five major soil groups ofMauritius. In ‘Proceedings 26th Congress of the International Society ofSugar Cane Technologist (ISSCT)’. pp. 384–387. (ISSCT: Durban)

Van Oost KG, Govers G (2000) USLE2D software for calculatingLS factor. University of Leuven, Laboratory for ExperimentalGeomorphology. Available at: http://geo.kuleuven.be/geography/modelling/erosion/usle2d/

Vrieling A (2006) Satellite remote sensing for water erosion assessment:a review. Catena 65, 2–18. doi:10.1016/j.catena.2005.10.005

Vrieling A, Sterk G, Beaulieu N (2002) Erosion risk mapping: amethodological case study in the Colombian Eastern Plains. Journalof Soil and Water Conservation 57, 158–163.

Wischmeier HW, Smith DD (1978) ‘Predicting rainfall erosion losses—Aguide to conservation planning.’ Agriculture Handbook No. 537.(USDA: Washington, DC)

WRU (2007) ‘Hydrology data book for period 2000–2005.’ (WaterResources Unit: Rose-Hill, Mauritius)

RUSLE-based soil erosion modelling Soil Research 651

www.publish.csiro.au/journals/sr