D.P. Shrestha

Land use classification in mountainous areas: integration of image processing, digital elevation data and field knowledge (application to Nepal)

Remote sensing data help in mapping land resources, especially in mountainous areas where accessibility is limited. In such areas, land degradation is a main concern. Land is degraded not only by natural processes but also by human activities through inappropriate land use practices. Land cover and land use mapping is thus very important for evaluating natural resources. Classification of remote sensing data in mountainous terrain is problematic because of variations in the sun illumination angle. This results in biased reflectance data, the distribution of which does not fulfil normality as required by the maximum likelihood classifier. In the present work the topographic effect is corrected by normalising the spectral bands by the total intensity. Classification results are further refined by using ancillary data and expert knowledge of the area. The integration of image processing and spatial analysis functions in GIS improves the overall classification result from 67 to 94 percent (a 27 percent increase).

Mapping and modelling mass movements and gullies in mountainous areas using remote sensing and GIS techniques

Abstract Natural as well as human-induced mass movements and gullies are severe environmental hazards. Remote sensing data offer promising possibilities for identification and monitoring. But their effective use in mountainous areas is hampered by cloud effects and relief-controlled factors, which cause geometric distortions and shadow areas, among other constraints. Nevertheless, aerial photographs and satellite images (visible, infrared and microwave bands), or combinations thereof, have been successfully used to discriminate and delineate landslide and gully types. GIS modelling of mass movements and gullies, using ancillary information in combination with remote sensing data, is rapidly developing. The shortcomings of deterministic modelling of such chaotic phenomena as mass movements and gullies highlight the relevance of GIS-assisted approaches to exploratory and predictive modelling. This paper describes practical applications of remote sensing and GIS for mapping, monitoring, exploring cause-effect relationships and assessing hazards of mass movements and gullies in hilly and mountainous areas.

Artificial Neural Network and Decision Tree in Predictive Soil Mapping of Hoi Num Rin Sub-Watershed, Thailand

The demand for high-resolution soil mapping is growing increasingly, in particular for the purpose of land degradation studies. The objective of this study focuses on applying the methods for digital predictive soil mapping in inaccessible, land degradation-prone areas. Artificial Neural Network (ANN) and Decision Tree (DT) were employed within the GIS environment to comply with the complexity of the soil forming factors governing the soil formation. Following the principles of the geopedologic approach to soil survey, a digital predictive soil mapping was carried out in Hoi Num Rin sub-watershed, covering an area about 20 km2. Both ANN and DT were applied to properly integrate the parameterized soil forming factors. To describe soil predictors to train the ANN and to formulate the decision trees, 4 organism types, 7 relief type units, 9 lithological units, 3 time series, 4 landscape units and 8 landform units were extracted from the map and databases. The results, the 10 soil class names were extrapolated to the unsampled areas to obtain the geopedologic map. In conclusion, the geopedologic approach to soil survey, which is based on understanding of landscape-soil relationship, is helpful to obtain spatial soil information in inaccessible areas, using ANN and/or DT are useful techniques in modeling the complex interactions among the soil forming factors. The difference, however, is that ANN, once it is well learnt, is faster, thus more recommendable in terms of time and cost saving.

Do the Emerging Methods of Digital Soil Mapping Have Anything to Learn from the Geopedologic Approach to Soil Mapping and Vice Versa

The use of soil maps and the feasibility of the existing soil survey procedure are often questioned by both surveyors and users. Thanks to the advances in the fields of remote sensing (RS) and geographic information system (GIS), a new trend – digital soil mapping – is emerging which might have answers to some of the questions. With a glance to some of the definitions and concepts, such as ‘what is a soil?’ and ‘what is the content of a soil map?’ we intend to highlight the complexity of the soil and its mapping. At the same time, we apply some of the geopedologic-oriented techniques of the digital terrain modelling to soil mapping in order to show the role of geomorphology in the mapping. The exercise was carried out as case studies in several areas in Thailand. Various soils at subgroup levels (Fluventic, Arenic, Aquic, Aeric, Ultic, Ustic, Vertic) belonging to the soil orders Entisols, Mollisols, Inceptisols, Alfisols, and Ultisols occur in different geomorphic surfaces, following well the physiographic setup of the landscapes. The case studies demonstrate the conventional predictive mapping (the ITC approach) and the geopedologic approach to soil survey, based on parameterisation of the soil-forming factors and their integration: in one case through applying decision trees, followed up by a statistical validation, and in another case by means of Artificial Neural Network (ANN). It is hoped to open up a discussion, which should lead to (1) clarifying the term ‘digital soil mapping’ and (2) finding out whether or not the shortcomings of the conventional approach of soil mapping can be recovered using the new trend and does the new trend suggest changes in the current definitions and concepts.

Modelling erosion on a daily basis, an adaptation of the MMF approach

Abstract Effect of soil erosion causing negative impact on ecosystem services and food security is well known. On the other hand there can be yearly variation of total precipitation received in an area, with the presence of extreme rains. To assess annual erosion rates various empirical models have been extensively used in all the climatic regions. While these models are simple to operate and do not require lot of input data, the effect of extreme rain is not taken into account. Although physically based models are available to simulate erosion processes including particle detachment, transportation and deposition of sediments during a storm they are not applicable for assessing annual soil loss rates. Moreover storm event data may not be available everywhere prohibiting their extensive use. In this paper we describe a method by adapting the revised MMF model to assess erosion on daily basis so that the effects of extreme rains are taken into account. We couple it to a simple surface soil moisture balance and include estimation of daily vegetation cover changes for calculating rain interception and for estimating effective rain. The runoff fraction is based on the available daily storage of the effective hydrological depth. Unlike the original MMF model which accumulates annual runoff, the daily model accumulates daily runoff from upstream contributing area in a predefined flow network according to steepest slope direction. Annual soil loss is calculated by adding daily erosion rates. We compare the obtained results with that obtained from applying the revised MMF model in two locations: (i) in sub-humid tropics in central Thailand which is affected by deforestation and land cover changes resulting in excessive soil losses, and (ii) in semi-arid environment in Morocco which is affected by severe gully formation. Since the model results are daily estimates it is possible to see the effects of exceptional rain. In Morocco, the effects of extreme rains are clearly shown which were absent in the results obtained by using the annual model. The results also show that erosion rates can be moderate when rainfall pattern is normal without extreme rains in a year although total rain may be similar. This is clearly shown in the erosion assessment in Thailand for the years 2005 (1390 mm) and 2006 (1366 mm, and with the presence of extreme rainy days).

Modelling spatially distributed surface runoff generation using SWAT-VSA: a case study in a watershed of the north-west Himalayan landscape

Soil and Water Assessment Tool (SWAT) and Soil and Water Assessment Tool-Variable Source Area (SWAT-VSA) models were employed to predict surface runoff generation in a watershed of the Himalayan landscape in GIS environment. Both the models differed in term of defining hydrological response units (HRUs) that serves as basis in assigning curve number for surface runoff estimation. HRUs in SWAT was derived by combination of hydrological soil groups based on soil types and land use/land cover (LULC) whereas in SWAT-VSA, it was based on soil wetness index derived from digital elevation model (DEM) and LULC. Both models were calibrated to predict surface runoff at watershed scale. SWAT-VSA predicted quite well [Root Mean Square Error (RMSE)

Geopedology Promotes Precision and Efficiency in Soil Mapping. Photo-Interpretation Application in the Henares River Valley, Spain

= 3.88

Modelling land degradation in the Nepalese Himalaya

=3.88, Nash–Sutcliffe coefficient of efficiency (NSE)