No abstract

No abstract

In the western USA. wildland \oil wrvey areas are often large. and the resources of money. personnel. and time required for conventional w i l wrvey techniques are in 4hort supply. We evaluated an alternative methodology for producing soil maps through a process of transecting, model construction. and projection onto a map base using .4RC/INFO geographic information system (CIS) technology. We conducted this study in the Libby Creek watershed in Wyoming where \oil distribution (Cryobonlfs. Cryoborolls, Cryaquolls. Cryaquents. Cqochrepts. and Cryrthents) is a function of geology, slope stability, and vegetation. The GIS-generated soils map was compared with existing general (Order 4) and detailed (Order 3) soils maps prepared for the U.S. Forest Service (USFS). Discrepancies noted between the GIS-generated map and USFS maps included Cryochrepts were the dominant soil on the CIS map (a?&), but comprised only 1506 on the USFS detailed soils map: Cryumbrepts occupied 1996 of the USFS general soils map but only 306 on the GXS-derived soils map: and no Cryumbrepts were delineated in the study area on the USFS detailed soils map. Only two of the eight Cryumbrepts sampled occurred within Cryumbrept delineations on the USFS general soils map. Of the 37 pedons sampled and classified along the five transects across Libby Creek watershed, 11 (30'6) corresponded to named soils of mapping units in the USFS general soils map, and 20 (54'6) coincided on the USFS detailed soils map. Results of this study suggest transecting and GIS-based mapping can be an effective technique for producing general soils maps, and can aid in placing soil boundaries for detailed soils maps. ONVENTIONXL SOIL SURVEY METHODS require re-C sources of money. personnel. and time, which are often in short supply. This is particularly true in the western USA where soil surveys of wildlands may encompass areas of > I million ha. In the USA. soils are mapped through identification of a limited number of key characteristics recognized in the soil taxonomic system (Soil Survey Staff. 1996). Only in the most detailed soil surveys (Order 1) are soils mapped directly by Sampling all individual pedons and the mapping unit boundaries observed in their entirety (Soil Survey Staff. 1993). In most soil surveys. soils are mapped using a model of soil occurrence. based on the five soil-forming factors (Jenny. 1941). and air photos or other remote sensing techniques. Topography, vegetation patterns, and color tones on the air photos are commonly used to help soil scientists place soil mapping unit boundaries. Thoughtful and consistent ground truthing is required to maintain the integrity of such maps. To construct useful soils maps, soil scientists must identify and understand the spatial variability of soils within the landscape. Statistics. and more recently geostatistics. allow soil scientists to quantify mapping unit

The general objective of the WATERMED project is to develop a comprehensive method for the study of the water use and the resistance to the drought of the natural and irrigated vegetation in the Mediterranean Basin (see Fig. 1), by means of a combined historical and current space-based remote sensing database, vegetation models and field measurements. The general concept is to integrate all available data of the studied environments.

In the present paper, hydro-environmental behaviour of two adjacent but characteristically different land areas of (i) semi-arid Barind Tract and (ii) Ganges flood plain situated in the north-western part of Bangladesh have been studied and possible mechanisms of variability in the two study sites have been investigated. Time series remote sensing data supplemented with climatic, geophysical and groundwater data and other ancillary information have been analysed for the purpose. Coupling of Landsat TM derived vegetation cover information with meteorological data depicts higher sensitivity of Barind Tract than the other study site towards vegetation cover changes in response to rainfall amount. Temporal sequence of albedo value as derived from Landsat TM data manifests appropriate seasonal trend that is consistent with field observations. Analysis of rainfall data over a period of about 50 years depicts substantially higher rainfall in both the study sites as compared to a typical semiarid region. However, temporal distribution of rainfall events is mostly concentrated within very short period of time in the year. Lithological information on the subsurface configuration depicts the existence of a relatively impermeable clayey layer near the top surface in these areas. Relatively higher thickness of this impermeable clayey layer over the semiarid Barind areas has been noticed that slows down the infiltration process significantly as compared to that in adjacent floodplain areas. Analysis of drainage network exhibits relatively higher drainage density in the Barind Tract area as compared to that in floodplain area. The drainage pattern shows predominantly unidirectional and distributary nature of drainage in the Barind Tract area and in general, distribution is oriented towards basin direction and most of the branches of the drainage system. Ultimately, relatively higher elevation, steeper slope result in relatively quicker surface runoff to carry out excess rain water to the nearest basins, results in decreased infiltration of rainwater. Eventually, the conversion efficiency of Barind Tract area to transform intercepted rainwater into underground reservoir water through infiltration seems to be lower as compared to that of floodplain areas.