HEADWATER CONTROL: Despatches from the Research Front

摘要

Headwater Control is about integrated land management. Effective headwater management requires better information, better technologies, better management structures, change in land husbandry and direct engineering interventions. Headwater regions often waste their resources. In Jammu and Kashmir, 40% (and in South Africa and Iranian Azerbaijan 60%) of the water resource runs to waste. Azerbaijan's planners believe that 25% could be collected for agriculture. In Xinjiang, China, by contrast, over-utilisation of the headwater resources in the Tarim Basin has caused desertification and large-scale farmland abandonment downstream. Better systems of water resource accounting are required. Eventually, the Lesotho Highlands Water Project may generate 25% of that nation's foreign exchange earnings. Acidification is linked to the human impact in headwaters. Previously, pH was controlled by climatic, lithologic, edaphic and biologic factors. New Japanese work stresses the role of the weathered layer beneath the soils of tropical steeplands in moderating the hydrologic impacts of forestation and acid precipitation. Elsewhere, pollution plays a critical role, especially the deposition of air pollutants. Time trends of NO_3 and SO_4~(2-) deposition may be dwarfed by natural geographical variations within a small basin. However, the chief agencies are decreasingly SO_4~(2-) ", increasingly NO_3 and more generally agricultural chemicals and other human wastes. Riverine microbial biofilms act as both sinks and sources for pollutants. In Canada, effective prevention of non-point-source pollution required a compromise between the net welfare of farmers, the wider non-farm community and the environment. Better systems of water quality management are required. The riparian zone is often counted critical. Fragmentation of both the riparian zone and its management rank among the most serious challenges for land managers. However, while restoring riparian vegetation reduces sediment loads and increases the biodiversity of affected rivers, it may not mitigate acidification. Mathematical models for predicting runoff from alpine catchments are evaluated. USDAHL with GIS, HEC-1 and RUNOFF-5 are generally satisfactory and HYRROM is useful for engineering. Empirical studies show that forests moderate hydrological and erosional processes. In Serbia, 15-years study in steep headwater basins finds that forest covers >70%, rather than <49%, yield smaller flood peaks and provide a more sustained flow (dry for 13.4 days/year-1 vs. 121.4 or more). In Japan's mountain headwaters, reforestation reduces water losses. Here, increased evapotranspiration is compensated by decreased deep seepage. In Alaska, tree roots prove as important as pore water pressure in regulating landsliding on steep slopes with shallow soils. Soil erosion remains a major problem. In Brazil, regional estimates of soil loss were a seventh of those due to rill and gully erosion in fields under conventional tillage. In Lithuania, soil depth declines ranging from 0.12 - 0.94 m and soil fertility declines ranging from 22% to 63% are reported. In the Himalaya, soil nutrient deficiencies are widespread and agricultural inputs are inadequate to the maintenance of fertility levels. Overgrazing, forest conversion, the harvesting of wild medicinal herbs, road construction and mining cause soil degradation and permit ecological invasion by alien weed species. The situation of the headwater farmer, struggling with land degradation, contrasts with that of the padi farmer downstream, whose fields are enriched with soils and nutrients lost from the uplands. Elsewhere, sediment and agrochemical pollution causes problems downstream. In Canada, conservation tillage proves best pollution control practice for both the farmers and non-farm community. In Brazil, rill and gully erosion was eliminated by the adoption of no-till agriculture. In Lithuania, returning steeper slopes to pasture reduced soil losses to zero