Causal effects of shelter forests and water factors on desertification control during 2000–2010 at the Horqin Sandy Land region, China

The Horqin Sandy Land（HSL）, the largest sandy land in the semi-arid agro-pastoral ecotone of Northeast China, has been subject to desertification during the past century. In response, and to control the desertification,government implemented the Three-North Shelter/Protective Forest Program, world＇s largest ecological reforestation/afforestation restoration program. The program began in1978 and will continue for 75 years until 2050. Understanding the dynamics of desertification and its driving forces is a precondition for controlling desertification.However, there is little evidence to directly link causal effects with desertification process（i.e., on the changing area of sandy land） because desertification is a complex process,that can be affected by vegetation（including vegetation cover and extent of shelter forests） and water factors such as precipitation, surface soil moisture, and evapotranspiration.The objectives of this study were to identify how influencing factors, especially shelter forests, affected desertification in HSL over a recent decade. We used Landsat TM imagery analysis and path analysis to identify the effects of spatiotemporal changes in water and vegetation parameters during2000–2010. Desertification was controlled during the study period, as indicated by a decrease in desert area at a rate of163.3 km2year-1and an increase in the area with reduced intensity or extent of desertification. Total vegetation cover in HSL increased by 10.6 % during the study period and this factor exerted the greatest direct and indirect effects on slowing desertification. The contribution of total vegetation cover to controlling desertification increased with the intensity of desertification. On slightly and extremely severe desertified areas, vegetation cover contributed 5 and 42 % of the desertification reduction, respectively. There were significant correlations between total vegetation cover and water conditions（i.e., evapotranspiration and precipitation）and the area of shelter forests（P ／ 0.0001）, in which water conditions and the existence of shelter forests contributed49.7 and 12.8 % to total vegetation cover, respectively. The area of shelter forests increased sharply due to program efforts, but only shrub forests had significant direct effects on reducing the area of desertification categorized as slightly desertified. The reason for the lack of direct effect of increased arbor forests（accounting for 95.3 % of the total increase in shelter forests） on reducing desertification might be that the selected arbor species were not suited to water conditions（low precipitation, high evapotranspiration） prevailing at HSL. The establishment of shelter forests aided control of desertification in the HSL region, but the effect was less than expected. Effective control of desertification in the HSL region or other similar sandy areas will require greater improvements in vegetation cover. In particular,shrub species should be selected for plantation with reference to their potential to survive and reproduce in the harsh climatic and weather conditions typical of desertified areas.

G-WADI PERSIANN-CCS GeoServer for extreme precipitation event monitoring

The Center for Hydrometeorology and Remote Sensing at the University of California, Irvine （CHRS） has been collaborating with UNESCO＇s International Hydrological Program （IHP） to build a facility for forecasting and mitigating hydrological disasters. This collaboration has resulted in the development of the Water and Development Information for Arid Lands-- a Global Network （G-WADI） PERSIANN-CCS GeoServer, a near real-time global precipitation visualization and data service. This GeoServer pro- vides to end-users the tools and precipitation data needed to support operational decision making, research and sound water man- agement. This manuscript introduces and demonstrates the practicality of the G-WADI PERSIANN-CCS GeoServer for monitor- ing extreme precipitation events even over regions where ground measurements are sparse. Two extreme events are analyzed. The first event shows an extreme precipitation event causing widespread flooding in Beijing, China and surrotmding districts on July 21, 2012. The second event shows tropical storm Nock-Ten that occurred in late July of 2011 causing widespread flooding in Thailand. Evaluation of PERSIANN-CCS precipitation over Thailand using a rain gauge network is also conducted and discussed.