Drought Monitoring And Early Warning In Mountain And Hilly Area

Drought disaster is one of the major natural disasters which affects agricultural production in China and annual loss accounts for more than15%of all kinds of natural disaster loss. Production of about5billion kg of grain reduced every year because of drought disaster. As global temperatures rising up, the intensity and frequency of the extreme drought climate increase. Droughts are increasingly serious phenomenon in southern humid regions of our country, especially in mountain and hilly area. Susceptibility of natural environment and vulnerability aggravated the negative impact of drought disaster. Under the new situation, the traditional emergency management models of drought have been unable to fully meet the actual needs of modern drought management. Drought management has transferred to risk management. The development of ’3S’ technology provide an effective approach for drought disaster monitoring, which has been widely applied to drought disaster monitoring and early warning in different areas except mountain and hilly area. Agricultural drought disaster is a gradually appeared disaster which is influenced not only by precipitation, evapotranspiration,but also by geographic and geomorphic conditions for they influcenced the redistribution of sunlight, temperature and water.So there was greater uncertainty in agricultural drought disaster of mountain and hilly area. In this paper, combined with the regional characteristics and "3S" technology, drought monitoring and early warning technology on two different scales firstly have been studied; secondly, from the point of disaster system dynamics, comprehensive evaluation methods of drought monitoring and early warning have been studied. The main results were as follows:(1) Based on precipitation anomaly index, the spatial and temporal distribution characteristics of meteorological drought of Wuling mountain area was analyzed. On quarterly scale, in spring and summer, the frequency of mild drought and moderate drought were high. Drought of light level were50.85%and57.38%,respectively,and the requency of moderate drought reached32.79%and37.71%,respectively, which distributed in middle and north of Wuling mountain area; In autumn and winter, frequency of different grades of drought were high. Drought hazard index was constructed, based on that, conclusions were drown that the extent of drought hazard of north area was high in spring and summer, same to that of middle area and east area in autumn, east area and middle area in winter. Drought risk was low in the whole area on yearly scale. Through analysis on the relationship with terrain factors, drought hazards of spring and summer had high relativity to elevation, whose correlation coefficient was0.70,and those of autumn and winter had negative correlation with elevation. Slope and surface rolling had no significant correlation with drought hazard of different seasons.(2) Based on the meteorological drought index (percentage of precipitation anomaly) and remote sensing drought index (temperature and vegetation drought index), integrated drought monitoring index was constructed. Relationship analysis with10cm soil moisture, the fitting accuracy reached significant level in the confidence level of95%, which had been applied to Wuling mountain area. Due to mountain and hilly area affected by the social and economic, strong defensive measures could not be taken when drought happened, which led to different degrees of drought disasters. From the perspective of drought risk management and system dynamics, interaction of the hazard-formative factors, hazard-formative environments and hazard-affected body was taken into account to drought monitoring. That is to say, integrated drought monitoring index which was based on two layers of hazard-formative factors and hazard-formative environments was used to combine with the vulnerability of hazard-affected bodies to build a new drought monitoring and evaluation standard. The method considers the disaster of hazard-affected bodies in order to objectively reflect the situation of drought.(3) Based on weighted markov model, drought early warning model was built and the average accuracy of predicting reached60.94%, which had been applied to evaluate the drought of wuling mountain area. It was thought that the model can be used for drought early warning in mountain and hilly area on month time scale. Due to the hilly region affected by different natural geographical environment and social economy, as meteorological drought occurred, the developing situation of different regions presented different drought level, and the disaster loss was also different. From the perspective of drought risk management and system dynamics, in this study the result of drought early warning based on of hazard-formative factors was coupled with environmental sensitivity and vulnerability of hazard-affected bodies. And a new drought early warning evaluation standard was built, which considered the hazarded-formative environment and hazard-affected bodies. That was advantageous to forecast regional drought disaster risk level objectively.(4) Hydrological model was introduced to drought monitoring and early warning technology. In drought monitoring, potential evaporation was simulated based on based on SWAT model. Crop water deficit index was established by evaporation and precipitation parameters. According to the results of monitoring in2009and2010, it was believed that Crop water deficit index objectively reflected the crop drought conditions and was consistent with the statistical results. The standardized precipitation index could only reflect the regional precipitation anomalies, not effectively reflected the surface water consumption or water deficit content. In the drought early warning, biological yield was simulated based on SWAT model. On the basis of determining quasi normal year, the agricultural drought early warning model was established based on biological yield rate, which was considered to effectively reflect crop drought degree in different region and the change trend of production.