| Precipitation has complex spatial variability because it influenced by atmospheric movement, sea-land position and underlying surface conditions and other factors. On the macro scale, the distribution of the precipitation followed by zonal distribution pattern, but it is complex and volatile on the micro scale. For a long time quantitative estimate the spatial distribution of the precipitation has been a research hotspot in meteorology and hydrology. Based on the previous research of the related topic, the study area of this paper is Tai-Hang mountains, following the idea of separation and combination, separating the precipitation estimation for the background field precipitation and the orographic precipitation modification. Firstly, using conventional precision index and terrain performance ability to evaluate different precipitation products(TRMM 3b43 V7, CMORPH 1.0, CMPA).Choosing the precipitation product which has the highest precision as precipitation background field. Then based on the semi-empirical mathematical model, blending meteorological station data, NCEP FNL data, digital elevation model and digital geomorphologic data for estimation of orographic precipitation. Finally, by using the partial least squares regression, the fine estimation model of monthly precipitation with complicated topography was established. And then the spatial distribution of precipitation over the rugged terrain in Tai-Hang mountains was estimated with 1km×1km resolution on GIS based on the model. On the basis, the accuracy of the estimation model was tried to improve by introduced in the topography precipitation condensation level.The main works in this study are as follows:(1) It is found that these three precipitation products detection results basically agree with climate characteristics of precipitation. CMPA has highest precision, minimum error fluctuation,highly related to topographic factors.Moreover,it has basic performance under various morphology of precipitation distribution so it can be utilized as background field precipitation.However,the predictions of CMPA in mountainsous areas and complicated regions still exist many generous questions. Its spatial dimension is too large to meet the needs of small regional meteorological studies, so we must make use of the orographic precipitation modification to modify its precision(2) The simulation result of correction factor for aspect could finely describe topographic characteristics of windward slope and lee slope. According to the results,the plus or minus range distributions of orographic precipitation modification is well consistent with distributions of windward and leeward slope. Besides that, orographic precipitation modification could describe the topoclimate distribution regular pattern and characteristics of precipitation.(3) The paper synthesize background field precipitation and orographic precipitation modification,using a stepwise regression algorithm with Partial least squares method to establish the fine estimation model of monthly precipitation with complicated topography. Comparisons of MRE predictions show that that error of our model for estimation of monthly precipitation is retained within 10% except winner.The winner months MRE within 20%, annual MRE within 5%. So this theoretical model was considered practicable and dependable.(4) Considering the topography lifting vertical speed, if the topography precipitation increment model was built with the help with the topography precipitation condensation level., the accuracy of the model would be improved. Taking the July precipitation of the Tai-Hang mountains, the correlation index R between the optimized model and the meteorological station is 0.947. Moreover, the MRE, RMSE of the optimized model are 4.64%,7.36mm respectively. Both of them are better than those of the model without optimized. |
PDF Full Text Request