Study On Lake Processes Effects On Regional Climate And Its Prediction Based On The Downscaling Method Of The WRF Model

In recent years,with the warming of the global climate,extreme weather occurs frequently all over the world.As a part of the terrestrial hydrosphere,lakes participate in the local and global water and energy cycles.As a part of the land surface processes,lake processes deeply affect the local and global weather and climate systems.Lake effect precipitation is the main manifestation of the influence of lakes on local weather.Forecasting accurate precipitation around lake and mastering the feature of lakes influence on local precipitation can reduce the loss caused by extreme precipitation around lake,which is very important for government to make use of local resources and make decision correctly.In this study,WRF_Lake model is used to simulate the lake surface temperature?TSK?and precipitation of Poyang Lake and the Great Lakes with the dynamic downscaling method,which are the largest freshwater lakes in China and the United States respectively.Using the results from CFS_Flake to drove WRF_Lake predicted precipitation.Firstly,we evaluate the applicability of WRF_Lake over Poyang Lake and analyze the influence of Poyang Lake on local precipitation during the summer from 2004 to 2008.Then we evaluate the simulated results from WRF_Lake over the Great Lakes during the winter from 2001 to 2010 and adjust the model.Finally,the modified WRF_Lake is used to forecast 20-year average precipitation from 1997 to December 2016 in the Great Lakes.The main results of the study are:?1?WRF_Lake model can reasonably simulate TSK,2 meter air temperature and the occurrence of precipitation in Poyang Lake area,with the correlation of 0.97,0.84 and 0.8,respectively.The simulated precipitation also show good spatial distribution,indicating that WRF_Lake model is suitable for Poyang Lake.?2?The influence of Poyang Lake on precipitation over the lake area changes with different time.In June,Poyang Lake showa a positive effect of promoting precipitation,and increases precipitation by 25-35%over the lake;but show a negative effect in July and August and inhibited 15-35%and 15-25%precipitation respectively over the lake.The influence of Poyang Lake on precipitation around lake area?27.75°N-30°N,115°E-117.5°E?also changes with different time.In June and July,the precipitation in the area increased by 1.5%and 3.4%respectively.In August,it shows the effect of inhibiting precipitation.?3?There are some limitations for WRF_Lake to simulate lake surface temperature in different lakes,so it is necessary to pay attention to the evaluation before use and the setting of relevant parameters.The original WRF_Lake simulates unreasonably the TSK of the Great Lakes in the winter of 2001 to 2010,while WRF_Lake can reasonably simulate the winter TSK of Poyang Lake from 2004 to 2008.?4?The surface roughness and depth of the lake in the WRF_Lake model are the main factors that affect the TSK simulation.The study found that the suitable surface roughness of the Great Lakes is 2×10^-5 m.WRF_MLake can significantly improve the simulation of TSK in winter in December.The spin-up time of the model has a greater impact on the lake temperature.The WRF_MLake model corrects the original model underestimation of precipitation in the lake area.The average deviation of 10-year precipitation has been reduced from the previous-2 mm to 0.07 mm.The model improved the distribution of precipitation in the lake area by simulating more reasonably.?5?The CFS_Flake did not show a significant decrease in forecast accuracy with the longer forecast duration when forecasting the average precipitation in the Great Lakes region in winter from 1997 to December 2016.However,the root mean-square error of precipitation for Lead1?18.4 mm?is better than Lead5?19.4 mm?.Using the modified WRF_Lake for downscaling can significantly reduce the root mean square error of the forecasting precipitation at all initial times of Lead1.The percentage of improvement at 1th,11th,and21th are 2.2%,14.4%,and 16.4%,respectively,indicating that downscaling can reduce CFS_Flake forecasting error,improving the accuracy of forecast results.The combination of forecasting model and regional climate model can achieve the purpose of accurate forecast,which is of great significance for the accurate prediction of precipitation in the great lakes region in China.