简介：AstochasticmodelfordailyprecipitationsimulationinChinawasdevelopedbasedontheframeworkofa′Richardson-type′weathergeneratorthatisanimportanttoolinstudyingimpactsofweather/climateonavarietyofsystemsincludingecosystemandriskassessment.ThepurposeofthisworkistodevelopaweathergeneratorforapplicationsinChina.Thefocusisonprecipitationsimulationsincedeterminationofotherweathervariablessuchastemperatureisdependentonprecipitationsimulation.AframeworkoffirstorderMarkovChainwithGammaDistributionfordailyprecipitationisadoptedinthiswork.Basedonthisframework,fourparametersofprecipitationsimulationforeachmonthat672stationsalloverChinaweredeterminedusingdailyprecipitationdatafrom1961to2000.Comparedwithpreviousworks,ourestimationfortheparameterswasmadeformorestationsandlongerobservations,whichmakestheweathergeneratormoreapplicableandreliable.Spatialdistributionsofthefourparametersareanalyzedinaregionalclimatecontext.Theseasonalvariationsoftheseparametersatfivestationsrepresentingregionaldifferencesarediscussed.Basedontheestimatedmonthlyparametersat672stations,dailyprecipitationsforanyperiodcanbesimulated.A30-yearsimulationwasmadeandcomparedwithobservationsduring1971-2000intermsofannualandmonthlystatistics.Theresultsaresatisfactory,whichdemonstratestheusefulnessoftheweathergenerator.