Comparison And Analysis Of Atmospheric Water Cycle Variables Based On Satellite And Reanalysis Data

Water cycle is defined as the cyclic movement process of the earth-atmospheric water in the form of evaporation,precipitation and run off under the effect of solar radiation and gravity.The variables of water cycle include atmospheric water vapor,cloud condensation and so on.Due to the water cycle variables directly affect the atmospheric circulation and climate change,water cycle is restricted to the climate system,at the same time it can feedback to the climate system conversely.Studying the water cycle variables can help us to have a more profound understanding of earth-atmosphere system,and has a profound significance to assess the climate change of the future.In this paper,the water cycle variables involved mainly include two categories,namely hydrometeors and water vapor.For the former,the horizontal and vertical climatological characteristics distribution of hydrometeors over China and its neighborhood based on International Satellite Cloud Climatology Project(ISCCP),Moderate Resolution Imaging Spectroradiometer(MODIS),CloudSat,European Center for Medium-Range Weather Forecasts Interim Re-Analysis(ERA-Interim),Climate Forecast System Reanalysis(CFSR)were analyzed.For the latter,we also use the tropical rainfall measuring satellite TRMM carry microwave imager(TMI)detection results,synchronous inversion on resolution in pixels,of moisture distribution and water vapor in the tropical ocean Diurnal variation were studied.The results show that:In hydrometeors,MODIS,ERA,CFSR show a relatively good agreement in spatial distributions and first dominate mode as well as monthly average timeseries in different regions.The ISCCP shows noticeable disagreement in absolute value and amplitude variation.Among them,the absolute value of MODIS data and amplitude of variation are the biggest,CFSR on land and other data is similar,and the ERA is the smaller of the three.By contrast,although the ISCCP can also capture the horizontal distribution of the hydrometeors and the major characteristics of monthly variation in some areas,but in some specific details and time and space relevance,there is a certain different from the other three datasets,absolute value and amplitude variation is the smallest.In LWP,there is higher consistency in several data.However,the absolute value of the observation data of MODIS and ISCCP significantly below the ERA and CFSR reanalysis data.In particular,the ERA of liquid water content of total Hydrometeor of the largest proportion,and is higher than CFSR both in the horizontal distribution and vertical distribution.These differences Lead to the "uncertainty" of LWP over 60%in low latitude ocean and the south of the qinghai-tibet plateau climate sensitive areas,it need to cause enough attention.In IWP,both horizontal and vertical distribution among different datasets,or specific numerical,there are obvious differences.So besides YunGuiChuan area,other area of the "uncertainty" is big.Among them,The MODIS data has the highest value,CFSR less,and the ratio of ERA of IWP in CWP is lowest,its absolute value is the lowest.By studying the seasonal distribution and seasonal variation of CWP,We can find that,the CWP of Summer in many parts is higher than in winter,and the seasonal variation of CWP in China and its neighborhood is significant.Different datasets when reflect the hydrometeor season distribution is roughly similar,but the observation datasets ISCCP,MODIS,reanalysis data CFSR,ERA in high value center of IWP reaction were significantly different.CWP distribution from different regions is influenced by the terrain,atmospheric circulation and water vapor transmission factors,regional characteristic is obvious.For the vertical distribution of the hydrometeor the liquid water content in southern China is significantly higher than the north.Reanalysis data compared with the observation results of CPR,ERA of liquid water content on the high side and CFSR is low,reanalysis data high value center and the observation data of the high value center are different.For the vertical distribution of ice water content,the reanalysis data CFSR,ERA of ice layer is deeper than CPR and the values of model data with observation data also have differences,shows that the model is likely to overestimate China region cloud thermal dynamic process,mode of ice cloud parameterization scheme is compared with the observed there are still large uncertainties.In aword,the hydrometeors of observations datasets ISCCP,MODIS,CPR and ERA,CFSR reanalysis datasets in China and its surrounding areas are similar,but because of the different satellite instruments,and different models to choose different parameterization schemes,the value of hydrometeor is different.Emphasize again,all datasets have possible error sources(some of them even cannot be overcome).For example,based on passive satellite-borne spectral sensor observations cannot obtain liquid water path information under the ice clouds,and the ice of the cloud based on the cloud radar is questionable,different models of parameterization schemes and underlaying surface treatment uncertainty.Therefore,The purpose of this article,not by comparing to clear what kind of data from the condensation of the result is better,but only is pointed out that there may be differences between different data,so that when using corresponding data analysis to recognize the extent of 'uncertainly' we can contribute a better understanding of cloud radiative properties,as well as learn more about the role of cloud water in climate change.On the other hand,for the water cycle,another important variable water vapor indicates that in terms of water vapor distribution,water vapor distribution of tropical oceans shows the decreasing trend from the equator to the poles.It mainly concentrated in the bay of Bengal and Indian Ocean waters near Indonesia and the Pacific warm pool area and equatorial convergence.On the coast of Peru,the west coast of the United States and other water vapor subsidence area,the value of the water vapor is low.Reanalysis data and observation inversion data consistency when reflect theoverall distribution of water vapor.Inversion water vapor products can be found after the daily variation of periodic analysis,much of the ocean of water vapor,generally has the very good cyclical change,cycle for 24 hours or 12 hours.We use data on ocean water inversion of the periodic analysis of daily variation of several typical water vapor.Results show that the water vapor in ocean generally has better cyclical when the diurnal variation of the water vapor peak in the local time evening or later at most of areas.Water vapor diurnal variation may be due to sea temperature wind speed,and the influence of solar shortwave radiation.