通过定性分析与定量计算相结合的方式探寻生活用水强相关因素，充分考量区域气候条件及经济发展水平差异，构建全国及七大地理分区的城市居民家庭生活需水函数，各区域需水函数模型均通过了显著性检验，拟合值与实际值误差在10%以内的城市占61.5%。分区城市居民生活需水函数可客观反映经济发展阶段特性和气候背景下的用水差异，提升了模型的应用性和可操作性，为分区阶梯定额的确定提供技术支撑。应用模型对省会及重要城市的居民生活用水定额和水价进行分析预测，结果表明：在同等居民收入和水价增幅下，北方城市居民生活用水定额降幅远低于南方城市，南方城市价格增长对居民生活用水定额的抑制作用更为明显；随着收入的增长，需要适度提升水价才能使规划水平年生活用水定额维持现状水平。

In the context of the prominent contradiction between supply and demand in China and the rapid growth of urban domestic water,reasonable control of the excessive growth of urban domestic water is the key to regulating economic and social water use and promoting sustainable social development.The tiered water pricing system can alleviate the trend of water stress through price leverage.However,the current domestic water quota standards for urban residents are generally too high,which can not effectively guide the implementation of the tiered water pricing system.The current water quota is determined concerning the relevant national and local quota standards.Although it has wide applicability,it is not typical and representative.How to determine a reasonable quota standard based on fully complying with regional differences is worthy of in-depth consideration. Through the combination of qualitative analysis and quantitative calculation,the strong correlation factors are explored that affect the domestic water quota.Pearson correlation analysis can explore the correlation between indicators from cross-sectional data.In addition,qualitative analysis can help for in-depth study of the law of urban domestic water use.Based on the full consideration of regional climatic conditions and economic development level,the model selected factors such as water price,per capita disposable income,and temperature as explanatory variables for urban household water quotas.The 2017 data of 293 prefecture-level cities across the country are used as a sample to construct a water demand function model.To better seek regional common features,the seven geographical regions of China were modeled separately.During data processing,the obvious centrifugal data and marginal data are excluded,and the number of valid samples accounts for about 98% of the total data volume.Then,a significant test on the constructed water demand function model is conducted to predict a reasonable water quota. The results show that the fit of the national and regional models is good,the coefficient of determination R2 of the model is above 0.5,and the corresponding significance level is much less than 0.05.The model is effective through the significance test.The model-fitting results are good.Cities with a model fitting deviation rate of less than 10% account for 61.5% of the effective samples,and cities with a deviation rate of less than 20% account for 86.5%.From the perspective of the zoning fitting effect,the deviation rate of north China and southwest region is mainly controlled below 10%,and the deviation rate of some cities in south China and east China region exceeds 20%.Among the three explanatory variables of the model,although the income variable conforms to a certain trend of change,with the increase of instability and uncertainty in the current international environment,the difficulty of income forecasting increases.The temperature variable has obvious regional characteristics.Although the fluctuation is small in a short period,as the risk of climate change intensifies.The temperature prediction method should be improved accordingly.Water price variables are affected by unpredictable policy factors,with the highest degree of uncertainty.The prediction model constructed according to geographical divisions reflects the common characteristics of the region.It is foreseeable that some cities have a large deviation rate when using the model to predict,but the model is still effective from the perspective of the region as a whole.In addition,it is found that the cities with high current per capita domestic water quotas are mostly located in the high water areas of southern China and the middle of China or near the water source areas.This part of the city has great water-saving potential.Under the guidance of water-saving awareness and policy pressure,the water-saving potential can be gradually released as a water-saving effect.By analyzing the elasticity coefficient of the model,it is found that in the era of rapid economic development in the country,although water prices and income are increasing at the same time,the impact of income on water demand is much greater than that of prices.Income growth will have a significant effect on living standards and water demand,but the price increase policy assisted by effective policies is still an important means to restrict water use.Finally,by setting the water consumption quota under the expected water fee expenditure and the water price under the expected quota standard,it can be found that,in 2025,the domestic water quota for residents in cities lacking water in the north will maintain the current level or slightly decline.However,the domestic water quota of cities with abundant water in the south has shown a significant downward trend,which means that the increase in prices in southern cities has a more obvious inhibitory effect on the domestic water quota of residents. To a certain extent,this shows that the rigid water consumption of residents in northern cities accounts for a relatively high proportion,while the enjoyment and luxury water consumption in southern cities accounts for a relatively large proportion.The results showed that:At annual scale,TRMM and CMORPH show higher detection accuracy than CHIRPS and PERSIANN_CDR,and terms of the spatial distribution characteristics of annual precipitation,only TRMM,CMORPH and CHIRPS could reflect its spatial distribution to varying degrees;At a monthly scale，TRMM and CMORPH still have high accuracy in precipitation estimation,and TRMM,CMORPH,and CHIRPS have different advantages in describing the spatial distribution of precipitation in different regions in YRB;At daily scale,the four satellite-based precipitation products have a strong ability to capture weak precipitation but have a poor ability to detect heavy precipitation;At spatial scale,the accuracy of the four types of satellite-based precipitation data is higher in high-altitude areas at daily and monthly scales,and there is no obvious change law at the annual scale. In general,among the four types of satellite-based precipitation products,TRMM and CMORPH have more advantages in different time scales in YRB,but the spatial distribution of the four types of products is significantly different in different regions and different time scales.This conclusion lays the foundation for the further development of high-temporal-resolution precipitation fusion products suitable for YRB.