【正文】 China can be found by the National Bureau of Statistics .With these data, the function about the both two and time can be fitted, and water consumption、water supply of various parts of China in 2025 could be obtained. Thus the dry zones and water shortage could be determined. With too many provinces and cities, and too much data, it is better to we split China into a few zones to study. As to the problem of water shortage, it needs to raise solutions. In general, storage、mobilization、 desalination, sewage treatment and other measures are the mon replenishment measures used by the government. In order to quantitatively study these programs, it needs to understand the specific process of each program, as well as each cost it produced. Among them, the diversion measures have been identified. To determine the water diversion route, taking the water mobilization costs into account, in view of the extremely uneven distribution of water resources in China, the best diversion route needs to be determined through the method of linear programming. In addition to water supply, reducing the demand for freshwater is also one of the ways, such as increasing the water prices, using agricultural irrigation drip technology, etc. They are all effective measures. On the other hand, the government should also improve the national watersaving awareness. Each of these measures has its own costs, economic benefits and the environmental impact. When determine the water strategy, it is necessary to using priorities to make decisions. So it needs to use the analytic hierarchy model to assess and rank the various schemes. When determine the best solution bination, it also needs to considerate its local conditions and its feasibility. Finally, it needs to fully pensate for the water shortage, and determine the best water strategy in 2025. 3. General Assumptions In order to have a better study on this paper, we simplify our model by the following assumptions: 1. There is no additional artificial water supply except our determined water strategy from 2013 to 2025. 2. There are no droughts、floods and strong geological disasters from 2013 to 2025. 3. Water transfer line is straight, and water diversion costs per kilometer ignore the impact of the terrain and the environment. 4. Desalination plants are all offshore, without considering the seawater transportation costs. 5. Do not consider increased freshwater for greenhouse effect. 6. Do not consider the effect of price changes. 4. Symbols and Definitions In this section, we will give some basic symbols and definitions in the following for convenience. Table Definition Variable Symbols Definition Unit P population M GDP 108yuan N Water demand of agricultural irrigation 104m3 t time year W Total water mobilization costs yuan C Desalination costs yuan N Pollution Equivalent 5. Mathematical Models . Prediction model Water demand is determined by the regional freshwater supply – demand balance, before discussing the water demand in 2025, the water demand and water supply need to be predicted firstly. . China partition according to the water resources situations Partition principles： I. Try to take care of the integrity of the administrative divisions, which is convenient for collection and statistics of the data. II. Each partition has the same one level of annual precipitation, and each partition has a relatively consistent water shortage situation. III. Try to ensure that the coastal areas could be divided together as much as possible, to be convenient for subsequent desalination processing. China has 32 provinces and municipalities, which can be divided into the North China, East China, northeast of China, southern part of China and Northwest according to the geographical location. And the annual precipitation also showed a decreasing grading from southeast to northwest. With prehensive considerations of location、climate、 annual precipitation and other factors, China was divided into seven areas A zone、B zone、C zone、D zone、E zone、F zone、G zone, for the convenience of our study. And we have studied respectively the water shortage situations of the seven regions： Specific dividing circumstances are as follows: Figure 1：China partition A zone： Tibet、 Qinghai、 Xinjiang B zone： Gansu、Inner Mongolia, Shaanxi, Shanxi and Ningxia C zone： Sichuan、 Yunnan、Guizhou and Chongqing D zone：Hunan、 Hubei and Jiangxi、 Anhui、Henan E zone: Fujian, Zhejiang, Shanghai and Jiangsu, Guangxi, Guangdong and Hainan F zone: Beijing and Hebei, Shandong, Tianjin G zone: Jilin, Liaoning and Heilongjiang of water demand： The demand of fresh water is constrained by the following several factors： Figure 2: Influencing factors on water demand Among them, population, GDP and agricultural irrigation have greater impacts. In view of those too much factors, it is hard to inquiry all the regional annual data of all those factors, so three important factors – population, GDP, the agricultural irrigation, have been chosen to research on the usage quantity of freshwater. Namely freshwater usage quantity is a function of such three factors – population、GDP and agricultural irrigation. Then their relationship can be explored using multiple regression analysis. 1. Multivariate regression analysis Take A zone for example. Water resources data of each province and city can be found from National Bureau of Statistics. Integrating the data of all the provinces and cities in A zone by Excel（Tibet、 Qinghai、 Xinjiang），the statistics results can be gotten as the following table: Table 2: Multivariate regression Year G zone/ 10 8m3 Population/104 GDP/108yuan Agriculture/108m3 2003 2004 2005 2006 2007 2008 200