Way Forward

The combination of fallowing and substituting groundwater by surface water was effective in reducing aquifer depletion in Guantao. The average annual depletion rate after 2014 was about half the value of the pre-project period 2000–2013 and basically limited to the deep aquifer. The goal of closing all deep aquifer wells has only been reached partially, their use being necessary in locations where the shallow aquifer is too saline.

imports, adding more reliable surface water imports at time of use, and saving water through agricultural techniques certainly have the necessary potential (see analysis in Box 5.1).
In its work in Guantao, the World Bank's GEF project (Foster and Garduño 2004) stressed that only reduction of consumptive water use (i.e. transpiration and evaporation) is real water saving. This remains of course true, but looking at saving of deep aquifer water, the statement has to be modified. As there is no irrigation back flow or any other seepage back to the deep aquifer, any reduction of pumping is 100% saving of deep aquifer water, irrespective of whether it turns into evapotranspiration, seepage to the shallow aquifer, or drainage outflow.

M Main aquifer rehabilitation measures, their remaining potentials in Guantao and unit costs
Reduc on of groundwater abstrac on can be achieved by supply side measures, adding new water resources, and by demand side measures, reducing abstrac ons by water saving. Costs of both can be expressed by CNY per cubic meter supplied or saved. Demand side measures (darker shading in the table) are subsidy driven. Supply side measures consist of water imports from reservoirs, the Yellow River or -via the SNWT -from the Yangtze River Basin.

Cost/Subsidy (CNY/m 3 )
Tradi onal water saving* (small fields and low pressured pipe irriga on) 0.4 1.6 Highly efficient water saving* (drip irriga on in green houses) 0.7 2.5 Highly efficient water saving* (sprinklers) 0.8 2.5 Mulching* 0.2 1.0 Conserva on llage* (maize) < 1 <0.5 Conserva on llage* (wheat) 0.6 1.3 Rain fed cropping** 0.9 4.5 Water saving cul vars*** 4 1.9 Fallowing of winter wheat** 10 3.0 Yellow River and reservoir water (No increase, improvement of efficiency in use of exis ng imports only) < 15 0.4 SNWT Central Route (Replace remaining deep aquifer pumping for drinking water supply) 4 2.5 SNWT Eastern Route (Extension to Hebei under construc on) < 1 Box 5.1: Aquifer rehabilitation measures: remaining potential in Guantao and specific costs. The box shows two prominent items, improved used of water imports from the south and increased fallowing of winter wheat, which taken together are sufficient to close a gap of 20 Mio. m 3 /year. They are followed in size by water saving through new cultivars (4 Mio. m 3 /year) and numerous "small" measures of water conservation, also adding up to a potential of about 4 Mio. m 3 /year. Deep aquifer pumping by households and industry should be completely substituted by additional imports of 4 Mio. m 3 /year of SNWT water. All measures together (30 Mio. m 3 /year) should be sufficient to fill the deep aquifer gap in the coming years provided the continued commitment of the administration.
The success of the last 7 years is mainly due to the fallowing measures and the import of surface water. It does not mean that no more management of the shallow aquifer is necessary. On the contrary, only from this situation as a starting point the storage of the shallow aquifer can be managed actively, hopefully increased, as a storage device to overcome future drought years. The system in place is also important to react to more systematic changes in the future, be it a change in crop water demand due to rising temperatures or changes in recharge due to changes in rainfall and rainfall patterns. The decision support system can keep track and help in designing the right answers in an adaptive management approach charted out by the red line concept.
The metering of water use and the collection of fees either for amounts overstepping quota or for every cubic meter of water pumped certainly serves the purpose of ensuring general discipline in water use. It is however only justified if the fees lead to a reduction of pumping. This can happen either directly or indirectly by using the collected fees to subsidize water saving measures.
Metering by electricity consumption is practicable and makes economic sense. It should be developed further with the electricity company using their daily available "big data". It would be ideal if the water fee could be collected automatically together with the electricity fee saving the effort for a second fee collection system.
Much has been achieved in the last years and a combination of crop system change and South-North Water Transfer (SNWT) clearly has sufficient potential to stop aquifer depletion, not only in Guantao but in the whole of the NCP. But stopping the depletion is not enough. Even if a cone of depression is stabilized by surface water imports and a hydraulic quasi-steady state can be reached, such a closed system will salinize with time. Only if drainage outflows to the sea via canals, streams, rivers and the aquifer are restored one can truly speak of a sustainable solution. In future, the focus should turn to groundwater quality. The salinity of pumped water should be monitored at least once per year. The increased agricultural pumping in the deep aquifer of the last few years is probably already related to an increase in salinity in the shallow aquifer's cones of depression. While the end of over-pumping is within reach, the salinity problem will not go away and requires further careful management.

Agriculture and Future Groundwater Management in the NCP
Groundwater depletion in the NCP was a consequence of the governments' will to battle famine in the early 1980s. Hebei Province played an important role in feeding the country since then. It increased its yield by 2-3 times over the past 3 decades. The intensification of food production coincided with a time of decreasing rainfall, which magnified the consequences regarding groundwater levels.
The centralized governmental structure in China usually allows the Chinese leaders to mobilize a vast amount of resources to pursue a goal or solve any problem in a determined manner once an issue is confirmed as extremely important. If a problem is perceived as in conflict with other objectives of interest, any decision is essentially a trade-off. This is the case for groundwater management in the NCP because mitigating groundwater over-pumping may hurt national food security and farmers' income, which are both considered at least as important as environmental protection, if not more important. In 2021, the agricultural department again received the task of increasing crop production (http://www.xinhuanet.com/2021-01/03/c_1 126940702.htm).
Population will peak between 2025 and 2030 at about 1.5 billion , who want to have an ever-richer diet with more meat, vegetables, and fruit. Climate change will decrease the yield of some of today's cultivars due to higher temperature and increase the water demand of crops due to higher ET (see Sect. 5.5). Water scarcity is exacerbated by deteriorating soil and water quality. Climate change may also put a question mark to the water transfers from the South, which must maintain a reserve for its own growth. So, will the solutions envisaged today still hold in 2050 or does it take more? To approach this question, a look at some trends of today is helpful.

Transition from Smallholder Agriculture to Larger Farms
The first trend is the transition from the present smallholder agriculture to modernized larger-scale farming in the NCP. The smallholder family farms with a cropping area of a third of a hectare, are economically unattractive and have no future. The annual income from a winter wheat and a summer maize crop on that area is less than 3000 CNY per person , which compares unfavorably with the average annual income in China of about 30 000 CNY per person. Young people leave the countryside and look for jobs in the cities. Their support of their parents is usually more significant than the parents' own income from farming. Most people working in the fields are above 45 and the question must be asked, who will produce China's food in 2050 (see Box 5.2). The situation is expected to change dramatically in the next decade(s) due to the fast change in rural population structure. With it, agricultural production activities and organization forms will change substantially too, from smallholder farming to larger farms through land aggregation. Take Switzerland for example: For an economically viable farm of 4 persons in Switzerland a farm area of about 100 ha (300 times the area of a Chinese family farm or almost the size of the whole cropland of a village in the NCP) is necessary. It is typically managed by the family plus one additional farm hand, which is feasible due to a high degree of mechanization, applying modern technology both for remotely controllable hardware, including machinery and irrigation equipment, and for information technology harnessing information on plant health, weather and markets for example.
Chinese farming is developing in the same direction. Farm sizes in the NCP are already increasing. Big entrepreneurs lease land from individual farmers, sometimes of whole villages, to practice a larger scale and more efficient style of farming. This opens up new opportunities also for increased efficiency in water use. However, it is unclear how grain production can eventually become profitable for big farms relying on land leased from small farm households. With the current leasing cost (800-1000 CNY/mu/year), planting grain crops (wheat and maize) is not sufficiently profitable to attract farming enterprises to take over. Agricultural land ownership and the right to trade ownership in a more permanent way will become more and more a critical issue in this transition process.
Using engineering and agronomic approaches, larger farm units can fully exploit the potential for improving water use efficiency beyond today's level. While highly efficient center-pivot sprinklers are neither applicable nor affordable for today's small family farms, they can unfold their full water saving potential on farms of at least 100 ha in size. A large farm can also afford to fallow a certain percentage of its area in order to let the soil rest or to plant green manure without having to be pushed by government. Today's pioneers of big farming go into niche products such as plant seeds, health foods, medicinal plants, or farm tourism to optimize income. At the next stage, grain production should be considered for scale-up to a more efficient and profitable operation.
Greenhouses are the most profitable production units, even at the scale of a family farm. When equipped with soil moisture sensors and automatic drip-or microsprinkling equipment they can save significant amounts of water compared to the traditional green houses. At the same time, they can save labor, which meanwhile has become expensive in China. In the US, vertical farms are quickly expanding. They produce vegetables in hydroponic cultures close to the consumer. Their water use is minimal as most of the transpired water can be recycled. And they do not need any agrochemicals except nutrients. Economic viability relies on the consumers' demand for healthy food. China's urban elite is also conscious of food quality, which shows in collective ordering or "crowd farming" of farm products from trusted producers in the periphery.
Bigger farms allow a more scientific approach to crop production, often termed "precision agriculture", which benefits the quality of the produce as well as the environment. Drones already today apply crop protection products. In future they will be able to deposit those products on plants in need only. Similar technology can improve fertilizer application to conserve nutrients and reach more uniform production over the area.
The smart water meters, which have been successfully applied in China's West will also be sustainable in the NCP if pumping in big farms is concentrated on fewer wells. Management of water quota by a county's water resources administration will become feasible when it needs to address only a small number of large farms instead of thousands of small farms. Already during the project, the biggest farm of Guantao enthusiastically volunteered to cooperate with our metering experiment. They want to save water-and thus energy-as they are much more conscious of cost-efficiency than the smallholder farmers. Smart devices such as smart water meters with data transmission to a central server and eventually a feedback from that server to the device, mark the beginning of the Internet of Things (IoT) in agriculture.
Information technology and the full use of available information have to play a bigger role in farming. Alibaba launched the "ET Agricultural Brain". This digital tool enables farmers to raise their crop yields and income by leveraging big data. The app allows farmers to digitally record information about their yields in order to optimize their entire production cycle, raising efficiency and capacity. It was preceded by similar efforts in North America where for example Climate Corporation analyses 50 terabytes of weather data every day and sends the information to users via its Software-as-a-Service platform Climate.com. The corporation's goal is to leverage data to help farmers deal with the increasingly volatile weather caused by climate change.
Agricultural research in China is quite advanced, but it is slow in trickling down into practice. A recent example shows how it can be made available to farmers faster: Through a national campaign, about 20.9 million farmers adopted the recommendations of a study, which increased productivity and reduced environmental impacts. As a result of the intervention, farmers were together US$12.2 billion better off through savings in fertilizer and increase in yield (Cui et al. 2018). The new type of agriculture requires human resources with the relevant education. To feed China's population in 2050, a scientifically educated workforce with an agricultural background will be vital (Tso 2004).
Increase of production per ha does not automatically mean increase of production per cubic meter of water, i.e. more "crop per drop". Therefore, optimization of production must go in parallel with other options of reducing water use. Agrotechnology has still some other methods in store, which under the name of "new green revolution" will help to optimize the combined efficiency of water and nitrogen use (see Sect. 5.3).

Consumer Behavior and Food Demand
Water saving must include the reduction of food waste post-harvest, in retail and on the level of the consumer. In recent years Chinese propaganda campaigns focused on the reduction of food waste in restaurants. In Europe initiatives push smart logistics for internet-based distribution of food close to expiration date. Food saved is water saved.
Future water demand is closely related to the diet people desire (McLaughlin and Kinzelbach 2015). Even a stagnating population will further increase water demand if the present trend towards increasing consumption of meat and milk products continues. In the past 30 years China's meat consumption has grown 2.5 fold with a population increase of about 15%, and the trend towards more meat continues, with values admittedly still low in comparison to many western countries (see Box 5.3)

C China's Population
According to the Ins tute of Health Metrics and Evalua on (IHME) China's popula on will peak between 2020 and 2030 at a popula on of 1.5 Bio. and then gradually decease to 730 Mio. at the end of the century. This is much less than the predic on by the UN bureau of popula on. This development presents a big advantage for China's resource availability.
The age pyramid will also change significantly with the popula on over 50 being in the majority. The overaging and lack of work force can be balanced out by more mechaniza on e.g. in agriculture. . However, the trend in meat consumption could change over health concerns. Young people in the west turn to vegetarianism. While the number of vegetarians in Germany was estimated to be 0.6% in 1983, it increased to about 10% of the population in 2016, with typical vegetarians being between 19 and 29 years old (Mensink et al. 2016). The market adapted to this dietary shift and the associated new demand. Sales of meat substitutes are increasing rapidly, creating a new billion-Dollar market.
The diet plays an important part in determining the water demand in agriculture. Of course, its meat component impacts in the first place the production of feed grain, which grows in summer and presents a smaller potential for water saving than wheat. But substitutes for wheat could also be considered. Potatoes yield the largest number of calories with the least input of water per unit area. Of course, dietary changes need time for people to get accustomed to. But did China not get accustomed to wheat after centuries of millet? Potatoes, which were unknown before Columbus, are the 4 th staple food crop (after rice, wheat and maize) in China since 2015 (Xinhua 2015).
Possibly China may also get used to Quinoa, which is advocated by the UN in its efforts to enhance the globe's food security while saving water (FAO 2011).

C China's Diet
In 2026 an average Chinese ci zen will consume about 53 kg of meat per year, 10% more than in 2017. Pork is the favorite Chinese meat making up about 60% of all meat consump on. It is followed by poultry, beef and veal, and mu on and lamb. Due to the African Swine Fever epidemic, the consump on of pork declined a er 2018 and rebuilding the swine herd will s ll take some years.
Data source: h ps://data.oecd.org/agroutput/meat-consump on.htm China is s ll ea ng less meat than others, such as the USA with an average yearly meat consump on of 100 kg/capita or the OECD countries with an average of about 70 kg/capita. This indicates that there is room for expansion.

Drought Resistant and High Yield Wheat Varieties
Managing groundwater sustainably depends to a high degree on agricultural strategies adopted in the NCP. Pursuing high yield to feed people was the major agricultural target pursued during the past 4 to 5 decades. The NCP played an important role in improving grain production during this period, when numerous high yield varieties were bred and widely cultivated. The yields of wheat and maize increased considerably from 1980 to 2010 (Fig. 5.1). They can reach 9 t/ha for wheat and 10.5 t/ha for maize today. This is much higher than the respective national average yields of 6.2 t/ha and 5.4 t/ha (2019 data) and opens up a yield gap. While the development of crop varieties has helped to achieve the goal of self-sufficiency in agricultural goods, the yield gap still presents a reserve to be activated.
The achievements in grain production were accompanied by a tremendous increase in fertilizer and irrigation water inputs, with the well-known consequences for groundwater depletion. While over the last 40 years yield increase was the primary objective, now the reduction of agricultural intensity has become an essential option to mitigate groundwater level decline.
Increasing attention has been given to the applications of biologic approaches, such as breeding of drought resistant varieties, or genetic modification and gene editing to largely improve photosynthetic efficiency. According to 30-year field experiments at the Center for Agricultural Resources Research, Institute of Genetics and Developmental Biology, CAS, a wheat variety with high water use efficiency could save 30-50 mm of irrigation water compared to the normal varieties planted in NCP (Fang et al. 2017) with almost the same level of yield. The WUE could be improved by 21.6% maximally by adoption of highly water productive varieties for wheat. On average, adopting high WUE varieties could achieve an increase of grain yield by about 1% and an increase in WUE by 0.5% annually . A recent study revealed that nitrogen status affects rice chromatin function through the modification of histones, and some specific genes regulating tillering. The transcription factor NGR5 mediates hormone signaling and increased NGR5 levels can drive increases in rice tillering and yield without increasing nitrogen fertilization requirements (Wu et al. 2020). This kind of breakthrough in genetic or biological sciences can be expected to greatly improve water and nitrogen use efficiencies of grain crops. Therefore, another "green revolution" through genetic editing technologies may happen in the near future. It should result in new drought resistant crop varieties and varieties with higher photosynthetic ability, using less water and nitrogen fertilizer.

Importing Surface Water: Benefit and Risk
For regions lacking large rivers while suffering from severe water shortage, groundwater overdraft seems to be inevitable and importing surface water from outside of the region is the most direct and effective way to alleviate the decline of groundwater levels. Based on the experience from groundwater overdraft control in the NCP so far, importing surface water should not only increase the availability of water but also the efficiency of its use. Present use of surface water imports is extremely inefficient as the timing of water transfers depends on flows in the upstream, which are not synchronous with the irrigation calendar. Storage capacity available in the plain area is by far not sufficient to allow more efficient use of the imports. Improvement requires further updating of water infrastructure including the water system's connectivity and maintenance, e.g. by dredging river channels, renovating pools and ponds, as well as using imported water for MAR. Import of surface water has been planned, designed, and implemented rather well compared to other methods of groundwater over-pumping control so far. Although it seems irreplaceable in the struggle to end groundwater depletion in the NCP, it is still facing considerable risks and uncertainties.
The surface water quota from the Yellow River will still be strictly limited in the near future. Due to constraints on water quota, after three years of groundwater overpumping governance, almost all cities and counties have used up their quota of Yellow River water. However, some counties' task of cutting down on groundwater extraction has not been achieved. In Hebei Province Yellow River water has mainly been used for agricultural irrigation. If no further imports are possible, groundwater over-pumping governance can only rely on optimization of the cropping structure, which depends heavily on government subsidies. With the decline of these, the further promotion of over-pumping control action in these mainly rural areas remains challenging.
The high cost of water supply from the SNWT project brings risk and uncertainty to the groundwater over-pumping control system. One important reason is that Hebei provincial government can only spend a small portion of its whole budget for projects supporting the SNWT. 78% of the investment relies on bank loans. The construction of water treatment plants and connecting water supply pipelines is mainly supported by financing through non-governmental investors. These investors expect high returns on their investments. The above two factors result in high water supply and transfer costs for the SNWT. The water price at the entrance of the main canal of the SNWT Project into Hebei Province is 0.97 CNY/m 3 , the price of water entering the surface water treatment plant is 2.76 CNY/m 3 , and the water price for the consumers is 7.43 CNY/m 3 , which is 24% higher than in Beijing, and 34% higher than in Tianjin. It amounts to 155% of the current average water price in urban areas, and to 210% of the current county average water price of Hebei Province.
Due to its high price, water of the SNWT Project is not affordable for agricultural and ecological purposes. Even redundant SNWT water is too expensive to replace groundwater in agricultural irrigation and preserve the groundwater system.
If the present water price of the SNWT's Central Route of 2.5 CNY/m 3 is assumed, the water needed for irrigation of winter wheat would cost about 6 000 CNY/ha. This amount is close to the present input cost for planting one ha of winter wheat and would make wheat planting with SNWT water economically infeasible. Today, the irrigation cost of wheat with groundwater (including maintenance and labor) is about 1000 CNY/ha or 1/6 th of the total input cost. The cost of water of the Eastern Route will be less than that of the Central Route, but it cannot be cheaper than groundwater as it has to be pumped up by about 100 m from the Yangtze River, about twice the lift presently necessary for deep aquifer groundwater in the NCP. If the management of all irrigation water would be put into the responsibility of one big company, they could make an average price, which would dilute the high marginal cost of the transfer water to an acceptable prize for food production.

Climate Change in the NCP
Climate change influences the fate of groundwater resources in the NCP in many ways. Precipitation determines not only groundwater recharge. It also determines the need of supplementary irrigation and therefore groundwater abstraction. Its distribution in time, extreme rain events or drought events can be as important for agricultural production as the annual total precipitation. Temperature, wind speed and radiation influence the potential evaporation, which finally determines the evapotranspiration potential of the crops and thus the water requirements. Finally, climate change may affect the flow of rivers, which are now the source of water transfers to the NCP.
The NCP has seen climate change in the past. Observations since the mid-twentieth century show that it has become warmer and drier. Most observed downward trends in annual precipitation amount were statistically significant Fu et al. 2009;Liu et al. 2005;Sun et al. 2017;Wang et al. 2012). Summer precipitation, which accounts for 50-75% of the total annual rainfall in NCP, also has decreased significantly Sun et al. 2020;Wang et al. 2012;Ye 2014), while precipitation in other seasons has shown contrasting trends Sun et al. 2020). The decrease in rainfall led to a decrease in groundwater recharge. Significant increase was not only observed in the mean annual temperature but also in temperature during both wheat and maize seasons Liu et al. 2014;Zhang et al. 2015). Surprisingly, a long-term decreasing trend has been observed in actual evapotranspiration, which can be explained by decreasing trends in precipitation, sunshine duration and wind speed which overcompensated the increase in ET due to temperature rise alone (Cao et al. 2014;Chen et al. 2010;Fan et al. 2012;Liu et al. 2014;Song et al. 2010). A recent study based on remote sensing and physical modeling found that actual evapotranspiration increased slightly since 2000, but concluded that the contribution of climate change was less than that of human activities . The result agrees with the measurement data of Luancheng Agro-Eco-Experimental Station of the Chinese Academy of Sciences showing that the actual seasonal ET of winter wheat and summer maize under wellwatered conditions gradually increased from the 1980s to the 2000s possibly caused by the increase in leaf stomatal conductance associated with the introduction of new cultivars . With global warming, more crops can be grown in Northeast China, which has better rainfall, and the recent decades already have shown some shift in production from the NCP towards the Northeast, reducing the pressure on the NCP.
The Fifth Assessment Report by the Intergovernmental Panel on Climate Change (Pachauri et al. 2014) provides the analysis of model outputs from the fifth phase of the Coupled Model Intercomparison Project (CMIP5). CMIP5 produced a state-of-theart multi-model dataset designed to advance our knowledge of climate variability and climate change including "long term" simulations of twentieth-century climate and projections for the twenty-first century and beyond for a number of Representative Concentration Pathways (RCPs) for atmospheric greenhouse gas (GHG) concentrations (Van Vuuren et al. 2011). Annual averages of temperature and precipitation for two RCPS and a time interval at the end of the century are shown in Fig. 5.2. Looking at Eastern China for the more pessimistic RCP8.5, one can find a strong increase in yearly average temperature (5-7°C) and an increase in annual rainfall (10-20%) compared to the baseline of 1986-2005. For a lower emission scenario, annual rainfall hardly changes, while average temperature increase is about 1°C.
To adapt these global projections to regions, a process called downscaling is required to obtain a higher spatial resolution suited for the scale of the region. The problem with the projections is their uncertainty. While different models usually agree well as far as the projected temperature rise is concerned, predictions of rainfall are extremely uncertain reaching from increase to decrease of rainfall. This is also true for the NCP, where the ensemble of climate models shows discrepancies among GCMs and downscaling methods, concerning the projected mean and extremes of precipitation. Some studies show a likely increase in annual precipitation but at varying rates (Fu et al. 2009;Tao and Zhang 2013), while others project a decline in annual precipitation . Being influenced by several climate variables, evapotranspiration estimates turn out to be very uncertain, making it difficult to draw conclusions about future trends. The projected changes of evapotranspiration during the wheat or wheat-maize growth period in the NCP could either increase or decrease within a range of around ± 10% depending on the GCM and greenhouse gas emission scenario employed (Guo et al., 2010;Lv et al. 2013;Mo et al. 2013;Tao and Zhang 2013).
We can distinguish direct and indirect impacts of climate change on groundwater resources. Changes in the amount, intensity and frequency of precipitation directly determine groundwater recharge and indirectly influence the demand of groundwater pumping for irrigation. The direct impacts of changes in precipitation on groundwater recharge have been found to contribute less to groundwater resources than the changes in pumping, not only in the NCP but also in other regions (Larocque et al. 2019;Li et al. 2014). However, it should be noted that studies on climate change assessments usually ignore the indirect impacts of precipitation on groundwater pumping, and  (Pachauri et al. 2014). The number of models used to calculate the multi-model mean is indicated in the upper right corner of each panel. Stippling (i.e., dots) shows regions where the projected change is large compared to natural internal variability and where at least 90% of models agree on the sign of change. Hatching (i.e., diagonal lines) shows regions where the projected change is less than one standard deviation of the natural internal variability these indirect impacts are usually connected to human activities. Similarly, evapotranspiration also directly influences groundwater recharge and indirectly determines groundwater pumping by crop water demand. Previous studies in the NCP focused on the influence of projected evapotranspiration on future crop water demand and crop yield (Tao and Zhang 2013;Xiao et al. 2020;Xiao and Tao 2016). Hardly any model takes into account the decrease of plant transpiration due to the impact of CO 2 on stomatal conduction, which also modifies estimates of irrigation water demand (Guo et al. 2010).
While an increase in transpiration will decrease recharge and instigate more pumping due to higher plant water demand, there are three mechanisms, which in other regions of the world have shown to increase groundwater recharge: The first is the increase in extremes such as strong rainfall events (Fischer and Knutti 2019). Recharge is not a linear function of rainfall. While small rain events may give zero recharge, large events lead to over-proportional recharge. The second is the decrease of average wind with global warming. As reported above, this trend has been observed in China over the last decades. If it continues, it will reduce the expected increase in evapotranspiration due to temperature increase. The third mechanism is due to the increase of winter precipitation also quoted from literature above. At warmer winter temperatures, the soil freezing depth will be reduced and more recharge will occur in the cold season when evaporation is low, especially when there is thawing snow.
To get a coherent picture for the trends in evapotranspiration and the resulting crop water demand, we did our own assessment. The results presented in Box 5.4 show that overall, the crop water deficit will increase by mid-century and increase even further by late century. The compensation by decreasing wind speed as in the past decades is not seen anymore. That means the second mechanism mentioned above will most probably not be available.
Given all the interactions and assuming that total annual precipitation will not decrease, recharge will most probably not change much compared to today. But, due to more extremes both wet and dry, the use of the aquifer as a buffering device is of growing importance in the future.
The water resources availability in the NCP is not only determined by the development in the NCP itself. As the North in recent years relies increasingly on water imports from the South, the influence of climate change on the water availability in the catchment of the Han River and the Yangtze River in general is also a relevant influence factor.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.