Hydrogeol J DOI 10.1007/s10040-017-1574-4
REPORT
Modelling the response of shallow groundwater levels to combined climate and water-diversion scenarios in Beijing-Tianjin-Hebei Plain, China Xue Li 1,2,3 & Si-Yuan Ye 1,2,3 & Ai-Hua Wei 4 & Peng-Peng Zhou 5 & Li-Heng Wang 5
Received: 18 July 2016 / Accepted: 8 March 2017 # Springer-Verlag Berlin Heidelberg 2017
Abstract A three-dimensional groundwater flow model was implemented to quantify the temporal variation of shallow groundwater levels in response to combined climate and water-diversion scenarios over the next 40 years (2011– 2050) in Beijing-Tianjin-Hebei (Jing-Jin-Ji) Plain, China. Groundwater plays a key role in the water supply, but the Jing-Jin-Ji Plain is facing a water crisis. Groundwater levels have declined continuously over the last five decades (1961– 2010) due to extensive pumping and climate change, which has resulted in decreased recharge. The implementation of the South-to-North Water Diversion Project (SNWDP) will provide an opportunity to restore the groundwater resources. The response of groundwater levels to combined climate and water-diversion scenarios has been quantified using a groundwater flow model. The impacts of climate change were based on the World Climate Research Programme’s (WCRP’s) Coupled Model Intercomparison Project phase 3 (CMIP3) multi-model dataset for future high (A2), medium (A1B), and low (B1) greenhouse gas scenarios; precipitation data
* Xue Li
[email protected]
1
Qingdao Institute of Marine Geology, Qingdao 266071, China
2
Key Laboratory of Coastal Wetland Biogeosciences, China Geological Survey, Qingdao 266071, China
3
Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266071, China
4
School of Water Resource and Environment, Hebei GEO University, Shijiazhuang 050031, China
5
Key Laboratory of Shale Gas and Geoengineering, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China
from CMIP3 were applied in the model. The results show that climate change will slow the rate of decrease of the shallow groundwater levels under three climate-change scenarios over the next 40 years compared to the baseline scenario; however, the shallow groundwater levels will rise significantly (maximum of 6.71 m) when considering scenarios that combine climate change and restrictions on groundwater exploitation. Restrictions on groundwater exploitation for water resource management are imperative to control the decline of levels in the Jing-Jin-Ji area. Keywords Climate change . Groundwater flow . Water diversion . Numerical modelling . China
Introduction The Jing-Jin-Ji area is one of the most important socioeconomic and agricultural regions in China. It contains approximately 8% of China’s total population and contributes approximately 11% of the GDP; however, these developments have been heavily dependent upon 1% of the total water resources. The current per capita water resources in this region are approximately 286 m 3/year, which represents only 1/32 of the world’s average amount of water use and is much less than the 500 m3/year that is considered to be the minimum for life (Falkenmark et al. 1989). Due to the scarcity of surface water, groundwater has become an important water resource and contributes 75% of the total annual water supply. Because water demand has exceeded the naturally renewable supply, this region has begun to experience regular water shortages, and overexploitation of the groundwater resources has led to adverse environmental effects such as groundwater drawdown cones and subsidence (Zhang et al. 2009;
Hydrogeol J
Zhu et al. 2013). Groundwater levels have decreased at an average rate of 0.36 m/year since 1960 based on numerical simulation (Zhang et al. 2013; Li et al. 2014), more than 20 cones with a total area of more than 70,000 km2 have formed, and subsidence has occurred over an area of more than 90,000 km2. The water crisis has restricted the integrated development of the Jing-Jin-Ji area. The implementation of the South-to-North Water Diversion Project (SNWDP) will supply imported water to supplement the water resources of this region and decrease the amount of groundwater pumping. This project thus provides an opportunity to restore the groundwater levels. Furthermore, climate change is occurring in China, including the Jing-Jin-Ji area (Ding et al. 2007; Qian and Zhu 2001), and is causing significant impacts on the hydrological environment (Chen et al. 2006; Piao et al. 2010)—for example, the average annual mean surface air temperature in China has increased by 1.1 °C over the past 50 years and by 0.5–0.8 °C over the past 100 years, which are slightly more than the global temperature increase for the same periods (Ding et al. 2007). Since the mid-1950s, precipitation in China has increased slightly, which is consistent with the global trend (Ding et al. 2006). Interdecadal variability and obvious trends which have recorded warmer and drier climate over the last four decades have been detected in North China. The annual precipitation of the North China Plain has decreased by 6.7% over the past 40 years (1958–1998), and the annual means of the daily mean, maximum, and minimum temperatures have increased by 0.838, 0.188, and 1.468 °C, respectively (Fu et al. 2009). Extremely large flood disasters occurred in Yangtze River basin in the summer of 1998. These symptoms of climate fluctuations have had impacts on the water resources. Groundwater, which is more stable than surface water, will be subjected to significantly increased demand due to climate change. Therefore, studies of the groundwater response to combined climate and water-diversion scenarios are necessary for the sustainable development of water resources. Over the last decade, large amounts of research have been published on how climate change might influence different aspects of the hydrological cycle (Burn 1994; Kabiri et al. 2015; Middelkoop et al. 2001; Vano et al. 2015). The majority of this research has focused on surface-water impacts; the limited research on the effects on groundwater is largely area-specific due to the significant influences of local geology, land use, and topography (Srivastava 2013). Climate change may have a direct impact on groundwater levels (Hao et al. 2008), but current studies have focused on the impact on recharge and discharge conditions instead of the groundwater levels (Bouraoui et al. 1999; Jyrkama and Sykes 2007; Scibek and Allen 2006; Waibel et al. 2013). There are still many
publications about uncertainties from global climate models and regional climate models (GCMs-RCMs) and downscaling (Crosbie et al. 2011; Goderniaux et al. 2011; Kurylyk and MacQuarrie, 2013; Stoll et al. 2011). Crosbie et al. quantified the relative uncertainties inherent in projections of future recharge contributed by multiple GCMs, downscaling methods and hydrological models at three locations; they concluded that impact studies should use multiple GCMs and give careful consideration to the choice of downscaling methods. About downscaling, most climate change impact studies on groundwater resources use the simple Bperturbation^ or Bdelta change^ method (Prudhomme et al. 2002). The method applies Bchange factors^ calculated as the difference (relative or absolute) between the control and future climate model simulations, to observed climatic data. Previous studies that used historical precipitation data as future climate scenarios for planning purposes are not suitable in cases in which the climate of a region has already changed (Risbey et al. 2007). To address such concerns, GCM outputs should be applied; furthermore, there is limited information on the impacts of combinations of climate change and water diversion scenarios. The primary objective of this report was to investigate the impacts of climate change on shallow groundwater levels. The study also analysed the impacts of combined climate and water-diversion scenarios on the water table by performing groundwater modelling for different scenarios.
Study area and hydrogeological setting The Jing-Jin-Ji Plain covers an area of more than 80,059 km2 in eastern China. It is bordered by the Taihang Mountains, Yan Mountains, Bo Sea, and the Zhang River to the west, north, east, and south, respectively (Fig. 1a). The landforms are typical of a plains landscape, and the topographic elevations are less than 100 m. From west to east, the region can be divided into three principal zones: the piedmont, central plains, and coastal plains. The region has a semi-arid continental monsoon climate with annual average precipitation of approximately 500–600 mm and annual potential evaporation of 1,100–2,000 mm. Due to decreasing precipitation and upstream reservoirs retaining runoff, most of the rivers are dry or become ephemeral streams in the flood season (Zhang et al. 2009). The stratigraphy of the study area consists mainly of unconsolidated Quaternary sediments, in which most of the groundwater is deposited. The aquifer structure varies from a single aquifer composed of gravels and pebbles in the upper parts of the piedmont fan in the west to multiple aquifers composed of sand, silt and clay in the east. Based on this stratigraphy, the groundwater system is divided into four aquifer groups (I, II, III, and IV). Aquifers I and II are shallow
Hydrogeol J Fig. 1 a Location and model boundaries of the study area, b hydrogeological cross section of the study area modified from Chen (1999)
aquifers, whereas III and IV are deep aquifers (Chen 1999; Fig. 1b). Recharge results primarily from precipitation infiltration, lateral flow in the mountains and return flow from irrigation, while discharge primarily occurs through phreatic evaporation
and exploitation. The groundwater flows from the north and northwest to the east under natural conditions; however, groundwater flows into the centres of the cones, and shallow water flows into the deep aquifers in the centre and eastern parts of the plain due the influence of exploitation.
Hydrogeol J
Groundwater flow model Conceptual model The groundwater system was generalized into a threedimensional (3D) heterogeneous anisotropic transient flow model. The study area is shown in Fig. 1a. Based on the aquifer structure, the model of the complex multilayer aquifer system included three aquifer groups and two aquitards. Aquifers I and II are combined into one aquifer group that represents the shallow aquifer system because there are several incomplete aquitards, and many pumping wells enhance the hydraulic connection between them, whereas aquifers III and IV represent the deep aquifer system. Aquitards are located between the aquifer groups. The northern and western boundaries along the mountains were defined as specified flux boundaries. The eastern boundary along the Bo Sea was defined as a constant head boundary for the shallow aquifers and a no-flow boundary for the deep aquifers. The southern and southeastern boundaries were treated as flux boundaries and were adjusted manually during model calibration. The top boundary was the water table, where the exchange occurred between the groundwater and the other systems through precipitation infiltration and phreatic water evaporation. The bottom of the system was considered to be impermeable because it is composed of loam and clay beneath the Quaternary strata. Precipitation infiltration is the dominant source of groundwater recharge in the study area. It is affected by the amount of precipitation and the infiltration coefficient. The monthly gridded precipitation product from the China Meteorological Data Sharing Service system, which has a resolution of 0.5° × 0.5°, was used in the model, and the monthly precipitation was fully distributed across the study area. The infiltration coefficient was based on a previous study in which the variance of the coefficient was explained by the lithology and the depth of the water table (Meng et al. 2013). In the model, the infiltration coefficient was spatially distributed within three zones, ranging from 0.17 to 0.25. The precipitation recharge area was then divided into 51 zones based on the monthly precipitation and infiltration coefficients. Lateral flow, which is from the north and west, is a significant source but is also poorly quantified (Kendy et al. 2004). The recharge rates from lateral flow were based on previous studies in which they were estimated by the profile method (Shao et al. 2013) and hydrological analysis (Chung et al. 2010; Zhang and Li 2013). The recharge rate from lateral flow in previous regional water resource evaluations between 1991 and 2003 using the profile method (1.533 billion m3/year; Zhang et al. 2009) was applied in this model and adjusted manually during model calibration.
Groundwater is the primary source of water for irrigation, accounting for 81–93% of the total irrigation water (Zhang et al. 2003, 2013). Irrigation return flow is significant because of its large amount. No detailed data of the total amount of irrigation were available, and the amount was estimated based on information on the ratio between the amount of irrigation water and the abstracted groundwater for agriculture (Zhang et al. 2012). The infiltration coefficient of irrigation was smaller than that of precipitation considering land use because the evaporation is larger during the irrigation season. In this study, irrigation infiltration was also distributed in 51 zones like it was done for precipitation recharge, and the recharge from precipitation and from irrigation return flow were added to form the total recharge. The leakage from the surface water decreased after the 1960s, and by the 1980s, the rivers had nearly dried up (Xue et al. 2010; Zhang et al. 2009). Only during the flood season do ephemeral streams have an effect on recharge rates, but according to groundwater resource evaluations, the river leakage recharge from 1991 to 2003 was 0.742 billion m3/year and accounted for 5.3% of the total recharge which is a small value, so the effect is very little. In the model, the recharge from ephemeral streams in the flood season was incorporate into total recharge. Under predevelopment conditions, groundwater discharged to rivers; therefore, rivers were gaining. Rivers only became losing after significant groundwater depletion since the 1980s; however, by this time, most of the surface water was dammed. Due to decreasing precipitation and upstream reservoirs retaining runoff, most of the rivers have become dry or become ephemeral streams in the flood season (Zhang et al. 2009). In the Modular Three-dimensional Finite-difference Ground-water Flow Model (MODFLOW), the calculation of the evaporation from the water table requires the evapotranspiration rate and the extinction depth. Evapotranspiration rate is the rate of evapotranspiration as it occurs when the water-table elevation is equal to the top of the grid cell elevation. Extinction depth is the depth below the top of grid cell elevation where the evapotranspiration rate is negligible. When the water table is at or above the ground surface, evapotranspiration loss from the water table occurs at the maximum rate specified by the user. When the elevation of the water table is below the extinction depth, evapotranspiration from the water table is negligible. Between these limits, evapotranspiration from the water table varies linearly with water table elevation. Evapotranspiration rate was generated by Thiessen Polygons using daily potential evaporation data from four meteorological stations provided by the China Meteorological Data Sharing Service system based on the phreatic evaporation coefficient which is decided by the characteristics of the vadose zone. The extinction depth was set at 4 m everywhere (Zhang et al. 2009).
Hydrogeol J
Artificial discharge from the aquifers occurs through pumping by more than 600,000 wells (Fei 2006; Zhang et al. 2009). The exploitation data were taken from the Hydrological Bureau of each municipality and previous field research (Fei 2006; Xu 2006; Yang et al. 2011; Zhang et al. 2000, 2009). Detailed information about the locations and depths of the wells was not available, but the layers of exploitation were known.
Numerical model The groundwater flow model was developed using MODFLOW (McDonald and Harbaugh 1984). MODFLOW includes a set of stress packages that allow the simulation of external flow stresses, such as wells, areal recharge, evapotranspiration, drains, and rivers (Harbaugh 2005; Harbaugh et al. 2000). The grid consisted of 334 rows and 225 columns with a cell size of 1.5 km × 1.5 km. Based on the hydrogeological conditions and data about the porous aquifers at the study site, the model was discretized into five layers. Layer 1 included aquifer groups I and II, layer 3 represented aquifer group III, and layer 5 included aquifer group IV. Layers 2 and 4 were aquitards. Due to data limitations, the simulation period extended from January 2004 to December 2005. Each calendar month was a stress period with constant source and sink terms. Contour maps for December 2003 were used as the initial head. All of the sources and sinks were input using MODFLOW packages such as Well (WEL), Recharge (RCH) and Evapotranspiration (EVT). Hydrogeological parameters such as the hydraulic conductivity, specific yield, and storage coefficient were input by subarea (Zhang and Fei 2009). Fig. 2 Fitting of simulated and observed groundwater flow fields in 2005: a shallow aquifer and b deep aquifer
Model calibration The calibration period extended from January 2004 to December 2004, and validation is from January 2005 to December 2005. The calibration process was divided into two parts: fitting of the observed and simulated groundwater flow fields and comparison between the observed and simulated hydrographs at typical observation wells (Shao et al. 2013). The simulated flow fields were similar to the observed water level contours in 2005 for the shallow and deep aquifers (layers 1 and 3; Fig. 2). A total of 232 observation wells were selected for comparison of the groundwater hydrographs. The goodness of fit between simulated and observed water levels in all monitoring wells is presented in Fig. 3. The mean error (ME) between the simulated and measured water levels for the 166 shallow observation wells (layer 1) was –0.872 m, and the root-mean-square error (RMSE) was 5.233 m. The ME and RMSE were 0.253 and 7.874 m, respectively, for 66 deep observation wells (layer 3). Among all the observation wells, the comparison between the simulated and observed water levels in six observation wells was shown in Fig. 4; therefore, the simulation results reflected the actual groundwater flow field and the features of the groundwater regime. The parameters including hydraulic conductivity and storage parameters were adjusted during the calibration manually. The variations of the parameters were consistent with the actual hydrogeological conditions in the model parameter zones with hydraulic conductivities of 1–200 m/day for the shallow aquifers and 0.3–35 m/day for the deep aquifers (Figs. 5 and 6; Table 1) An equilibrium analysis of the transient groundwater flow model was used to compute the equilibrium of groundwater in the Jing-Jin-Ji Plain from 2004 to 2005. The recharge values
Hydrogeol J Fig. 3 Observed vs. simulated hydrographs at a 166 shallow observation wells, and b 66 deep observation wells. Dotted lines indicate the 95% prediction intervals
for the entire groundwater system were 13.274 billion m3 in 2004 and 13.076 billion m3 in 2005, and the differences between the recharge and discharge were –3.567 billion m3 in 2004 and –2.431 billion m3 in 2005. These results suggest that the groundwater recharge was less than the groundwater discharge for several years, which has caused the continuous reduction of groundwater storage.
Scenario analysis and discussion Through the model calibration, a scenario analysis based on the model was performed to predict and evaluate the
groundwater levels for the next 40 years (2011–2050). Because the shallow aquifers were more susceptible to climate change than the deep aquifers, only the shallow water levels were analysed. The study focused on precipitation changes and their direct effects on the groundwater levels for two reasons. On one hand, the water table depth of most of the area is 4–50 m; the area with water-table depth greater than 50 m accounts for 0.53% of the total area, and less than 4 m accounts for 18.99% according to observation data (Zhang et al. 2009), so phreatic evaporation is not very large considering the water table is relatively deep. On the other hand, precipitation is the dominant source which accounts for 86.3% of the total groundwater recharge in the study area based on
Fig. 4 Observed vs. simulated hydrographs at observation wells: a–d shallow aquifer and e–f deep aquifer
Hydrogeol J
Fig. 5 Hydraulic conductivity: a shallow aquifer and b deep aquifer
groundwater budget analysis, and exploitation is the dominant sink which accounts for 83.2% of the total groundwater discharge; that is to say, the phreatic evapotranspiration accounts for only a little in the total discharge and is very small compared to the precipitation infiltration, so the groundwater level changes caused by changes in temperature and evaporation over the next 40 years will be small compared with those caused by precipitation. Five scenarios were developed based on these assumptions.
Scenarios Scenario 1: baseline scenario The baseline scenario was simulated to analyse the water table of the study area under the present climate and the intensity of human activity in 2011–2050. The present climate for 1971– 2000 was used as the baseline input for comparison with the other scenarios.
Fig. 6 Storage parameters—specific yield (Sy) controls movement of phreatic water while specific storage (Ss) controls movement of confined water: a shallow aquifer and b deep aquifer
Hydrogeol J Table 1
Summary of model parameters of aquifers in the model
Model layer
Parameters Hydraulic conductivity (m/day)
Storage parametersa
Infiltration coefficient of precipitation
Infiltration coefficient of irrigation
Extinction depth (m)
1
1–200
0.065–0.2
0.18–0.25
0.17–0.24
6
3
0.3–35
0.000027–0.00023
–
–
–
5b
0.3–35
0.000027–0.00023
–
–
–
a
Specific yield (Sy) for layer 1; specific storage (Ss) for layers 3 and 5
b
No data are available for parameters in layer 5; therefore, parameter values for layer 3 were used for layer 5; layers 2 and 4 are aquitards
Scenarios 2–4: climate change scenarios The most widely used methodological framework to assess the impact of climate change on a catchment uses a limited number of global or regional climate model outputs as follows. First, scenarios that describe the future climate of the catchment are derived using climate model outputs either by applying them directly to the catchment or by downscaling them (Wilby and Wigle 1997; Crosbie et al. 2010). Second, these scenarios are run to derive future time series of the catchment’s state variables. Changes are calculated by comparing the indicators derived from these future series with the same indicators derived from modelled historic or baseline series (Jackson et al. 2011). This study used the weighted average precipitation calculated by the reliability ensemble averaging (REA) method (Xu et al. 2010) using the CMIP3 data that were obtained from the Beijing Climate Center. A delta change approach was used. The projected precipitation changes for 2011–2050 under climate change scenarios A1B, A2, and B1 (IPCC 2007) were derived by calculating the difference between the GCM’s simulated baseline (1971– 2000) and future (2011–2050) climate variables. The projected precipitation change data were a yearly gridded product with a resolution of 1.0° × 1.0°. First, a delta change of precipitation was achieved with the resolution of 0.5°–0.5° by resampling, which is a function of GIS, then the resolution was the same as precipitation data of 2004–2005; thus, the same subzones of precipitation change were achieved for the constructed model. Just like the input before, there were still 51 zones of precipitation infiltration (Fig. 7). Under scenarios A1B, A2, and B1, the precipitation all showed increasing trends in 2011– 2050 with a maximum change of 28.31 mm and a minimum change of 20.02 mm. The change ratios were 4.31, 4.31 and 4.41%, respectively. Due to the complexity of the monsoon climate in the study area, precipitation pattern was neglected; only yearly changes of precipitation infiltration were used as the inputs for the three emissions scenarios for 2011–2050.
Scenario 5: combined climate change and water-diversion scenarios The SNWDP will deliver water from southern China to the dry northern China. It consists of three routes: the eastern, middle, and western routes (Liu and Zheng 2002). The middle route, which is 1,276 km long, will transfer water from the Danjiangkou Reservoir on the Huang River, which is the largest tributary of the middle reaches of the Yangtze River, to Beijing and Tianjin as well as to the western part of the North China Plain (Liu and Du 1985). It is expected to significantly alleviate water shortages and reduce groundwater extraction in the Jing-Jin-Ji area. The first phase of the middle route was completed in December 2014 and is projected to transfer 4.95 billion m3/year of water into the Jing-Jin-Ji area. The start date and water diversion plan of the second phase has not been reported. To study the combined impact of the climate change and water-diversion scenarios on the water table, the following assumptions were made. The first phase of the water diversion plan of the middle route during 2011–2050 is assumed to be effective at improving water use efficiency and encouraging recycling of water, and the demand for groundwater is assumed to not increase due to societal and economic development, so groundwater exploitation is reduced by 30%. For simplicity, the average of the three emissions scenarios was adopted as the input for the future climate. Results and discussion The simulated averaged water-table depths of all grids in the study area under these scenarios are shown in Fig. 8. Under the baseline scenario, which includes the present climate and human activity, the water level decreases by 7.41 m in 2050 compared with 2010, and the rate of decline reaches 0.21 m/ year. The recharge from precipitation and irrigation is 10.17 billion m3/year, and the groundwater continues to be depleted; therefore, the average water level is projected to decline continuously in the absence of climate change and waterdiversion measures. Based on the projected changes in precipitation under the three emissions scenarios, there is almost no difference in water-table depth between the three emissions
Hydrogeol J Fig. 7 Precipitation absolute changes in the study area in the future (2011–2050) compared with the baseline (1971–2000)
scenarios. This occurs because the increases in total precipitation are nearly identical under the three scenarios. Moreover, the water levels are projected to decline under the three climate change scenarios by 1.23, 1.21, and 1.24 m, respectively, in 2050 compared with 2010. The recharge from precipitation and irrigation is 2.58 billion m3/year more than that under the baseline scenario, but the groundwater continues to be depleted. These results indicate that the increase of precipitation over the 40 years will not allow the water levels to recover but will decrease the rate of decline compared to the baseline scenario. In the combined climate change and SNWDP scenarios, the water levels will increase gradually at a rate of 0.12 m/year. The water table will increase by 6.71 m in 2050 compared with 2010, that is to say, water levels will recover 6.71 m over the next 40 years. These results suggest a recovery of water levels under the combined climate and water-diversion scenarios. Complete recovery of the groundwater levels is not easy to achieve and may be impossible to realize depending on the magnitude of climate change, so strict groundwater management is necessary. Many studies have predicted the impact of the SNWDP on groundwater (Yang et al. 2012; Cao et al. 2013; Ye et al. 2014; Zhang and Li 2014). In the case of a reduction in groundwater pumping in the North China Plain of 6.0 km3/year, the local groundwater levels in the piedmont region, particularly in Beijing, will recover significantly by the 2030s, and the
Fig. 8 Water-table depth over the next 40 years under different scenarios (the curves are superimposed under scenarios A2, A1B, B1)
declines in the groundwater levels in the central and coastal plain will slow down (Cao et al. 2013). In Beijing, the areas of the cones of depression will decrease by different amounts, and the water levels in the centres of some cones of depression will rise (Yang et al. 2012). The areas of the large cones have been calculated, and the deep groundwater levels in the JingJin-Ji area will be more difficult to recover than the shallow levels (Zhang and Li 2014). These studies have analysed the spatial variations of the water levels in detail and do not indicate complete recovery after the implementation of the SNWDP. In this study, the water level did not recover due to the impact of climate change but did recover due to the impact of the SNWDP, which indicates that the SNWDP will help to control the decrease of water levels. These results are consistent with other studies (Cao et al. 2013; Zhang and Li 2014; Chen et al. 2012); moreover, the integral water levels were calculated for the next 40 years, and the temporal variations were analysed. The results improve the understanding of the impact of the SNWDP. In all studies of climate change impacts, the results are associated with considerable uncertainties. These uncertainties are most directly related to the predictions from GCMs. Climate change projections from GCMs have large uncertainties, especially at regional scales (Murphy et al. 2004; Trenberth 1997). The uncertainty related to the use of different GCMs is the primary source of uncertainty in the projections of climate change on the North China Plain (Fu et al. 2009). In this study, the uncertainty in precipitation in the GCMs is significant. This uncertainty was minimized by not using only one GCM but rather using the multi-model dataset CMIP3. Averaging across the GCMs and emissions scenarios projects annual precipitation increases of 28.2–40.7 mm (3.2– 4.8%) for 2025 (2005–2044; Fu et al. 2009). The projected increase of precipitation in this study is within this range. The second uncertainty is that the precipitation impacts the groundwater level in direct and indirect ways. The amount of precipitation recharging the groundwater will be changed as a result of direct influence, while the groundwater level will be affected indirectly when pumping amount, irrigation amount and surface-water diversion amount get changed as
Hydrogeol J
a result of precipitation fluctuation (Zhang et al. 2014). The precipitation was considered in the climate change scenarios, but other changes in socio-economic activity caused by climate change and their impacts on climate change were not considered (Holman 2006). The third uncertainty is about downscaling methods. There have been several studies about methods of downscaling from GCM to local scale (Maraun et al. 2010; Schmidli et al. 2006; Wetterhall et al. 2005). The methods have impacts on groundwater level projections. The simplest method (change factor) is inappropriate and more rigorous methods should be used (Mileham et al. 2009; Holman et al. 2009). In this study, the area is very large, and the “delta change” method was used. The fourth uncertainty is about the flow model including model structure and model parameter uncertainty (Kuczera and Parent, 1998; Delhomme 1979). In the studies of impact of climate change upon groundwater, the hydrological model is the source of the least uncertainty (Crosbie et al. 2011). Moreover, to build a model for a large study area is very time demanding, especially the parameter identification, and the software and CPU power are limited when facing large volumes of data, therefore, this work has not been done in this study.
Conclusions A conceptual groundwater flow model was developed by analysing the hydrogeological conditions in the Jing-Jin-Ji Plain, and a 3-D transient numerical flow model was then developed using MODFLOW. By fitting the groundwater flow field and the hydrographs of observation wells, the hydrogeologic parameters were identified and the model was calibrated. The model was then used to simulate the response of the groundwater levels to the impacts of future climate change and water diversion. The average annual rainfall increased by 4.31, 4.31, and 4.41% according to GCM forecasts of the A1B, A2, and B1 emission scenarios, respectively, over the next 40 years. The water levels were simulated and analysed under five scenarios. Under the present climate and human activity, the water levels will decline continuously at a rate of 0.21 m/year, and the water levels will decrease by 7.41 m over the next 40 years. Due to the impact of climate change, the water levels will still decrease but at a slower rate. The groundwater levels will increase significantly when water diversion is superimposed, and the maximum increase is projected to be 6.71 m; therefore, the recovery of groundwater in the Jing-Jin-Ji area requires stricter pumping management and it will need to take longer than 40 years. Climate change can cause changes to potential irrigation and other management strategies. This study only considered the direct precipitation recharge for groundwater in the modelling. The interaction of climate change and socio-economic
activity still requires further research to be integrated into the model. Acknowledgements This study was jointly funded by the Key Program for International S&T Cooperation Projects of China (2016yee0109600), Governmental Public Research Funds of China (No. DD20160144) and the Natural Science Foundation of Science and Technology Department in Hebei Province (No. D2016403044). We acknowledge several modelling groups for providing their data for the analysis, including the Program for Climate Model Diagnosis and Intercomparison (PCMDI) and the World Climate Research Programme’s (WCRP’s) Coupled Model Intercomparison Project for collecting and archiving the model output and organizing the model data analysis activity. The data were collected, analysed, and provided by the National Climate Center. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References Bouraoui F, Vachaud G, Li L, Le Treut H, Chen T (1999) Evaluation of the impact of climate changes on water storage and groundwater recharge at the watershed scale. Clim Dynam 15(2):153–161. doi: 10.1007/s003820050274 Burn DH (1994) Hydrologic effects of climatic change in west-central Canada. J Hydrol 160(1):53–70. doi:10.1016/0022-1694(94)90033-7 Cao G, Zheng C, Scanlon BR, Liu J, Li W (2013) Use of flow modeling to assess sustainability of groundwater resources in the North China Plain. Water Resour Res 49(1):159–175. doi:10.1029/ 2012WR011899 Chen W (1999) Groundwater in Hebei Province. Earthquake Publishing House, Beijing Chen Y, Takeuchi K, Xu C, Chen Y, Xu Z (2006) Regional climate change and its effects on river runoff in the Tarim Basin, China. Hydrol Process 20(10):2207–2216. doi:10.1002/hyp.6200 Chen HR, Gao ZY, Wang SL et al (2012) Modeling on impacts of climate change and human activities variability on the shallow groundwater level using MODFLOW. J Hydraulic Eng 43(3):344–363 Chung I, Kim N, Lee J, Sophocleous M (2010) Assessing distributed groundwater recharge rate using integrated surface water–groundwater modelling: application to Mihocheon watershed, South Korea. Hydrogeol J 18(5):1253–1264. doi:10.1007/s10040-010-0593-1 Crosbie RS, Mccallum JL, Walker GR, Chiew FHS (2010) Modelling climate-change impacts on groundwater recharge in the MurrayDarling Basin, Australia. Hydrogeol J 18(7):1639–1656. doi:10. 1007/s10040-010-0625-x Crosbie RS, Dawes WR, Charles SP, Mpelasoka FS, Aryal S, Barron O et al (2011) Differences in future recharge estimates due to GCMs, downscaling methods and hydrological models. Geophys Res Lett 38. doi:10.1029/2011gl047657 Delhomme JP (1979) Spatial variability and uncertainty in groundwater flow parameters: a geostatistical approach. Water Resour Res 15(2): 269–280. doi:10.1029/WR015i002p00269 Ding Y, Guoyu R, Shi G, Deliang C (2006) National assessment report of climate change (I): climate change in China and its future trend. Adv Clim Change Res 2(1):3–8 Ding Y, Ren G, Zhao Z, Xu Y, Luo Y, Li Q, Zhang J (2007) Detection, causes and projection of climate change over China: an overview of recent progress. Adv Atmos Sci 24(6):954–971. doi:10.1007/ s00376-007-0954-4 Falkenmark M, Lundqvist J, Widstrand C (1989) Macro-scale water scarcity requires micro-scale approaches. Nat Resour Forum 13(4):258– 267. doi:10.1111/j.1477-8947.1989.tb00348.x
Hydrogeol J Fei Y (2006) Evolution and conservation of groundwater in Hebei Plain to the south of Beijing and Tjianjin. Hohai University, Nanjing, China Fu G, Charles SP, Yu J, Liu C (2009) Decadal climatic variability, trends, and future scenarios for the North China Plain. J Clim 22(8):2111– 2123. doi:10.1175/2008JCLI2605.1 Goderniaux P, Brouyère S, Blenkinsop S, Burton A, Fowler HJ, Orban P et al (2011) Modeling climate change impacts on groundwater resources using transient stochastic climatic scenarios. Water Resour Res 47. doi:10.1029/2010wr010082 Hao X, Chen Y, Xu C, Li W (2008) Impacts of climate change and human activities on the surface runoff in the Tarim River Basin over the last fifty years. Water Resour Manage 22(9):1159–1171. doi:10.1007/ s11269-007-9218-4 Harbaugh AW (2005) MODFLOW-2005, The US Geological Survey modular ground-water model: the Ground-Water Flow Process. US Geological Survey, Reston, VA Harbaugh AW, Banta ER, Hill MC, McDonald MG (2000) MODFLOW2000, The US Geological Survey modular ground-water model: user guide to modularization concepts and the Ground-Water Flow Process. US Geological Survey, Reston, VA Holman IP (2006) Climate change impacts on groundwater rechargeuncertainty, shortcomings, and the way forward? Hydrogeol J 14(5):637–647. doi:10.1007/s10040-005-0467-0 Holman IP, Tascone D, Hess TM (2009) A comparison of stochastic and deterministic downscaling methods for modelling potential groundwater recharge under climate change in East Anglia, UK: implications for groundwater resource management. Hydrogeol J 17(7): 1629–1641. doi:10.1007/s10040-009-0457-8 IPCC (2007) Climate Change 2007: the physical science basis—contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK Jackson CR, Meister R, Prudhomme C (2011) Modelling the effects of climate change and its uncertainty on UK Chalk groundwater resources from an ensemble of global climate model projections. J Hydrol 399(1):12–28. doi:10.1016/j.jhydrol.2010.12.028 Jyrkama MI, Sykes JF (2007) The impact of climate change on spatially varying groundwater recharge in the Grand River watershed (Ontario). J Hydrol 338(3):237–250. doi:10.1016/j.jhydrol.2007. 02.036 Kabiri R, Bai VR, Chan A (2015) Assessment of hydrologic impacts of climate change on the runoff trend in Klang Watershed, Malaysia. Environ Earth Sci 73(1):27–37. doi:10.1007/s12665-014-3392-5 Kendy E, Zhang Y, Liu C, Wang J, Steenhuis T (2004) Groundwater recharge from irrigated cropland in the North China Plain: case study of Luancheng County, Hebei Province, 1949–2000. Hydrol Process 18(12):2289–2302. doi:10.1002/hyp.5529 Kuczera G, Parent E (1998) Monte Carlo assessment of parameter uncertainty in conceptual catchment models: the Metropolis algorithm. J Hydrol 211(1):69–85. doi:10.1016/S0022-1694(98)00198-X Kurylyk BL, MacQuarrie KTB (2013) The uncertainty associated with estimating future groundwater recharge: a summary of recent research and an example from a small unconfined aquifer in a northern humid-continental climate. J Hydrol 492:244–253. doi:10.1016/j. jhydrol.2013.03.043 Li X, Li G, Zhang Y (2014) Identifying major factors affecting groundwater change in the North China Plain with grey relational analysis. Water 6(6):1581–1600. doi:10.3390/w6061581 Liu C, Du W (1985) Areal reallocation of China’s water resources. GeoJournal 10(2):157–162. doi:10.1007/BF00150734 Liu C, Zheng H (2002) South-to-north water transfer schemes for China. Int J Water Resour Dev 18(3):453–471. doi:10.1080/ 0790062022000006934 Maraun D, Wetterhall F, Ireson AM, Chandler RE, Kendon EJ, Widmann M et al (2010) Precipitation downscaling under climate change: recent developments to bridge the gap between dynamical models
and the end user. Rev Geophys 48(3):633–650. doi:10.1029/ 2009RG000314 McDonald MG, Harbaugh AW (1984) A modular three-dimensional finite-difference ground-water flow model. US Geological Survey, Reston, VA Meng S, Fei Y, Zhang Z, Lei T, Qian Y, Li Y (2013) Research on spatial and temporal distribution of the precipitation infiltration amount over the past 50 years in North China Plain. Adv Earth Sci 28(8): 923–929 Middelkoop H, Daamen K, Gellens D, Grab JC et al (2001) Impact of climate change on hydrological regimes and water resources management in the Rhine basin. Clim Change 49(1–2):105–128. doi:10. 1023/A:1010784727448 Mileham LRG, Taylor M, Tindimugaya TC, Thompson J (2009) The impact of climate change on groundwater recharge and runoff in a humid, equatorial catchment: sensitivity of projections to rainfall intensity. Hydrol Sci J 54(4):727–738. doi:10.1623/hysj.54.4.727 Murphy JM, Sexton DM, Barnett DN, Jones GS, Webb MJ, Collins M, Stainforth DA (2004) Quantification of modelling uncertainties in a large ensemble of climate change simulations. Nature 430(7001): 768–772. doi:10.1038/nature02771 Piao S, Ciais P, Huang Y, Shen Z et al (2010) The impacts of climate change on water resources and agriculture in China. Nature 467(7311):43–51. doi:10.1038/nature09364 Prudhomme C, Reynard N, Crooks S (2002) Downscaling of global climate models for flood frequency analysis: where are we now? Hydrol Processes 16(6):1137–1150. doi:10.1002/hyp.1054 Qian W, Zhu Y (2001) Climate change in China from 1880 to 1998 and its impact on the environmental condition. Clim Change 50(4):419– 444. doi:10.1023/A:1010673212131 Risbey JS, Hamza K, Marsden JS (2007) Use of climate scenarios to aid in decision analysis for interannual water supply planning. Water Resour Manage 21(6):919–932. doi:10.1007/s11269-006-9064-9 Schmidli J, Frei C, Vidale PL (2006) Downscaling from GCM precipitation: a benchmark for dynamical and statistical downscaling methods. Int J Climatol 26(5):679–689. doi:10.1002/joc.1287 Scibek J, Allen D (2006) Modeled impacts of predicted climate change on recharge and groundwater levels. Water Resour Res 42(11), W11405. doi:10.1029/2005WR004742 Shao J, Li L, Cui Y, Zhang Z (2013) Groundwater flow simulation and its application in groundwater resource evaluation in the North China Plain, China. Acta Geol Sin 87(1):243–253. doi:10.1111/17556724.12045 Srivastava R (2013) Groundwater resources under a changing climate. J Indian Institute Sci 93(2):251–263 Stoll S, Franssen HJH, Butts M, Kinzelbach W (2011) Analysis of the impact of climate change on groundwater related hydrological fluxes: a multi-model approach including different downscaling methods. Hydrol Earth Syst Sci 15(1):21–38. doi:10.5194/hess-1521-2011 Trenberth KE (1997) The use and abuse of climate models. Nature 386(6621):131–133. doi:10.1038/386131a0 Vano JA, Nijssen B, Lettenmaier DP (2015) Seasonal hydrologic responses to climate change in the Pacific Northwest. Water Resour Res 51(4):1959–1976. doi:10.1002/2014WR015909 Waibel MS, Gannett MW, Chang H, Hulbe CL (2013) Spatial variability of the response to climate change in regional groundwater systems: examples from simulations in the Deschutes Basin, Oregon. J Hydrol 486:187–201. doi:10.1016/j.jhydrol.2013.01.019 Wetterhall F, Halldin S, Xu CY (2005) Statistical precipitation downscaling in central Sweden with the analogue method. J Hydrol 306(1–4): 174–190. doi:10.1016/j.jhydrol.2004.09.008 Wilby RL, Wigley TML (1997) Downscaling general circulation model output: a review of methods and limitations. Prog Phys Geog 21(4): 530–548. doi:10.1177/030913339702100403
Hydrogeol J Xu X (2006) A study of numerical simulation for groundwater flow in the Beijing plain, China. University of Geosciences, Beijing Xu Y, Gao X, Giorgi F (2010) Upgrades to the REA method for producing probabilistic climate change projections. Clim Res 41(1):61–81. doi:10.3354/cr00835 Xue L, Li W, Yang W, Li J, Zhu X (2010) Numerical modelling on exploited shallow groundwater flow in the plain of Haihe River basin. Geotech Investig Surv 3:50–55 Yang Y, Li X, Wang L, Li C, Liu Z (2011) Characteristics of the groundwater level regime and effect factors in the plain region of Tianjin City. Geolog Surv Res 34(4):313–320 Yang Y, Guo-Min L, Yan-Hui D, Fan-Dong Z (2012) Influence of south to north water transfer on groundwater dynamic change in Beijing plain. Environ Earth Sci 65(4):1323–1331. doi:10.1007/s12665011-1381-5 Ye A, Duan Q, Chu W, Xu J, Mao Y (2014) The impact of the south– north water transfer project (ctp)’s central route on groundwater table in the Hai River Basin, North China. Hydrol Process 28(23): 5755–5768. doi:10.1002/hyp.10081 Zhang G, Fei Y, Wang J, Yan M et al (2012) Adaptation between irrigation and groundwater in North China. Science Press, Beijing
Zhang G, Fei Y, Liu C, Yan M, Wang J (2013) Adaptation between irrigation intensity and groundwater carrying capacity in North China Plain. Trans Chin Soc Agr Eng 29(1):1–10 Zhang X, Pei D, Hu C (2003) Conserving groundwater for irrigation in the North China Plain. Irrigation Sci 21(4):159–166. doi:10.1007/ s00271-002-0059-x Zhang Y, Li G (2013) Long-term evolution of cones of depression in shallow aquifers in the North China Plain. Water 5(2):677–697. doi:10.3390/w5020677 Zhang Y, Li G (2014) Influence of south-to-north water diversion on major cones of depression in North China Plain. Environ Earth Sci 71(9):3845–3853. doi:10.1007/s12665-013-2771-7 Zhang Z, Fei Y (2009) Atlas of sustainable utilization of groundwater resources in the North China Plain. SinoMaps Press, Beijing Zhang Z, Shen Z, Xue Y (2000) Groundwater environment evolution in the North China Plain. Geological Publishing House, Beijing Zhang Z, Fei YH, Chen ZY (2009) Investigation and assessment of sustainable utilization of groundwater resources in the North China Plain. Geological Publishing House, Beijing Zhu L, Gong H, Li X, Li Y, Su X, Guo G (2013) Comprehensive analysis and artificial intelligent simulation of land subsidence of Beijing, China. Chin Geogra Sci 23(2):237–248. doi:10.1007/s11769-013-0589-6