Next Article in Journal
The Integration of Nature-Inspired Algorithms with Least Square Support Vector Regression Models: Application to Modeling River Dissolved Oxygen Concentration
Next Article in Special Issue
Uncertainty in Irrigation Technology: Insights from a CGE Approach
Previous Article in Journal
Validation of ECMWF Multi-Layer Reanalysis Soil Moisture Based on the OzNet Hydrology Network
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Optimization-Based Agricultural Water-Saving Potential Analysis in Minqin County, Gansu Province China

Centre for Agricultural Water Research in China, China Agricultural University, Beijing 100083, China
*
Author to whom correspondence should be addressed.
Water 2018, 10(9), 1125; https://doi.org/10.3390/w10091125
Submission received: 3 July 2018 / Revised: 29 July 2018 / Accepted: 8 August 2018 / Published: 23 August 2018

Abstract

:
To deal with the contradictions that are caused by natural conditions and unreasonable water allocations in Minqin County, which are located downstream of the Shiyang River basin in arid northwest China, an optimization-based multi-scale calculation method was proposed for analyzing agricultural water-saving potential. Firstly, an optimization model was developed for allocating water and land resources legitimately with the conjunctive use of surface water and groundwater. Secondly, the groundwater equilibrium was fully considered in developing optimization model to achieve the ecological value of agricultural water savings. Then, multi-scale agricultural water-saving potentials were analyzed based on optimal results under different water-saving levels. These results provide local water managers with satisfactory economic benefit with higher water use efficiency. With reasonable management strategies of water and land resources, the ecosystem of Minqin County could gradually recover in the future. The results of the multi-scale water-saving potential analysis can help decision makers to identify desired water-saving plans that consider the coordinated development of the local economy, society, and ecology.

1. Introduction

Due to the rapid urbanization and industrialization in China, the increasing conflict between limited water resources and increased water demands calls for reasonable water allocation schemes. As the biggest consumer in most areas, especially in arid and semi-arid areas that are characterized by low rainfall and high evaporation, irrigation water plays an important role in agriculture production [1,2]. For example, in the arid regions of northwest China, irrigation water accounts for approximately 90% of the total water use [3], while the irrigation water use efficiency is far lower than the advanced regions [4]. That is, enormous potentials exist in the conservation of limited agricultural water resources. The accurate assessment of agricultural water-saving potential can not only help local sustainable development by affecting local food security, ecological security, and water security, but also provide specific water-saving direction to local decision makers. Minqin County, which is located downstream of the Shiyang River Basin, is a typical arid area suffering serious ecological deterioration due to the long-term exploitation of groundwater and unguided water management. It is of great significance to study the water-saving potential of Minqin County as a typical case.
The agricultural water-saving potential was measured using various techniques, engineering projects, and water-saving management, and involved four scales, including crop, field, irrigation area, and regional/basin [5]. Many researchers have tried to measure the agricultural water-saving potential [6,7,8,9]. Different measures have been adopted, including efficient irrigation technology, deficit irrigation regulation, and irrigation scheduling improvement [10]. However, comprehensive analysis would be obtained by multiple scales of water-saving measures and considering multiple effects [11]. Although every method has its appropriate applications, according to its advantages and disadvantages, unreasonable results might be obtained when overlooking the important effects that are created by agricultural water and land allocation. Zhang and Guo [12] integrated water management optimization model and agricultural water-saving potential analysis methods to analyze local irrigation water-saving potential based on optimization results. However, how to combine multiple scales of water-saving measures with water and land resource optimal allocation has rarely been considered and needs to be addressed.
Decreasing of available irrigation water and the increasing of water demand call for the reasonable allocation and effective management of agricultural water resources. Additionally, crop planting structure optimization, which allocates the optimum planting proportion to each crop to achieve the goals of increasing agricultural economic benefits and decreasing irrigation water use [13], is equally essential for sustainable agricultural development and to address serious water shortage problems. Therefore, irrigation water and planting structure optimization models were extensively employed to provide theoretical guidance for decision makers [14,15,16]. The interaction and exchange between surface and groundwater resources were considered throughout the conjunctive use of surface water and groundwater [17,18], in which the available groundwater depths were introduced [12,19]. In addition, some studies tried to combine various optimization models with a water-saving potential analysis, and thus contribute to the optimal management of water and land resources [12,20]. However, with more attention being paid to optimizing agricultural water and land resources separately via different optimization models, the interactions between them were not fully considered and few researchers realized the ecological value in agricultural water savings.
Therefore, to deal with the multiple contradictions that are caused by natural conditions and unreasonable water allocation, which manifest in ecological harm and further affect agricultural water and land resource management, the objectives of this study are to (1) establish an integrated optimization model for water and land resource allocation based on the conjunctive use of surface water and groundwater; (2) introduce the groundwater equilibrium constraint to obtain several irrigation water schemes and planting areas under different water-saving levels; and, (3) conduct the analysis of multi-scale irrigation water-saving potential with water-saving measures. Finally, an optimization-based multi-scale research method for agricultural water-saving potential is developed. The method was applied to Minqin County, which is located downstream of the Shiyang River basin in northwest China, to demonstrate its effectiveness and practicality for agricultural water and land management and agricultural water-saving potential analysis. The results may support local decision makers formulate better allocation schemes, and the study methods should be helpful for managers to identify the desired water and land allocation plans in similar areas.

2. Study System

2.1. Study Area

Minqin County is a typical arid region, located downstream of the Shiyang River basin, northwest China, within the east longitude 102°52′–103°50′ and the north latitude 38°22′–39°6′. As the green barrier of ecological security in northwest China with the title of the desert oasis, Minqin County is surrounded by the Badain Jaran desert and Tengger desert (Figure 1), and is located in an arid climate zone, with an average annual precipitation and potential evaporation of about 110 mm and 2623 mm, respectively. Precipitation varies greatly among different seasons, with precipitation from July to September, accounting for nearly 66% of the whole year’s precipitation [2]. The average annual precipitation from 1985–2014 in Minqin County is shown in Figure 2, and the precipitation distribution in 2014 of Minqin County is expressed in Figure 3. Minqin County has three main irrigation areas: Hongyashan, Huanhe, and Changning, among which Hongyashan’s irrigation area occupies 92.8% of the whole Minqin irrigation area [21]. All of the surface water supply for Minqin County comes from the Hongyashan reservoir, and agriculture is the highest water use sector.
Because of economic development and population growth in the upper and middle reaches of the Shiyang River, the available surface water in Minqin County is rapidly decreasing. Meanwhile, a more serious problem is the over-exploitation of the groundwater. In the past few decades, the over-exploitation of the groundwater has caused a series of ecological problems, such as the continuous decline of the groundwater level (Figure 4), deterioration of the groundwater environment, and further leads to serious desertification, increased salinity, and oasis shrinking [22]. To restore the ecological system of the Minqin oasis, local water managers limited the groundwater exploitation, and thus led to a corresponding reduction of agricultural production. Although the government has taken action by transferring water from other areas in recent years, the water availability for the study area is far less than the water requirement. Thence, both the water utilization efficiency and water production efficiency need to be improved.
The economic development of Minqin County mainly depends on agricultural production, and the main crops in Minqin include spring wheat, maize, cotton, vegetables, melons, and oil crops. The current irrigation quotas of the typical crops in Minqin County are displayed in Figure 5. The statistical data of Minqin County in recent years can be seen in Table 1, which shows that the proportion of economic crops has increased gradually, and the cultivated areas of water-consuming crops are increasing too, indicating that fierce contradictions exists between the crop planting structure and limited available water. This contradiction is a great challenge faced by local managers in practice.
With the improvement of agriculture water management, great water-saving potentials exist in local irrigation water allocation system. Therefore, it is of great significance to analyze the agricultural water-saving potentials with the efficient utilization of limited water and land resources in order to improve agricultural production, promote socioeconomic development, and help ecological restoration in Minqin County.

2.2. Data Collection

Natural environmental data, socio-economic information, and typical crops data were used in this study. Meteorological data from 1985–2014 in the Minqin County were obtained from the China Meteorological Data Sharing Service System; crops data [23], including the typical crop sensitivity index, crop growth stages division, crop coefficient, and maximum crop yield, were used in this study; crop planting data and population were obtained from The Statistical Yearbook of Wuwei City; streamflow data, water supply data, and current irrigation water distributions originated in the survey of Wuwei City; the groundwater management objectives were available in The Shiyang River Basin Key Governance Projects; the current irrigation quotas were obtained from The Wuwei Industry Water Quota; and finally, the value range of the parameters expressed in Table 2 [24] were references of the coefficient including the regional specific yield, the leakage coefficient of the canal system, the precipitation infiltration coefficient, and the field leakage coefficient. All were used for the optimization model.

3. Study Methods

3.1. System Framework

This paper attempts to combine the optimization model for agricultural water and land resource allocation with agricultural water-saving potential. This section contains two major components (Figure 6): (1) integrated optimization model for agricultural water and land resource allocation and (2) multi-scale agricultural water-saving potentials. The joint connect point between these two parts is the total available water of Minqin County.
The optimization model is developed to achieve the efficient utilization of water resources and optimally allocate the land resources of crops via adjusting the crop planting acreage. The appropriate adjustments of agricultural water resources and cultivated areas are explored to support decision makers for better agricultural water management.
When considering the realities of the study area, analysis with the multi-scale water-saving potentials, which consists of the crop-scale, field-scale, irrigation area-scale, and region-scale, are conducted. The water-saving potentials of different scales are calculated by corresponding water-saving measures. By comparing the results among different scales, a relatively reliable analysis could be obtained to help local managers to alleviate agricultural water shortage.

3.2. Groundwater Equilibrium Analysis

Irrigated groundwater balance refers to the relationships between the total groundwater recharge, the total discharge, and the storage during a period of time. That is, the difference between the groundwater recharge and the discharge in a certain period is equal to the change of the underground reservoir inventory, which can be expressed as follows [25]:
Δ Q = Q s Q e  
where ΔQ is the change in the amount of groundwater during the certain period; Qs is the total amount of groundwater recharge during the certain period; and, Qe is the total amount of groundwater discharge during the certain period.
The main sources of recharge in the study area include precipitation infiltration, field infiltration, canal leakage, well irrigation seepage feedback, and lateral recharge. The main discharges include well water exploitation, evapotranspiration, and lateral discharge. The change of the groundwater level indicates the amount of variation in the stored groundwater. The amount of lateral supply and lateral discharge is so small as to be ignored. The limit depth of groundwater evaporation is 5 m [26]. Minqin’s groundwater depth is far more than 5 m, so the evapotranspiration is assumed to be zero in this study. Therefore, the groundwater equilibrium function can be expressed, as follows:
μ A Δ H = Q p + Q c + Q f + Q w Q o  
where the μ is the regional specific yield; A (hm2) is the area; ΔH (m) is the increase of the groundwater level during the certain period; Qp (104 m3) is the precipitation infiltration during the certain period; Qc (104 m3) is the canal leakage during the certain period; Qf (104 m3) is the field infiltration during the certain period; Qw (104 m3) is well irrigation seepage feedback during the certain period; and, Qo (104 m3) is the amount of well water exploitation during the certain period.

3.3. Integrated Optimization Model for Water and Land Resource Allocation

When considering the transformation of surface water resource, the Jensen model (which Jensen M.E. first proposed in 1968) is introduced in this study to optimize the farmers’ net economic benefit with an objective formulation. The decision variables are the allocation of the surface water and groundwater for each crop in each period, and the crops’ cultivated areas. The objective function is expressed as:
max = i = 1 m A i B i Y m i j = 1 n ( S W i j + E P i j + G W i j E T m a x i j ) λ i j C 1 i = 1 m A i t = 1 T ( S W i t η 1 ) C 2 i = 1 m A i t = 1 T ( G W i t η 2 ) i = 1 m A i D i  
S W i j = t = 1 T q i j t S W i t  
G W i j = t = 1 T q i j t G W i t  
where i is the crop type; j is the number of crop growth stage; t is the index of time periods; Ai is the crop area of crop i (hm2); Bi is the price of the crop per unit (CNY/kg); Ymi is the maximum yield of crop i per unit under full irrigation (kg/hm2); SWij is the irrigated surface water of crop i during stage j (mm); GWij is the irrigated groundwater of crop i during stage j (mm); EPij is the effective precipitation of crop i during stage j (mm); ETmaxij is the maximum evapotranspiration of crop i during stage j (mm); λ i j is the water sensitivity index of crop i within stage j; SWit is the irrigated water for surface water of crop i during period t (mm); GWit is the irrigated water from the groundwater of crop i during period t (mm); C1 is the price of surface water per unit (CNY/m3); C2 is the price of groundwater per unit (CNY/m3); η 1 is the utilization coefficient of surface water; η 2 is the utilization coefficient of groundwater; D i is the planting cost of crop i (CNY); and, qijt is the proportion of crop i during stage j in the period t. There are some constraints in this model, as described below.

3.3.1. Groundwater Equilibrium Constraint

In order to prevent land salinization and desertification, respectively, caused by a high groundwater level and low groundwater level, the groundwater equilibrium constraint was introduced into the model. The available upper and lower bounds of the groundwater depth were used to control the groundwater level in a reasonable range in order to protect the groundwater resources. The specific constraint is expressed as follows:
H t = H t 1 α P t μ i = 1 m A i [ β S W i t η 1 + β S W i t η + β G W i t G W i t λ η 2 ] μ A + E G t  
H m i n H t H m a x  
where Ht is the groundwater level depth in the period t (mm); Ht−1 is the groundwater level depth in the period t − 1 (mm); Pt is the effective precipitation in the period t (mm); α is the precipitation leakage coefficient; β is the canal leakage coefficient; β is the irrigation return flow coefficient; η is the field water use coefficient; λ is the ratio of agricultural water consumption to the total amount of well water exploitation; β is the well regression coefficient; EGt is the groundwater evaporation in the period t (m3); Hmin is the groundwater level upper limit (m); and, Hmax is the groundwater level lower limit (m).

3.3.2. Water Supply Constraints

The amount of surface water and groundwater calculated in the model cannot exceed the total available agricultural water supply, thus the following constraints are proposed.
Surface water supply constraint
i = 1 m A i t = 1 T S W i t Q η 1  
Groundwater availability constraint
i = 1 m A i t = 1 T G W i t G η 2  
where Q (104 m3) is the surface water supply and G (104 m3) is the groundwater availability.

3.3.3. Evapotranspiration Constraint

In the model, the sum of precipitation and irrigation water during each growth stage should be less than the maximum water requirement, and satisfy the minimum evapotranspiration of crops during the corresponding growth stage. Therefore, the constraint is formed, as follows:
E T m i n i j S W i j + E P i j + G W i j E T m a x i j  
where ETminij is the minimum evapotranspiration of crop i during stage j (mm).

3.3.4. Food Security Constraint

In Minqin County, the main grain crops are spring wheat and maize. The sum of the two outputs should meet the minimum food requirements of the total population (420 kg/person) [20]. This constraint can be expressed, as follows:
i = 1 k A i Y m i j = 1 n ( S W i j + E P i j + G W i j E T m a x i j ) λ i j Y ¯ × S  
where Y ¯ is the per capita minimum food weight (kg/person) and S is the local population.

3.3.5. Planting Area Constraints

In the process of optimization, the total acreage of the crops should less than the maximum area. The maximum and minimum planting area constraints are formed to avoid the extreme cultivation adjustment of certain crops. Constraints are expressed, as follows
i = 1 m A i A m a x  
A i A i , m a x  
A i A i , m i n  
where Amax is the total crop acreage (hm2); Ai,max is the maximum acreage of the crop i (hm2); and, Ai,min is the minimum acreage of the crop i (hm2).

3.3.6. Non-Negative Constraints

The total amount of irrigation water in each crop growth period must be greater than zero, and the expression of these constraints are as follows:
S W i t 0 , G W i t 0  
Spring wheat, maize, cotton, seed melon, and oil flax are selected for model optimization in this study. The number of crop growth stages is five, and the number of time periods is 12. The average prices of spring wheat, seed melon, cotton, maize, and oil flax are 2.5 CNY/kg, 1.6 CNY/kg, 14.5 CNY/kg, 1.6 CNY/kg, and 6.5 CNY/kg, respectively. The precipitation meteorological data multiplied by the effective utilization factor is calculated as the effective precipitation. The values of ET0 and ETmax are obtained from the FAO-56 (Food and Agriculture Organization) Penman-Monteith method. The available water supply is calculated when considering the proportion of planting and the irrigation water use coefficient based on the total agricultural irrigation water consumption in 2014. Other relevant parameters are shown in Table 3, Table 4 and Table 5. The optimization model is programmed by LINGO mathematical optimization software (LINDO Systems Inc., Chicago, IL, America), and the results are the optimization schemes of the comprehensive model of water and land resource allocation.

3.4. Multi-Scale Agricultural Water-Saving Potential Analysis Method

The agricultural water-saving potentials mainly include four scales: crop, field, irrigation area, and region/basin. There are notable differences in the agricultural water-saving potential in each scale, and the implementation of water-saving measures in the corresponding scales inevitably affects the others [5]. The researches on the multi-scale agricultural water-saving potentials are as follows:
  • The crop physiological process is taken into account in the process of the crop-scale agricultural water-saving potential calculation. To obtain satisfactory crop yields, the irrigation mode of water deficit is applied to save irrigation water. Different degrees (5%, 10%, 15%, 20%, 25%, and 30%) of water deficit are divided and implementation proportions of 10%, 20%, and 40% are set up, from which the most advisable scheme can be picked out. The difference between the previous irrigation water value and the current irrigation water value is calculated accordingly as the crop-scale water-saving potential.
  • The field-scale water-saving potentials are obtained by calculating the differences between the previous irrigation water consumption and the optimized irrigation water allocation with crop planting structure adjustments. As displayed in Figure 5, the irrigation water consumption of spring wheat and maize is larger than the other crops. Thus, reducing the planting area of these two crops and increasing the area of cotton and oil flax appropriately could save more irrigation water. However, to ensure regional food security, the area of spring wheat and maize cannot be reduced too much.
  • The irrigation area-scale water-saving potentials are reflected by the leakage and the loss of water in the process of irrigation water delivery, which relate to engineering measures, including canal system seepage, pipeline water supply, sprinkler irrigation, and drip irrigation. Currently, in Minqin County, the surface water irrigation utilization coefficient, and the groundwater irrigation utilization coefficient reach 0.61 and 0.85, respectively, and could be further enhanced. Assuming that the head flow is constant, the irrigation water delivered to the field is increased, while the water utilization coefficient is improved. The increased amount of water is calculated as the water-saving potential in the irrigation area.
  • The optimal combination of water-saving measures in crop, field, and irrigation areas is considered in the calculation of the region-scale water-saving potential. The detailed calculation steps (Figure 7) are as follows: (1) the optimization scheme of the integrated optimization model for agricultural water and land resource allocation is used as the optimal combination of crop-scale and field-scale water-saving measures; (2) the agricultural net irrigation water is calculated; (3) water savings with different water utilization coefficient improvements are calculated with engineering water-saving measures applied; (4) the amount of gross irrigation water is calculated; and, (5) the calculation results are compared to the current irrigation water consumption, and the reduction of the headwater diversion is calculated as the region-scale agricultural water-saving potential.

4. Results Analysis and Discussion

4.1. Integrated Optimization Model for Agricultural Water and Land Resource Allocation

The integrated optimization model for agricultural water and land resource allocation is used in order to obtain the distribution of water and crop area during the crop growth period (Table 6, Figure 8). As expressed in Table 6, the optimization results increase the water that is allocated to maize because the maize has the highest water productivity among five crops. Simultaneously, the irrigation water of other four crops is reduced when compared with the current scheme. After optimization, 8.66 × 104 m3 irrigation water can be saved. As shown in Figure 8, excessive groundwater exploitation in the peak period of water supply can be effectively avoided by the conjunctive allocation of surface water and groundwater. Thence, the groundwater depth can maintain a safe level with a slight degree of lifting, which contributes positively to regional ecological restoration. The optimal total net benefit is 0.4 billion CNY, and the net benefit per unit water is 3.3 CNY. When considering maximum economic benefit, land resources are allocated to the crops with high yield and low water demand to make full use of limited resources. For local water managers, adjusting the local crop irrigation quota and crop acreage could promote water savings, production increases, and expanded revenue. For local farmers, reducing the cultivation of spring wheat and oil flax, and replacing them with maize, cotton, and seed melon, may be a better choice and bring greater benefits.

4.2. Multi-Scale Agricultural Water-Saving Potential

4.2.1. Crop-Scale Water-Saving Potential

The Jensen model was introduced to calculate the yield of crops in each scenario (Table 7). The highest yields minus the actual yields are regarded as the reduction of the crop yields. The relationships between the reduction of crop yields and the degree of water deficit are expressed in Figure 9. As shown in Figure 9, 15% is a desirable degree of water deficit while considering both the water-saving potential and crop yields. Based on the water-saving schemes, water-saving potentials were obtained by reducing the irrigated water by in different implementation proportion of 10%, 20%, and 40% (Figure 10). In this study, 100% represents totally applying water-saving measures to save water in the region. Although more water has been saved with 40% implementation, assurance in technology and high cost jointly hinder its application. Thus, a scheme of 20% implementation with a 15% water deficit seems more recommendable for local managers, in which the water-saving potential reaches 9.68 × 105 m3. In addition, as shown in Figure 9, reducing the irrigation water of oil flax and seed melon firstly contribute to minimizing farmers’ losses.

4.2.2. Field-Scale Water-Saving Potential

In this section, it is assumed that water quantity in the typical crop growth period is constant with the change of planting area. The saved water consumption was calculated, and the results were shown in Figure 11, where 100% represents that the planting structure is totally adjusted. Although the water consumption of grain crops is high, it is inadvisable to decrease grain crops’ acreage too much, due to the restriction of regional food security. From the data in Table 1, the current proportion of grain crops, including spring wheat and maize is 41%. Therefore, it can be concluded that 15% is a proper adjustment proportion, with the field-scale water-saving potential reaching 2.34 × 106 m3.

4.2.3. Irrigation Area-Scale Water-Saving Potential

From the data of water consumption of different industries in 2014, the current surface irrigation water is 1.40 × 108 m3, and the gross ground irrigation water is 3.49 × 107 m3. In this study, the available irrigation water is 1.74 × 108 m3, which is the calculation result of the total agricultural water supply multiplied by the planting proportion (70%). Although plenty of leakage water could be saved by irrigation area-scale water-saving engineering measures, it is difficult to substantially improve the water use coefficient of the canal system and field water use coefficient. Based on previous researches [5], reasonable targets are set up, and the water-saving potentials under different water use efficiency levels are calculated (Figure 12). When the surface water utilization coefficient is 0.66 and the groundwater utilization is 0.90, the water-saving potential of the irrigation area-scale is 8.62 × 106 m3. Therefore, it is necessary to choose suitable regional development water-saving targets, when implementing irrigation area-scale water-saving measures in irrigation areas.

4.2.4. Region-Scale Water-Saving Potential

From the results of the optimization model, it can be found that the net irrigation water requirement is 9.98 × 107 m3, of which the surface water is 8.26 × 107 m3, and the groundwater is 1.71 × 108 m3. Based on the optimization results, the water-saving potentials under different improvement levels of water use efficiency are calculated (Figure 13). As shown in Figure 13, the possible region-scale water-saving potential can theoretically reach 1.69 × 107 m3. The region-scale water-saving potential is greater than any other scales, with harmoniously considering the regional ecological, social, and economic stable development. Furthermore, it can be concluded that region-scale water-saving measures can effectively alleviate the contradiction between agricultural water supply and demand and improve the irrigation water use efficiency. For local managers, it is meaningful to execute a suitable water-saving scheme via the optimal combination of water-saving measures in crop, field, and irrigation areas. It is equally important to choose suitable development goals on the basis of local reality.
To sum up, multi-scale agricultural water conservation measures can save considerable irrigation water and ease the regional water pressure. Through saving water intuitively, the water-saving measures at the crop scale have a negative impact on certain crops’ yields, while the field-scale water-saving measures can increase crop yields for farmers with saving a large amount of irrigation water. Irrigation area-scale water-saving measures can save a large quantity of water leakage. Finally, the region-scale water-saving measures embody the greatest advantage in saving water to ameliorate the status quo of water shortages. That is, a reasonable scheme can contribute to ecological restoration, effectively alleviating the deterioration of the ecology and groundwater environment in Minqin County. Thus, the preferred choice is to adopt the region-scale water-saving measures for local ecological environment protection and normal agricultural production. Several recommendations are presented, as follows:
  • To ensure the reasonable allocation of limited water resources, an optimal amount of irrigation water for five crops, which not include fallow or non-growing-season periods, should be allocated, with 366 mm for spring wheat, 237 mm for maize, 238 mm for cotton, 473 mm for seed melon, and 286 mm for oil flax.
  • To achieve the efficient utilization of limited land resources, moderate adjustments of crop cultivated areas could be made, including a 25% reduction of spring wheat, 13.5% increase of seed melon, 17.5% increase of cotton, 3.35% reduction of maize, and 7.9% increase of oil flax.
  • To obtain a larger degree of water-saving, stronger financial support should be provided for engineering water-saving renovation measures from the government’s investment. The comprehensive promotion of efficient water-saving irrigation technology should be the development target for local agriculture water managers.
Furthermore, the entire region was considered in the optimization, to deal with the unreasonable water allocation of the Minqin County. These efforts would support local managers to develop more rational water allocation schemes. However, water resource managers face complicated problems in the practice of optimization. For example, China’s agricultural land is decentralized. Farmers have the right to choose the growing crops based on individual interests, rather than overall interests, which makes the implementation of the optimization scheme difficult. These factors would be fully considered in future research.

5. Conclusions

This study proposed an integrated optimization model for agricultural water and land resource allocation, and an optimization-based multi-scale water-saving potential analysis framework for support irrigation water management. It can offer alternative water-saving schemes in corresponding conditions for decision makers.
The characteristics of this study are as follows:
  • The integrated optimization model incorporated planting structure optimization and water resource allocation with the conjunctive use of surface water and groundwater, and it was applied to Minqin County in Gansu Province, China. The optimization results provide comprehensive and efficiency water-saving solutions with higher economic benefits.
  • The groundwater equilibrium constraint was used in the model. Taking full advantage of the interaction between surface water and groundwater, this model keeps the groundwater environment a stable and safe condition with a lifting tendency of the groundwater depth. Through balancing agricultural water and ecological water, optimal solutions contribute to both irrigation water saving and ecological restoration.
  • Four scales were chosen in order to analyze the agriculture water-saving potential via adopting water-saving measures, including cutting down the water supply, adjusting the planting structure, improving the utilization coefficient of irrigation water, and the optimal combination of above measures based on integrated optimization model. By analyzing the results of the four scales, it was found that region-scale water-saving scheme not only demonstrates a huge potential of water-saving and agricultural benefit, but also contribute to balancing agricultural prolificacy and ecological restoration.
In addition, the framework, the allocation optimization model and the analysis method in this study can also be applied to other similar regions to help better water-allocation scheme formulation. However, some limitations exist in the proposed method. For example, local policies, year-to-year variability in supply and demand, as well as precise relationships in model formulation could be taken into account in further research.

Author Contributions

Conceptualization, Q.Y.; Methodology, Q.Y. and F.Z.; Software, Q.Y.; Writing-Original Draft Preparation, Q.Y.; Writing-Review & Editing, Q.Y., F.Z. and P.G.; Supervision, P.G.; Project Administration, P.G.; Funding Acquisition, P.G.

Funding

This research was supported by the National Key R&D Program of China (No. 2017YFC0403201) and National Natural Science Foundation of China (No. 51621061).

Acknowledgments

The authors are grateful to the editor and the anonymous reviewers for their insightful comments and suggestions.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Li, M.; Guo, P.; Singh, V.P. Biobjective Optimization for Efficient Irrigation under Fuzzy Uncertainty. J. Irrig. Drain. Eng. 2016, 142, 05016003. [Google Scholar] [CrossRef]
  2. Li, M.; Guo, P.; Singh, V.P. An efficient irrigation water allocation model under uncertainty. Agric. Syst. 2016, 144, 46–57. [Google Scholar] [CrossRef]
  3. Huang, Y.; Li, Y.P.; Chen, X.; Ma, Y.G. Optimization of the irrigation water resources for agricultural sustainability in Tarim River Basin, China. Agric. Water Manag. 2012, 107, 74–85. [Google Scholar] [CrossRef]
  4. Ye, Y.X. Development Status of agricultural water use and water saving agriculture in China. Mod. Agric. 2015, 75. (In Chinese) [Google Scholar] [CrossRef]
  5. Kang, S.Z.; Yang, J.Z.; Pei, Y.S.; Feng, S.Y.; Huo, Z.L.; Sun, J.S.; Zhang, X.Y.; Wu, J.W.; Wang, J.L.; Tong, L.; et al. Farmland Water Cycle Process and Agricultural Efficient Water Use Mechanism in Hai River Basin; Science Press: Beijing, China, 2013; pp. 204–252. (In Chinese) [Google Scholar]
  6. Zhao, X.; Wang, Y.; Xueming, M.A. Comprehensive evaluation of agricultural water-saving potential in the middle reaches of Hei River using genetic projection pursuit model. Chin. J. Eco-Agric. 2014, 22, 104–110. [Google Scholar] [CrossRef]
  7. Karimov, A.; Molden, D.; Khamzina, T.; Platonov, A.; Ivanov, Y. A water accounting procedure to determine the water savings potential of the fergana valley. Agric. Water Manag. 2012, 108, 61–72. [Google Scholar] [CrossRef]
  8. Horst, M.G.; Shamutalov, S.S.; Pereira, L.S.; Goncalves, J.M. Field assessment of the water saving potential with furrow irrigation in Fergana, Aral Sea basin. Agric. Water Manag. 2005, 77, 210–231. [Google Scholar] [CrossRef]
  9. Yan, N.; Wu, B.; Perry, C.; Zeng, H. Assessing potential water savings in agriculture on the Hai basin plain, China. Agric. Water Manag. 2015, 154, 11–19. [Google Scholar] [CrossRef]
  10. Christian-Smith, J.; Cooley, H.; Gleick, P.H. Potential water savings associated with agricultural water efficiency improvements: A case study of California, USA. Water Policy 2012, 14, 194. [Google Scholar] [CrossRef]
  11. Liu, J.G.; Zhao, Y.; Pei, Y.S.; Tan, X.M. Agricultural water saving potential at different scales in Tuhaimajia River Basin. Adv. Sci. Technol. Water Resour. 2012, 32, 50–53. (In Chinese) [Google Scholar]
  12. Zhang, D.M.; Guo, P. Integrated agriculture water management optimization model for water saving potential analysis. Agric. Water Manag. 2016, 170, 5–19. [Google Scholar] [CrossRef]
  13. Zeng, X.T.; Kang, S.Z.; Li, F.S.; Zhang, L.; Guo, P. Fuzzy multi-objective linear programming applying to crop area planning. Agric. Water Manag. 2010, 98, 134–142. [Google Scholar] [CrossRef]
  14. Li, Y.P.; Huang, G.H. Planning agricultural water resources system associated with fuzzy and random features. J. Am. Water Resourc. Assoc. 2011, 47, 841–860. [Google Scholar] [CrossRef]
  15. Hassaballah, K.; Jonoski, A.; Popescu, I.; Solomatine, D.P. Model-based optimization of downstream impact during filling of a new reservoir: Case study of Mandaya/Reseires reservoirs on the Blue Nile River. Water Resour. Manag. 2012, 26, 273–293. [Google Scholar] [CrossRef]
  16. Zhang, Z.Y.; Si, H.; Feng, B.P.; Hu, C.; Lu, M.X. An optimal model for agriculture water and land resources configuration in water shortage irrigation area. J. Hydraul. Eng. 2014, 45, 403–409. [Google Scholar]
  17. Safavi, H.R.; Darzi, F.; Mariño, M.A. Simulation-Optimization Modeling of Conjunctive Use of Surface Water and Groundwater. Water Resour. Manag. 2010, 24, 1965–1988. [Google Scholar] [CrossRef]
  18. Chen, C.W.; Wei, C.C.; Liu, H.J.; Hsu, N.S. Application of Neural Networks and Optimization Model in Conjunctive Use of Surface Water and Groundwater. Water Resour. Manag. 2014, 28, 2813–2832. [Google Scholar] [CrossRef]
  19. Yue, W.F.; Yang, J.Z.; Zhan, C.S. Coupled model for conjunctive use of water resources in the Yellow River irrigation district. Trans. Chin. Soc. Agric. Eng. 2011, 27, 35–40. (In Chinese) [Google Scholar]
  20. Xu, W.L.; Su, X.L. Agricultural water-saving potential analysis based on crop planting structure optimization—A case study of Liangzhou. Agric. Res. Arid Areas 2010, 28, 161–165. (In Chinese) [Google Scholar]
  21. Li, M.; Guo, P.; Fang, S.Q.; Zhang, L.D. An inexact fuzzy parameter two-stage stochastic programming model for irrigation water allocation under uncertainty. Stochastic Environ. Res. Risk Assess. 2013, 27, 1441–1452. [Google Scholar] [CrossRef]
  22. Liu, Z.G. Analysis on the Reform of Water Resources Management in Minqin County. Gansu Sci. Technol. 2016, 32, 10–11, 13. (In Chinese) [Google Scholar]
  23. Kang, S.Z.; Su, X.L.; Du, T.S.; Feng, S.Y.; Tong, L.; Shen, Q.L.; Shi, P.Z.; Yang, X.Y.; Zuo, Q.; Wang, F.X.; et al. Transformation of Water Resources in Watershed Scale in Northwest Arid Basin and Its Water Saving Regulation Model—A Case Study of Shiyang River Basin in Gansu Province; China Water & Power Press: Beijing, China, 2009; pp. 247–251, 524–530. (In Chinese) [Google Scholar]
  24. Wang, X.H. Study on the Coupled Simulation and Regulation of Groundwater and Surface-Water in Sanjiang Plain, Northeast China; Northeast Institute of Geography and Agroecology Chinese Academy of Sciences, University of Chinese Academy of Sciences: Huairou, China, 2015. [Google Scholar]
  25. Virginia, V.; Krepper, C. Water balance of the Salado-Juramen to river basin (Argentina). Hydrol. Processes 2013, 27, 3825–3832. [Google Scholar] [CrossRef]
  26. Zhang, R.L. The Characters of Distribution and Transformation of Water Resources in Shiyangriver Basin; China University of Geosciences: Beijing, China, 2006. [Google Scholar]
Figure 1. Sketch map of the Shiyang River basin in northwest China.
Figure 1. Sketch map of the Shiyang River basin in northwest China.
Water 10 01125 g001
Figure 2. Precipitation from 1985–2014 in Minqin County.
Figure 2. Precipitation from 1985–2014 in Minqin County.
Water 10 01125 g002
Figure 3. Precipitation distribution for 2014 in Minqin County.
Figure 3. Precipitation distribution for 2014 in Minqin County.
Water 10 01125 g003
Figure 4. Groundwater availability and groundwater depth from 2000–2015 in Minqin County.
Figure 4. Groundwater availability and groundwater depth from 2000–2015 in Minqin County.
Water 10 01125 g004
Figure 5. Current irrigation quotas of typical crops in Minqin County.
Figure 5. Current irrigation quotas of typical crops in Minqin County.
Water 10 01125 g005
Figure 6. System framework.
Figure 6. System framework.
Water 10 01125 g006
Figure 7. Region water-saving potential calculation flow chart.
Figure 7. Region water-saving potential calculation flow chart.
Water 10 01125 g007
Figure 8. Water distribution and groundwater level changes in each month of 2014.
Figure 8. Water distribution and groundwater level changes in each month of 2014.
Water 10 01125 g008
Figure 9. The relations between the reduction of crop yields and the degree of water deficit.
Figure 9. The relations between the reduction of crop yields and the degree of water deficit.
Water 10 01125 g009
Figure 10. The water savings by different implementation proportions.
Figure 10. The water savings by different implementation proportions.
Water 10 01125 g010
Figure 11. Water savings of different planting structure adjustments.
Figure 11. Water savings of different planting structure adjustments.
Water 10 01125 g011
Figure 12. Irrigation area-scale water-saving potentials of different levels.
Figure 12. Irrigation area-scale water-saving potentials of different levels.
Water 10 01125 g012
Figure 13. Region-scale water-saving potentials of different levels.
Figure 13. Region-scale water-saving potentials of different levels.
Water 10 01125 g013
Table 1. Summary of crop acreage from 2011–2015 in Minqin County (unit: hm2).
Table 1. Summary of crop acreage from 2011–2015 in Minqin County (unit: hm2).
Spring WheatMaizeCottonOil FlexSeed Melon
20117400.045386.6912,826.737160.042106.68
20123853.359560.0511,480.067773.372173.34
20133353.359373.388880.049666.722633.35
20144473.339646.677553.3310,666.672320.00
20154673.579967.175000.2511,513.912473.46
Table 2. The range of the main parameters.
Table 2. The range of the main parameters.
Specific YieldPrecipitation Infiltration CoefficientCanal Leakage CoefficientIrrigation Supply Coefficient
Clay0.02–0.050.08–0.150.08–0.150.08–0.12
Silt 0.05–0.100.12–0.180.12–0.180.12–0.18
Fine sand0.06–0.200.15–0.220.15–0.220.15–0.20
Sand and gravel0.12–0.250.18–0.25--
Sandstone0.04–0.100.05–0.10--
Basalt0.05–0.080.08–0.15--
Sedimentary rocks0.03–0.100.05–0.08--
Table 3. Typical crops’ water product function parameters.
Table 3. Typical crops’ water product function parameters.
Growth Stage12345
Spring wheatDate3.21–4.294.30–5.125.13–6.16.2–6.136.14–7.16
EP (mm)5.552.724.394.0322.72
ETmax (mm)72.6961.04113.4978.76233.29
λ−0.2230.3300.0940.5900.329
Seed melonDate5.1–6.206.21–7.57.6–7.318.1–8.208.21–9.17
EP (mm)12.798.6727.6418.5214.63
ETmax (mm)97.1440.02115.6042.7750.35
λ0.1080.0620.2560.200−0.038
CottonDate4.21–6.226.23–7.167.17–9.149.15–10.20
EP (mm)15.2819.6948.3210.58
ETmax (mm)55.6389.08170.1216.59
λ0.2450.1720.4690.063
maizeDate4.14–5.205.21–6.286.29–7.267.27–9.13
EP (mm)7.0111.5628.3237.44
ETmax (mm)71.13139.54151.58209.26
λ0.1930.1150.0050.100
Oil flexDate4.17–6.96.10–6.186.19–7.57.6–8.27
EP (mm)11.643.029.3452.65
ETmax (mm)176.9949.0695.16196.99
λ0.1720.0840.1020.011
Table 4. Growth stage-time period transformation variable (qijt).
Table 4. Growth stage-time period transformation variable (qijt).
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
Spring wheat10.000.000.350.970.000.000.000.000.000.000.000.00
20.000.000.000.030.390.000.000.000.000.000.000.00
30.000.000.000.000.610.030.000.000.000.000.000.00
40.000.000.000.000.000.400.000.000.000.000.000.00
50.000.000.000.000.000.570.520.000.000.000.000.00
maize10.000.000.000.570.650.000.000.000.000.000.000.00
20.000.000.000.000.350.930.000.000.000.000.000.00
30.000.000.000.000.000.070.840.000.000.000.000.00
40.000.000.000.000.000.000.161.000.420.000.000.00
50.000.000.000.000.000.000.000.000.000.000.000.00
Cotton10.000.000.000.331.000.730.000.000.000.000.000.00
20.000.000.000.000.000.270.520.000.000.000.000.00
30.000.000.000.000.000.000.481.000.470.000.000.00
40.000.000.000.000.000.000.000.000.530.650.000.00
50.000.000.000.000.000.000.000.000.000.000.000.00
Seed melon10.000.000.000.001.000.670.000.000.000.000.000.00
20.000.000.000.000.000.330.160.000.000.000.000.00
30.000.000.000.000.000.000.840.000.000.000.000.00
40.000.000.000.000.000.000.000.650.000.000.000.00
50.000.000.000.000.000.000.000.350.570.000.000.00
Oil flex10.000.000.000.471.000.300.000.000.000.000.000.00
20.000.000.000.000.000.300.000.000.000.000.000.00
30.000.000.000.000.000.400.160.000.000.000.000.00
40.000.000.000.000.000.000.840.870.000.000.000.00
50.000.000.000.000.000.000.000.000.000.000.000.00
Table 5. Variable declaration.
Table 5. Variable declaration.
VariableVariable Meaning DescriptionValue
η 1 Surface water irrigation utilization coefficient0.61
η 2 Groundwater irrigation utilization coefficient0.85
C1Surface water price, CNY/m30.24
C2Groundwater price, CNY/m30.26
αPrecipitation leakage coefficient0.20
βCanal leakage coefficient0.16
β′Irrigation backflow coefficient0.20
β″Well irrigation compensation coefficient0.20
η′Field water delivery efficiency0.85
λ Agricultural water consumption ratio0.44
EGtEvapotranspiration, mm0.00
μSpecific yield 0.16
AArea, 106 hm21.59
H1Initial value of groundwater depth, m18.75
QSurface water supply, 104 m38511.39
GGroundwater availability, 104 m32794.85
Y ¯ Per capita minimum food weight, kg/person420.00
SPopulation, 10424.11
Table 6. Comparison of the indicators before and after optimization.
Table 6. Comparison of the indicators before and after optimization.
Spring WheatSeed MelonCottonMaizeOil Flax
Irrigation (mm)Present390360300420300
Optimization366237238473286
Crop area (hm2)Present447323207553964710,667
Optimization33532633888099709823
Yield (kg/hm2)Present73212250147610,1432200
Optimization72012809153014,6143150
Water productivity (kg/m3)Present2.080.610.442.410.69
Optimization1.780.880.462.620.87
Water benefit (CNY/m3)Present5.004.866.415.564.52
Optimization4.277.046.696.035.65
Table 7. Summary of typical crop yields at different levels of water deficit.
Table 7. Summary of typical crop yields at different levels of water deficit.
Water DeficitYield (kg/hm2)
Spring WheatSeed MelonCottonMaizeOil Flax
5%7158.12676.51457.314,045.53322.1
10%6738.02592.71384.413,319.83256.7
15%6320.62507.01311.312,593.33188.9
20%5906.12419.31238.011,866.03118.5
25%5494.72329.21164.411,137.93045.3
30%5086.52236.61090.610,408.92968.9

Share and Cite

MDPI and ACS Style

Yue, Q.; Zhang, F.; Guo, P. Optimization-Based Agricultural Water-Saving Potential Analysis in Minqin County, Gansu Province China. Water 2018, 10, 1125. https://doi.org/10.3390/w10091125

AMA Style

Yue Q, Zhang F, Guo P. Optimization-Based Agricultural Water-Saving Potential Analysis in Minqin County, Gansu Province China. Water. 2018; 10(9):1125. https://doi.org/10.3390/w10091125

Chicago/Turabian Style

Yue, Qiong, Fan Zhang, and Ping Guo. 2018. "Optimization-Based Agricultural Water-Saving Potential Analysis in Minqin County, Gansu Province China" Water 10, no. 9: 1125. https://doi.org/10.3390/w10091125

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop