Spatial and Temporal Variation in Water Productivity and Grain Water Utilization Assessment of Heilongjiang Province, Northeast China

Introduction

In recent years, rapid economic development and population growth have increased the consumption of water to meet additional demands for food, intensified the competition for water among agriculture, industry,and the environment.Agriculture is the largest waterconsuming sector (FAO, 1994; Rosegrant et al., 2002),and in many developing countries, irrigated agriculture has been expanding rapidly in recent decades.Under the pressures of water scarcity and increasing population growth, China's agricultural production is being challenged to produce more agricultural products despite limited water resources (Brown and Halweil,1998; Kang et al., 2010).Specifically, Heilongjiang Province is an important grain production region located in southeastern China.Analyzing the current agricultural water use and improving water productivity are necessary to meet this challenge.

香豆素对小麦发芽指数、种子萌发、幼苗生长及根系活力有抑制作用，且随处理浓度的增加而增强，肉桂酸、羟基苯甲酸和苜蓿素在处理浓度大于0.010 g/L之后抑制作用逐渐明显。

省界断面监测指标包括水量和水质两类，监测站点分为驻测站（指有人值守站）、自动监测站（指有人看管无人值守站）和巡测断面（指水质监测断面、定时取样实验室化验方式）。

The term "water productivity" (WP) was originally promoted by Molden (1997) and is used to express production per unit of water consumed (Perry et al., 2009).Many spatial-temporal studies of regional or basin-scale WP have been reported in the literature.So far, two major procedures for assessing regional WP have been widely applied.The first uses statistical or model-simulated yield and water use data to assess WP (Garg et al., 2012; Amarasingha et al.,2015; Wang et al., 2014).For instance, Chapagain and Hoekstra (2004) used statistical data from the Food and Agricultural Organization (FAO) to determine the WP of each country in the world from 1997 to 2001.Based on a review of 84 literature sources containing experiments from the past 25 years, Zwart and Bastiaanssen (2004) estimated the average values and global ranges of crop water productivity (CWP)for irrigated wheat, rice, cotton and maize.Igbadun et al.(2006) quantified CWP of a maize crop cultivated under irrigation from field experimental data in the Mkoji subcatchment of the Great Ruaha River Basin in Tanzania.Huang and Li (2010) assessed basin-scale CWP for the staple grain crops rice, wheat, maize and soybean in the major breadbasket basins of China over the period 1997-2004 using a hydro-modelcoupled-statistics approach.Sharma et al.(2016)quantified and mapped long-term spatial-temporal variability in CWP, evapotranspiration water productivity and associated environmental variables at regional scales for maize- and soybean-producing counties in Nebraska.

The second procedure integrates remote sensing(RS) and geographical information system (GIS)technology with models to obtain spatial-temporal expressions of yield and water use and to then assess WP (Zwart et al., 2010a; Cai et al., 2010; Yan and Wu,2014; Campos et al., 2017).Liu et al.(2007) estimated the global production and WP of wheat by developing a GIS-based EPIC model.Zwart et al.(2010b) studied global wheat WP by developing WATPRO model,which is based on RS data combined with normalized difference vegetative index (NDVI) and surface albedo data sets.Yan and Wu (2014) analyzed the CWP of winter wheat based on RS data and found a steady increase in CWP in recent years.Cai et al.(2010) studied rice yield, water consumption and WP in the Ganges basin in India based on RS, statistics and meteorological data.Li et al.(2016) evaluated the spatial pattern of CWP in the middle reaches of the Heihe River Basin using the Aqua Crop model.

The previous studies, however, have been chiefly concerned with single crops (Phogat et al., 2017;Howell et al., 2015) or with a kind of water input.This paper attempted to incorporate the staple grain crops rice, maize and soybean into a single integrated analytical framework using four WP indices to represent different water inputs.Based on data collected from the main irrigation districts of 12 prefecture-level cities in Heilongjiang Province, our study quantified temporal variation and spatial distribution in WP using spatial statistics.Aimed to address the issues associated with grain crop production and water use at the regional scale across Heilongjiang Province.In addition, this paper discussed WP-based policy suggestions for improving agricultural water management in irrigated land and provided suggestions for targeted irrigation programs in various regions of Heilongjiang Province.

Materials and Methods

Study area

Based on Eq.(8), the four WPs of each city were weighted to obtain provincial averages for the four indicators (Fig.2).Fig.2 showed that the four provincial WPs of grain crop in the irrigation area of Heilongjiang Province increased from 2007 to 2012,reaching a maximum in 2012 and then decreasing.Between 2013 and 2015, these values were stable.The decline in WP after 2012 was mainly caused by Heilongjiang Province implementing a project in 2013 to change dryland fields into paddy fields in some irrigation districts, which increased production, but led to a surge in irrigation water consumption.During the study interval, the average annual rates of change in WPg, WPa and WPI for rice were negative, but WPs of maize and soybean showed a slightly increasing trend.The average annual growth rates of WPET were 1.07%for rice, 4.29% for maize and 1.02% for soybean,which indicated that the utilization rates of water resources in the field were improved.

WP at the provincial scale was weighted by the effective irrigation area of each city based on the calculated WP results of each city:

2.4 2组术后24周骨折愈合情况评估 术后24周，2组患者X线检查均有骨痂通过骨折线，患肢无纵向叩击痛，不扶拐能行走3min，对比之前X线片骨折无变形，内固定无松动开裂，无骨不连发生。

Fig.1　Map of Heilongjiang Province

Data resources

Based on the availability of data, the production of rice, maize and soybean in 12 prefecture-level cities in Heilongjiang Province from 2007 to 2015 was selected as the study topic.The Daxing'anling region is not suitable for the growth of food crops due to terrain and climatic conditions and was therefore not included in our analysis.The meteorological data included the highest and the lowest temperature,sunshine hours, wind speed, relative humidity and precipitation, of these 12 cities were obtained from the China Meteorological Data Service Center (http://data.cma.cn/).Grain yield per unit area, county and city land area and other data were from the Heilongjiang Statistical Yearbook.Irrigation water, irrigated area,irrigation water utiliza-tion coefficient and other irrigation data were from the Heilongjiang Water Conservancy Construction Yearbook and Heilongjiang Province irrigation water utilization analysis and analysis results summary.The crop coefficient Kc was determined using crop yields in related studies.

WP indices

WP can be expressed as agricultural production per unit volume of water.The numerator may be expressed in terms of crop yield (kg · hm-2).A number of options are available to define the volume of water per unit of area (m3 · hm-2) in the denominator (Playan and Mateos, 2006).Based on data availability, four in dices were selected and calculated in this study.Eqs.(1)-(4) represented the ratios of grain yield to gross water inflow, generalized agricultural water input,field evapotranspiration (crop water consumption)and irrigation water withdrawal, respectively; the corresponding WP indices were called gross inflow water productivity (WPg), generalized agricultural water productivity (WPa), evapotranspiration water productivity (WPET) and irrigation water productivity(WPI), respectively.The indices differred in scientific connotation and numerical performance, but any of them could be used to quantify the relationship between water utilization and grain production.Gross water inflow (10P+Ig) referred to the maximum quantity of water resources a region could supply within a specific time, so WPg reflected the climate and condition of water resources in the region.Generalized agricultural water input (10Pe+Ig) was the amount of water resources used for agricultural production during the crop growth period, which consisted of effective precipitation and irrigation water in flow (Li and Huang, 2010).Field evapotranspiration was the actual amount of water consumed by crop production,and improving WPET was a direct method of enhancing the efficiency of water use at the field scale.

Where, Y was the yield of grain crops (kg · hm-2); P was the precipitation during crop growth period (mm);Ig was irrigation water diverted (m3 · hm-2); Pe was the effective precipitation during crop growth period(mm); and ETc was growing-season crop evapotranspiration (mm).Pe was the fraction of the total precipitation P as rainfall that was available to the crop and did not run off.Without detailed site-specific information, Pe was very difficult to determine.A simple approximation was provided by the U.S.Department of Agriculture Soil Conservation (Döll and Siebert, 2002).

Where, ET0 was the crop reference evapotranspiration (mm · d-1); Rn was the net radiation at the crop surface (MJ · m-2d); G was the soil heat flux density(MJ · m-2d); T was the mean daily air temperature (℃);ea was the actual vapor pressure (kPa); eb was the saturation vapor pressure (kPa); Δ was the slope vapor pressure curve (kPa · ℃-1); γ was the psychometric constant (kPa · ℃-1); U2 was the wind speed at a height of 2 m (m · s-1); and ETc was actual crop evapotranspiration.Kc was the crop coefficient, which was dimensionless.

Grain evapotranspiration

Daily reference evapotranspiration was computed using the standardized Penman-Monteith (Djaman et al., 2016).The Penman-Monteith method recommended by FAO 56 was used in this paper to estimate ET0 as expressed by Eq.(6).Actual evapotranspiration for a given agricultural crop could be related to the potential evapotranspiration, ET0, according to Eq.(7)(Jenson, 1968).

Where, P10-day and Pe10-day were the 10-day precipitation and effective precipitation (mm), respectively.

Provincial WP

分布式温湿度传感器可以监测环境异常，有异常则报警。动态密码电子锁代替普通的机械锁和电子锁，结合热释电传感器检测非法侵入。动态密码电子锁代替传统锁具锁闭集装箱，其开关状态信息通过锁具内嵌的蓝牙实时传输至处理核心。热释电传感器安装在集装箱内用于检测是否有人进入，检测到有人进入后与主控制器收集到的电子锁的开关状态进行比较，如此时电子锁状态为打开则判定为正常开箱，如强行开锁或未开锁暴力拆解进入则会被判定为非法侵入，此时树莓派控制核心向平台发送报警信息并调用摄像头录制现场视频并发送。安全保障模块工作流程如图2所示。

[2]Doug K.Evaluting the impact of environmental inter protection,a review of three research studies,Comlotion for Education in the outdoors:Research Symposium Proce edings,1996,4（2）.

Where, WPprov.was the provincial water productivity(kg · m-3); n was the number of prefecture-level cities;WPi was WP of city i (kg · m-3); and Si was the effective irrigated area of city i (hm2).

Implications of relationships among WP indices

Spatial distribution of city-scale WP

The gross inflow water resource was the maximum amount of water that flowed into the field in the period of crop growth.However, the field evapotranspiration represented the effective water consumption in the agricultural production process, and thus the amount of water actually required for grain production.Therefore, the ratio of WPg to WPET reflected the generalized water use efficiency for grain production, a dimensionless ratio called the generalized agricultural water resource consumption ratio (α).

Because α was the proportion of water resources actually consumed by farmland out of the total input water resources, 1–α represented the theoretical watersaving potential of regional food production.A higher α indicated a smaller proportion of water wasted in production and a larger amount of water that could be saved or used in other sectors.α indices could be improved on both fronts by reducing the water lost through irrigation water conveyance (evaporation from the water surface and during percolation) and increasing the effective utilization of rainwater.

Comparing the statistical values of 2007 and 2015,the difference between the maximum value and the minimum value was reduced, as was CV among cities,which indicated that the difference in irrigation WP among the cities had narrowed.However, CVs of rice and soybean did not decrease consistently but rather showed maximum values in 2012, indicating that WP's changes in these 12 prefecture-level cities were not synchronous and that the degree of spatial difference changed over time.Many factors affected the utilization of irrigation water; the study area was large, involving different terrains and climate types; and the differences in economic factors and management levels between regions were significant.These issues had a vital impact on irrigation water-use efficiency and WP.The increased regional differences in rice and soybean WPs from 2007 to 2012 were due to these issues, which caused the grain yield and irrigation water consumption of cities to vary.However, the minimum value of WP increased as the cities with lower WP improved their irrigation; thus,the difference in WP among the cities was gradually narrowing.

However, in practical water management, managers were more likely to reduce irrigation water waste than to increase the efficient use of rainwater.Hence,an indicator β was introduced to measure the rate of irrigation water-saving potential with respect to the total generalized agricultural water resources:

Where, β was the proportion of the total amount of irrigation water that could be saved in generalized agricultural water resources, called the irrigation water use potential index.A higher β indicated that irrigation water represented a larger proportion than natural precipitation of the generalized agricultural water resources that were not used for crop evapotranspiration.Reducing β was beneficial to the efficient use of water resources.

Since β was similar to α in that the numerator and denominator were both water inputs, β could also be written as a relationship between WP indicators.Because field evapotranspiration was also called agricultural water consumption and to facilitate rewriting formulas,ETc was approximated as (Ig× η+Pe).

Therefore, Eq.(10) could also be written as:

Where, η was the irrigation water utilization coef ficient (irrigation efficiency), which was a dimensionless ratio of the volume of irrigation water used beneficially to the total volume of irrigation water applied.

According to Eq.(11), β reflected a relationship among the four WP in dices.This relationship reflected the fact that reducing the consumption of water resources for food production, especially irrigation water resources, could improve the efficiency of water use in the region.However, due to differences in natural conditions and water resources in different areas,the urgency of reducing irrigation water input also differred.Therefore, the relationship between WPET and WPI was used to establish an additional index γ:

If γ was below 100%, the amount of irrigation water loss was greater than the field evapotranspiration supplied by precipitation, which might cause increased agricultural water use.Therefore, γ was called the irrigation water reduction priority index.The urgency of reducing β in regions with γ values less than 100%was stronger than in those with γ values above 100%(Cao et al., 2015).

Results and Discussion

Temporal and spatial distribution of water productivity (WP)

Trend analysis of provincial WPs over time

“其他的苦情都是编出来的？”“也不全是，我真失恋了，那天老妈给我介绍的对象看不上我，说我太性感。”“这也算失恋？”

The study region, Heilongjiang Province, is located in the northeastern part of China between 121°13'-135°06'E and 43°26''-53°34''N (Fig.1).Its topography is characterized by mountains to the north and east which surrounds the central great plains.The Songnen Plain to the west and the Sanjiang Plain to the northeast are two of the most important regions of agricultural production in China (Hu et al., 2017).The climate ranges from humid to semi-humid, with an average precipitation of approximately 540 mm/year,and the growing season precipitation (May-September)accounts for 80%-90% of the yearly rainfall.Heilongjiang Province has 15.94 million hectares of arable land, ranking second in China, per capita arable land is three times the country average.Food crops are mainly rice, maize and soybean, which cover more than 90% of the total sown area.Water is essential for agricultural development and food production.China's water resources are abundant, but the per capita water availability is only 2 039.2 m3, less than a quarter of the global level, while the per capita water availability in Heilongjiang Province is 2 129.8 m3, which is near the average for China.The water demands of grain production seriously restrict the distribution of water resources to other industries in Heilongjiang Province.Therefore, studying the grain water productivity (WP)and water use situation in Heilongjiang Province is very important.

Comparing four parts of Fig.2 showed that the annual trends of WPg, WPa and WPI in rice and maize were basically the same, while all these indices were lower in soybean due to its lower yield, which for the same amount of water provided a productivity far lower than those of rice and maize.The differences of WPET among the three kinds of grain were large,which indicated significant differences in the water requirements of these crops during the growth period.There were no significant differences between rice and maize for the first three indices of water productivity,but WPET showed that the water demand of maize was lower than that of rice.

In 2007, the average rice yield was 7 020 kg · hm-2,higher than those of other years.In this year, the water input and field evapotranspiration were also high,and the four rice WP indices were at their lowest.Similarly, the yield of maize was the highest in 2014,but its WP indices were not high.From 2008 to 2012, the yields of rice, maize and soybean increased steadily.During this period, the fluctuation of unit irrigation water was minimal, and WP index values showed an increasing trend.Based on the statistical data of the Yearbook, the above results showed that WP indices depended on the change of grain yield per unit area and the total amount of irrigation water.In addition, changes in rainfall had some influences on WP, but compared to the amount of irrigation water,the impact was small.

紧扣教学内容，需要教师进一步强化学科教学的“文本”意识，根据文本特点，进行设计合作学习。合作学习的运用不是程式化的，应该随着教学内容做到因需选择合作学习。

Statistical analysis of provincial WP over time

As the basis for a numerical analysis of the spatial dispersion and time variation in the four WPs, it could be calculated the major statistics of water productivity at the city scale, including the maximum and minimum values, standard deviations (SDs) and coefficients of variation (CVs).CV was the ratio of each SD to its respective mean and reflected the degree of dispersion in WP over different years.The years 2007, 2012 and 2015 were selected as representative years, as shown in Table 1, because 2007 and 2015 were the starting and ending years of the study interval and 2012 had the largest number of statistical maxima.

Fig.2　Provincial WPs from 2007 to 2015

主题出版是“围绕国家政治、经济、社会、文化等方面的工作大局，就党和国家发生的一些重大事件、重大活动、重大题材、重大理论问题等主题而进行的选题策划和出版活动”。[1]传统的主题出版相对被动，存在选题陈旧、内容空洞、速生速朽等诸多问题，难于达到主题出版的初衷。

Comparing WPs within the same crop, WPs could be ranked by spatial difference as WPI＞WPa＞WPg＞WPET.As the quantity of water input decreased, the inter-regional dispersion increased.Comparing the same indicators among the different crops, the spatial difference ranking was maize＞soybean＞rice.

Annual growth rate of WP at city scale

The annual growth rate of WP in every region of Heilongjiang Province was calculated according to the above theory, and the specific results are shown in Fig.3.

Table 1　Main statistics of four WP indicators in representative years

Grain　Year　Index　Max　Min　SD　CV　Index　Max　Min　SD　CV Rice 2007 WPI 0.792　0.301　0.129　0.221 WPa 0.623　0.270　0.095　0.197 2012　0.916　0.415　0.156　0.221　0.690　0.328　0.113　0.207 2015　0.700　0.423　0.097　0.166　0.538　0.344　0.065　0.143 2007 WPg 0.595　0.265　0.088　0.192 WPET 1.051　0.701　0.112　0.124 2012　0.630　0.305　0.096　0.196　1.501　0.891　0.183　0.158 2015　0.507　0.320　0.062　0.144　1.262　0.855　0.135　0.129 Maize 2007 WPI 0.678　0.222　0.144　0.298 WPa 0.539　0.199　0.109　0.275 2012　0.983　0.413　0.189　0.265　0.772　0.326　0.140　0.255 2015　0.759　0.478　0.097　0.160　0.583　0.389　0.068　0.143 2007 WPg 0.515　0.195　0.101　0.268 WPET 1.329　0.667　0.229　0.224 2012　0.704　0.304　0.120　0.243　2.229　1.175　0.350　0.214 2015　0.546　0.362　0.065　0.145　1.974　1.276　0.199　0.131 Soybean 2007 WPI 0.183　0.064　0.033　0.250 WPa 0.141　0.057　0.023　0.215 2012　0.276　0.100　0.048　0.251　0.208　0.079　0.034　0.234 2015　0.181　0.121　0.022　0.145　0.139　0.098　0.015　0.127 2007 WPg 0.135　0.056　0.021　0.206 WPET 0.396　0.226　0.059　0.195 2012　0.177　0.074　0.028　0.216　0.600　0.325　0.077　0.163 2015　0.130　0.092　0.014　0.132　0.482　0.345　0.047　0.118

Fig.3 showed that during the study interval (2007-2015), WPg and WPa for rice had positive average annual rates of change only in the cities of Hegang City,Qitaihe City and Jiamusi City, while the rest of the cities showed negative growth.The annual growth rate of WPI in 12 cities was between –2.1% and 6.9%.WPET in Daqing City grew the fastest at 4.2% per year, while most other cities grew by 2%.Harbin City, Jixi City, Qitaihe City and Mudanjiang City showed little change, growing by 0.1%, 0.4%, 0.6%and 0.8%, respectively.The annual growth rates of the four WPs in maize were higher than those in rice, and WPs were positive for all cities except for Mudanjiang City.WPs in Hegang City and Jiamusi City grew relatively rapidly, and all the indices in these two cities showed an annual growth rate of more than 6.5%.The average annual growth rate of maize WP was higher than that of rice over the whole study period, which was consistent with the dominant position of maize cultivation in Heilongjiang Province.

任何一种犯罪调查模式的存在都有其自身的基础理论，这是职务犯罪调查的理论指引，也是“如何开展信息化调查”的理论起点。“信息三论”包括信息论、系统论、控制论，从监察立案至调查终结的过程。职务犯罪调查信息化在运用过程中可以将信息三论和侦查学基本原理相融合，从而产生其基本理论框架，包括信息转移原理、信息流原理、信息能动平衡原理。［2］

Fig.3　Annual average growth rate of each indicator from 2007 to 2015

Because of the small annual variation of soybean yield per hectare, the annual growth change of WP in most cities was not obvious.WPI, WPg and WPa of Hegang City increased by 9.1%, 6.7% and 7%,respectively, which were the fastest rates among the cities.The cities with the highest growth rates for WPET were Qiqihar City and Heihe City, with values of 6.3% and 6%, respectively.

According to the above analysis, the areas with rapid growth of WP could be divided into two categories: one included cities with the ability and funding to invest in agricultural production technology and irrigation water management measures, and the other comprised the main grain-producing regions in Heilongjiang Province,where soil, precipitation, temperature and other natural conditions were most suitable for crop growth.

“轻、松、脆、滑”主要是指演奏的技巧表现，这些音乐技巧对演奏者的表现手法提出要求，为音乐本身服务。要求演奏者在技巧的支撑下，抒发个人情怀，表达幽远的意境。然而就演奏者本身也好，就音乐意境也好，其中都不同程度得暗含儒家、道家思想。

The maximum and minimum values of the indicators of the three grain crops were different in each year,but the annual average was more stable and better reflected the management level of an irrigation area.WPs of the three grain crops under study are presented in Fig.4.Comparing different grain WPs at the same water input amount, rice and maize had higher values because of their relatively high yields.Thus, water productivity increased with decreasing water input when comparing different WP indices of different grain crops.

Rice is the most important cereal crop in Heilongjiang Province, which is becoming a national rice basket region due to the superior nutritional quality of the rice, which also has flavors favored by Chinese rice consumers.For rice, high WPs were found in the central and southern regions of Heilongjiang Province and the eastern region, including Harbin City, Suihua City and Shuangyashan City.Harbin City and Suihua City are on the Songnen Plain, and Shuangyashan City is on the Sanjiang Plain.These two plains, possessing relatively favorable biophysical conditions and management practices, registered the highest WPs.Hegang City and Heihe City, which were both insignificant as rice-growing areas, had the lowest rice WPs.

Fig.4　Spatial distribution of water productivity indices in Heilongjiang Province

Maize, the most abundant cereal crop grown in Heilongjiang Province, which accounted for about 40% of the total planted area of food crops, displayed a similar spatial pattern to rice with respect to WP indices, with Harbin City being the highest and Hegang City the lowest.Cities on the Songnen Plain had the largest maize per hectare yield, so the resulting WPs were very high.Since the requirement of maize for water during the growth period was much less than that of rice, the regional WPET of maize was higher, with values in Suihua City, Jiamusi City and Shuangyashan City of more than 1.5 kg · m-3.

WPs of soybean were not consistent with its total production.Heilongjiang Province was the largest and most important soybean producer in China, but the per hectare yield of soybean was low compared to those of rice and maize, resulting in a low WP.The spatial pattern showed an apparent descending gradient stretching from southern to northern cities.Compared to those of rice and maize, WP indices for soybean in Heihe City and Qitaihe City achieved a high yield ranking, because the biophysical conditions of Heihe City and Qitaihe City were more suitable for soybean sowing than for other two grains.

每天我们都会面对着无数的选择，有些事芝麻小事不值得一提，但有些对于我们人生却又较大的影响。我们不断受到无数选择的轰炸，这使我们无法每次都能够做出最恰当的选择。但有个非常有效的技巧——批判性思维。要做到批判性思维有许多方式，但以下四步可以帮助你更好的建立你的批判性思维。

From a regional perspective, in summary, WPs of cities on the Songnen Plain (i.e., Harbin City, Suihua City and Daqing City) were consistently higher.These cities' natural conditions and water resources were suitable for crop growth, especially for rice and maize;these cities were not major soybean producers, but with a reasonable level of irrigation, soybean WP was also high.Cities with moderate values on Sanjiang Plain mainly planted the high-water-consumption grain crop rice; precipitation was abundant in these areas but relatively concentrated, so it cannot be efficiently used to grow grain.Because growing rice in these areas still required a large amount of irrigation water, the irrigation water grain productivity indicators remained moderate.The lower WP values in the mountainous region of Qitaihe City were due to lower yield, which was mainly caused by higher elevations and lower temperatures.In agricultural areas with a semi-arid climate, such as Qiqihar City, dryness combined with soil and water loss was a major constraint for agricultural production.In that reclamation area, sandy soil texture, drought and windy climate resulted in high in filtration and low irrigation water-use efficiency.Hence, WP was lower as a result of higher application of irrigation water.

Water utilization evaluation of grain production

Potential enhancement of regional water-use efficiency

Given the changing agricultural management and climatic conditions, identifying the areas with the highest potential for improved WP management was very important.Eq.(13) was used to calculate the consistency (C) of percent chance of improvement in WPs in the rice-, maize- and soybean-growing districts of cities.

Where, n was number of study years (n=9).This process scales the cumulative WP for each city by nine times the maximum for each city expressed as a percentage.The consistency values (C) were calculated and are listed in Table 2.

Table 2 showed that the consistency of WPI, WPg and WPa for rice were similar in each city, with values between 47% and 85%, while WPET fluctuated between 56% and 81%.The variation for maize was similar to that for rice; the average consistency value of WPET ranged from 50% to 77%, and other three indicators were 40%-84%.The difference among the four indices of soybean was small, and all were 47%-80%.C values of soybean were lower than those of rice and maize in each city, indicating that the soybean crop had a high potential for improving water-use efficiency during the growing period.

Cities with high mean WP and low consistency had the highest potential for improving WP.Table 2 showed that the consistency indicators for four WPs of three crops were lowest in Hegang City, indicating a high potential for the promotion of WP there.The consistency values in Harbin City and Suihua City were higher, indicating that irrigation management and water resource allocation methods were better optimized in those cities.

Analysis of regional water consumption based on WP

Using Eqs.(9)-(11), three indicators were analyzed to describe water resource utilization and efficiency improvement state.These three indicators could be used to explore ways to improve the water-use efficiency of grain production in different irrigated regions.These results are shown in Figs.5 and 6.

Table 2　Calculated consistency values of WP indices by city (%)

City　Harbin Qiqihar　Jixi　Hegang Shuangyashan Daqing　Qitaihe Jiamusi Qitaihe Mudanjiang　Heihe 　Suihua Rice CWPI 84　48　65　47　79　65　70　63　73　64　57　81 CWPg 86　54　68　47　81　73　69　67　72　68　59　81 CWPa　85　53　68　48　81　71　71　67　74　68　59　85 CWPET 77　56　70　57　71　76　81　75　61　71　61　76 Maize CWPI 84　44　59　41　74　62　52　60　62　57　46　79 CWPg　84　47　61　41　74　68　50　62　61　60　47　78 CWPa　81　46　60　40　73　65　51　61　60　59　47　80 CWPET 77　52　64　50　68　73　61　72　53　64　52　76 Soybean CWPI　74　44　59　34　66　47　56　52　68　59　54　69 CWPg 83　54　68　38　74　58　61　61　74　69　61　76 CWPa 74　48　62　35　67　52　57　55　68　62　56　72 CWPET　76　58　72　47　68　60　76　71　66　73　65　74

Fig.5　Provincial α, β and γ from 2007 to 2015

Fig.6　Values of indicators α, β and γ in 12 cities from 2007 to 2015

(1) Utilization rate of generalized agricultural water resources

These results showed that α values of rice, maize and soybean in irrigated farmland during the study period were 50.4%, 37.0% and 46.9%, respectively.The effective utilization rate of generalized agricultural water resources for rice in the irrigated land of Heilongjiang Province was just over 50.0%, while those for maize and soybean were lower.α values of the three crops in each city fluctuated, but overall, the value of α was lower in 2015 than that in 2007.

The spatial distribution analysis revealed that the annual α of rice, maize and soybean in Qitaihe City were the highest in the province.The values of rice α in Harbin City and Shuangyashan City were both approximately 50.0%.The water-saving irrigation technology in Harbin City was of very high quality,and Qitaihe City had superior management of irrigation water input.In Shuangyashan City, irrigation investment funds accounted for 60.0% of the total water conservation construction, and rainfall was abundant.The above factors led to the higher α values in these three cities.Some areas with low α values also had abundant precipitation, but poor agricultural water management resulted in a substantial waste of irrigation water resources and caused the low α indices in these cities.Values of α for maize and soybean were between 20.0% and 40.0%, indicating that the irrigation efficiency of these two crops urgently needed improvement and that emphasis should be placed on reducing the water supply from source irrigation.

(2) Rate of irrigation water-saving potential in the total generalized agricultural water resources

Indicator β was calculated by WPg, WPI, WPa, WPET and the provincial average value of η.The results of this calculation showed that the provincial value of β for rice decreased from 90.2% to 50.9%, while β value for soybean decreased from 85.5% to 50.5%.Large fluctuations were observed in the annual values, likely because the precipitation and the effective precipitation occurred during the growth season fluctuated in each year.β value for maize decreased steadily from 51.9%to 29.0% without major fluctuations during the study period.The utilization efficiency of irrigation water was greatly influenced by irrigation construction investment and irrigation technology.As the utilization coefficient of irrigation water increased with time, the total irrigation water amount decreased, so β values of grain crops in recent years showed a significantly decreasing trend.The annual average β values of the provincial irrigation area were 58.7% for rice, 34.0%for maize and 57.1% for soybean, indicating that in the water resources that were not fully utilized by rice and soybean crops, the proportion of irrigated water was nearly 60.0%, higher than that of precipitation.The lower β for maize indicated that the utilization rate of irrigation water was better for maize crops.

At the city scale, seven cities showed downward trends in β value, while the remaining five showed upward trends.The annual rate of decrease in β value of all the three crops in Mudanjiang City reached more than 5.5%, close to the average for the province.β of rice in Hegang City decreased faster than those of maize and soybean, while in Daqing City, β of soybean decreased more than those of rice and maize.In other cities, the differences in the rate of β change between the three types of grain crop were smaller.In the spatial analysis, regions with higher β were mostly distributed in the western semi-arid and eastern parts of the province.From 2007 to 2008, the western semiarid and eastern regions had higher β values.The high β in the western region was mainly due to reduced rainfall, which caused a high degree of precipitation utilization by crops because almost all precipitation was effective precipitation.In other words, the gross inflow water resources that were not used by crops primarily consisted of irrigation water, so these areas had higher β values.Measures for improving WP in these regions should include the adoption of advanced water-saving irrigation technology to decrease conveyance water loss and the total irrigation diversion.Though precipitation (P) and effective precipitation (Pe) was high during the growth period of the grain crops, irrigation water waste was still a lot,the irrigated WP indices in the cities in the eastern part of the province were low.

(3) Index for urgency of reducing irrigation water use

Overall, the provincial irrigation water γ value was less than 100%, which indicated inefficient use of water resources.Thus, reducing the water resources used for irrigation would be appropriate.Most cityscale values of β were less than 50.0%, meaning that half of the irrigation water could not be effectively used by grain crops, resulting in a great waste of water resources.The values of γ in Harbin City, Shuangyashan City and Qitaihe City were the highest;although γ values in these cities were less than 100%their irrigation water control was better than that of other cities, reflecting a higher management level.Because rice was a paddy-field crop, its irrigation water demand was higher and the values of γ was much higher than those of maize and soybean; thus,irrigation water for rice should be properly controlled.

According to Figs.4 and 6, the cities with high WPs were concentrated in the southern and eastern areas of Heilongjiang Province.The values of α and β in Harbin City, Shuangyashan City and Qitaihe City were close to the provincial average, while their γ values were far higher than those of other cities.Water supply and demand in these areas were more reasonable than in other regions, so the most important factor in raising WP would be to increase crop yields.In some areas with low α and β but higher γ, such as Hegang City, WP was low; to raise WP there, the amount of irrigation water should be reduced while also improving crop yields.In cities with all three indices low, such as Qitaihe City, WP indices were not only determined by water resources but also by crop varieties and yields.Managers could develop targeted irrigation programs according to the different wateruse efficiencies of different crops.

Conclusions

To understand water productivity (WP) in irrigated cropland, this work contributed a collection of actual water input data, a quantification of WP in Heilongjiang Province from four perspectives, a quantitative assessment of the temporal and spatial variation in WP indices, and a discussion of the implications of the grain-water relationship.The conclusions were as below.

(1) The values of WP in Heilongjiang Province first increased and then decreased but showed an overall increase from 2007 to 2012.The trends of the annual growth rates of maize and soybean WP indices were similar and positive for most cities, while the annual growth rate of rice WP was low and negative for most cities.WP showed spatial differences among the irrigated areas in Heilongjiang Province, and the causes of these differences were closely related to changes in grain yield and the total irrigation water consumption.

两者关系为相互补充、相互促进。医院文化对思想政治工作具有引导、规划和促进作用，医院思想政治工作的开展应重视医院文化的积极作用，[3]并善于主动营造良好的医院文化环境，以促进医院思想政治工作的实际效果；思想政治工作能有力促进医院文化建设，保证医院文化建设的正确方向。

(2) Spatially, the cities with high WPs tended to be distributed in the southwestern region and the eastern region, while WP indices in the northern region were lower.Over time, the spatial differences in WP among cities in the province were gradually decreasing.An analysis of the relationships among the four indicators of grain production water evaluation revealed that more than half of the inflow water was not being used effectively by crops.As the provincial irrigation water-use coefficient increased and irrigation wateruse potential decreased over time, irrigation water-use efficiency generally showed a rising trend.However,the efficiency of irrigation water use in each region was still at a relatively low level.

(3) The key to improving WP was to control irrigation water input and improve irrigation management.From the perspective of grain cultivation, the maize sowing area in Heilongjiang Province was relatively large, but the utilization efficiency of water resources was not high during the growth period.Due to low yield, soybean WP was low.To improve the wateruse efficiency of grain production in Heilongjiang Province, it could be recommended that managers adjusted crop irrigation system, controlled rice irrigation water quantity and actively increased the yield per unit area of soybean.

3.1.3 “农业+民俗文化”：文化创意带动产业的相互融合与促进 在现代农业产业规划中，以当地的传统文化、风俗习惯及区域特色为切入点，进一步把握其文化的深层次含义，通过合理策划来为农业生产及展示等活动注入更多活力，充分体现其文化性与地域性。基于现有的农业产业链及三大产业的共同发展，重点发展乡村文化，实现文明创新，有机结合农村产业与乡村旅游、区域文化[6]。通过多元化的产业文化和一目了然的体验形式来深化游客对文化的理解与体会。

References

Amarasingha R P R K, Suriyagoda L D B, Marambe B, et al.2015.Simulation of crop and water productivity for rice (Oryza sativa L.)using APSIM under diverse agro-climatic conditions and water management techniques in Sri Lanka.Agric Water Manag, 160:132-143.

Brown L R, Halweil B.1998.China‘s water shortage could shake world foods security.World Watch, 11(4): 10-21.

Cai X L, Sharma B R.2010.Integrating remote sensing, census and weather data for an assessment of rice yield, water consumption and water productivity in the Indo-Gangetic river basin.Agric Water Manag, 97(2): 309-316.

Campos I, Neale C M U, Arkebauer T J, et al.2017.Water productivity and crop yield: a simplified remote sensing driven operational approach.Agric For Meteorol, doi:10.1016/j.agrformet.2017.07.018

Cao X C, Wang Y B, Wu P T, et al.2015.Water productivity evaluation for grain crops in irrigated regions of China.Ecol Indic, 55:107-117.

Chapagain A K , Hoekstra A Y.2004.Water footprint of nations, vol.1.main report.Value of Water Research Report Series 16, UNESCOIHE, Delft, the Netherlands.

Djaman K, Mel V C, Balde A B, et al. 2016.Evapotranspiration,irrigation water requirement, and water productivity of rice (Oryza sativa L.) in the Sahelian environment. Paddy Water Environ, 15(3):469-482.

Döll P, Siebert S.2002.Global modeling of irrigation water requirements.Water Resour Res, 38(4): 1037-1048.

Food and Agriculture Organization (FAO).1994.The state of food and agriculture 1993.Food and Agriculture Organization, Rome, Italy.

Garg K K , Bharati L, Gaur A, et al.2012.Spatial mapping of agricultural water productivity using the swat model in Upper Bhima catchment, India.Irrig Drain, 61(1): 60-79.

Howell T A, Evett S R, Tolk J A, et al.2015.Evapotranspiration, water productivity and crop coefficients for irrigated sun flower in the U.S.Southern High Plains.Agric Water Manag, 162: 33-46.

Hu Q, Wu W B, Song Q, et al.2017.How do temporal and spectral features matter in crop classification in Heilongjiang Province,China? J Integr Agric, 16(2): 324-336.

Huang F, Li B G.2010.Assessing grain crop water productivity of China using a hydro-model-coupled-statistics approach.Part II:Application in breadbasket basins of China.Agric Water Manag,97(9): 1259-1268.

Igbadun H E, Mahoo H F, Tarimo A K P R, et al.2006.Crop water productivity of an irrigated maize crop in Mkoji sub-catchment of the Great Ruaha River Basin, Tanzania. Agric Water Manag, 85 (1-2):141-150.

Jensen M E.1968.Chapter 1: Water consumption by agricultural plants.Plant Water Consumption and Response.New York,U.S.A.

Kang S Z, Shi W, Zhang J.2000.An improved water-use efficiency for maize grown under regulated deficit irrigation.Field Crops Res, 67(3): 207-214.

Li B G, Huang F.2010.Trends in China's agricultural water use during recent decade using the green and blue water approach.Advan Water Sci, 21(4): 575-583.(in chinese)

Li J, Zhu T, Mao X M, et al.2016.Modeling crop water consumption and water productivity in the middle reaches of Heihe City River Basin.Comput Electron Agric, 123: 242-255.

Liu J G, Wiberg D, Zehnder A J B, et al.2007.Modeling the role of irrigation in winter wheat yield, crop water productivity, and production in China.Irrig Sci, 26(1): 21-33.

Molden D.1997.Accounting for the water use and productivity.SWIM.paper 1.International Irrigation Management Institute, Columbo, Sri Lanka.

Perry C, Steduto P, Allen R G, et al.2009.Increasing productivity in irrigated agriculture: agronomic constraints and hydrological realities.Agric Water Manag, 96(11): 1517-1524.

Phogat V, Skewes M A, McCarthy M G, et al. 2017.Evaluation of crop coefficients, water productivity, and water balance components for wine grapes irrigated at different deficit levels by a sub-surface drip.Agric Water Manag, 180: 22-34.

Playán E, Mateos L.2006.Modernization and optimization of irrigation systems to increase water productivity.Agric Water Manag, 80(1):100-116.

Rosegrant M W, Cai X M, Cline S A.2002.Global water outlook to 2025 averting an impending crisis.In: a 2020 vision for food,agriculture and the environment initiative.International Food Policy Research Institute, Washington, DC, U.S.A.

Sharma V, Irmak S, Djaman K, et al.2016.Large scale spatial and temporal variability in evapotranspiration, crop water use efficiency, and evapotranspiration water use efficiency of irrigated and rainfed maize and soybean. J Irrig Drain Eng, 142: 4015063.doi:10.1061/(ASCE)IR.1943-4774.0000985

Wang Y B, Wu P T, Engel B A, et al.2014.Application of water footprint combined with a unified virtual crop pattern to evaluate crop water productivity in grain production in China.Sci Total Environ, 497-498: 1-9.

Yan N, Wu B F.2014.Integrated spatial-temporal analysis of crop water productivity of winter wheat in Hai Basin.Agric Water Manag, 133(1): 24-33.

Zwart S J, Bastiaanssen W G M, de Fraiture C, et al. 2010.A global benchmark map of water productivity for rainfed and irrigated wheat.Agric Water Manag, 97(10): 1617-1627.

Zwart S J, Bastiaanssen W G M, de Fraiture C, et al.2010.WATPRO:a remote sensing based model for mapping water productivity of wheat.Agric Water Manag, 97(10): 1628-1636.

Zwart S J, Bastiaanssen W G M.2004.Review of measured crop water productivity values for irrigated wheat, rice, cotton and maize.Agric Water Manag, 69(2): 115-133.