Climate change has changed planting structure greatly in cold regions. Studies are needed that understand the relationship between climate change and agriculture in cold regions and to serve as references for studies of the impact of climate change on agriculture in similar areas. This paper uses Heilongjiang Province as a case study; seven test methods and mutual information were used to analyse the variation trend, abrupt changes and relationship between climate and planting structure. The following was concluded. (1) The precipitation trend was not significant; temperature showed a significant upward trend, the minimum temperature showed the sharpest increase. (2) The proportion of area planted in rice and maize showed a significant upward trend. The trend of rice was the most pronounced, the trend of wheat significantly decreased. (3) Abrupt changes in temperature occurred in the 1980s; abrupt changes in wheat were concentrated at the end of the 1990s. (4) The relationship between temperature and planting structure was stronger than that of precipitation, and the relationship between minimum temperature and planting structure was stronger than that of maximum temperature. The results show that temperature variables, especially minimum temperature, are the main factors affecting the change in planting structure in cold regions.

INTRODUCTION

In recent years, with climate change, heat resources in cold areas (high latitudes) have generally increased. As a result, the agricultural climatic zone has moved northward, causing the agricultural planting structure in cold areas to change greatly (Liu et al. 2013a). The agricultural planting structure considered in this paper is the proportion of crop planting area in a year or in a production season (Li & Wang 2010). In general, the increase in temperature will shorten the crop growth period, and it may also lead to drought, pests and other disasters (Liu et al. 2013b; Yang et al. 2013; Lin et al. 2015). For cold regions, because of the low base temperature, the increase in the temperature can increase the grain yield. For example, studies have shown that temperature and rice yield showed a positive correlation in cold regions; as the minimum temperature increased by 1 degree, the rice yield increased by 3.6% (Zhou et al. 2013). Climate warming is not conducive to the production of corn if maize varieties are fixed, and an increase in temperature will lead to the stagnation of corn production. However, for cold areas, an increase in temperature will lead to a prolonged frost-free period, a long growth period and high-yield varieties will increase maize yield. This is also one of the reasons for the continuous increase in the size of the area planted in maize in cold regions against the background of temperature increases (Meng et al. 2013). Therefore, the influence of climate change on agricultural production in cold regions is different from that in other regions, and in particular for the cold regions with agriculture as the main economic activity, the changes in the climate and agricultural planting structure and the relationship between the two have considerable research value. This paper uses Heilongjiang Province as an example because it is the largest grain-producing and highest-latitude province in China. Taking the climate and planting structure as research objects, change trends and abrupt behaviour were analysed, and the relationship was discussed. The results can provide a reference for similar regional climate effects on planting structure.

In recent years, many scholars have studied the relationship between climate change and agricultural production. In a study on the impact of climate change on agricultural production, the relationship between maize planting structure and climate was studied based on the average resource suitability index, and the results showed that the changes in climate affected the planting structure of maize in northeastern China, which significantly expanded the planting of maize to the north and east (Zhao et al. 2015). The effect of extreme temperature on rice yield was quantified, and rice was more sensitive to low-temperature stress than to high-temperature stress (Zhang et al. 2016). The relationship between climate change and wheat yield was studied, and it was concluded that extreme temperature was the most important factor that caused the decline in wheat yield; the warming of the climate had a negative effect on wheat production. The increase in temperature and the decrease in precipitation resulted in regional drought, which further reduced the yield of wheat (Tack et al. 2014). The effects of climate change on the yields of maize, soybean, rice and wheat were studied quantitatively, and a change in the world yield of 21% could be explained by an agro-climatic index (Iizumi & Ramankutty 2016). The effects of integrated climatic factors and other factors on agricultural production have been studied. The effects of climate, groundwater, reservoirs and other factors on agricultural production in Tunisia were comprehensively considered, and the results showed that agriculture in Tunisia is strongly dependent on temperature and precipitation and that groundwater and reservoirs have a positive effect on irrigation agriculture (Zouabi & Peridy 2015). In addition to climate change, the impacts of crop variety and management planning on crop yield were studied, and the results showed that the improved crop variety and management plan compensated for the negative effects of climate change (Wang et al. 2012). In addition, numerical simulations of agricultural production under climate change have been investigated. In one study, the impact of extreme weather on Chinese agricultural production was simulated based on the Reliability Ensemble Average method (Sun et al. 2016). In other research, the effects of climate change on rice growth and yield were simulated under two planting methods and water stress conditions (Sumathi et al. 2014), and the effects of climate change on rice, wheat yield, water and nitrogen balance were simulated (Jalota et al. 2013). Furthermore, studies regarding the identification, optimization and prediction of agricultural planting structure have been conducted. Based on remote sensing data and a spatial production allocation model, three main crops – rice, wheat and maize – were identified in China, and the results showed that maize had higher precision (Tan et al. 2014). In other work, to maximize the efficiency of the Irrigation District, the agricultural planting structure in the Shandong area of China was optimized (Zhang et al. 2013). Another study used four methods to predict the distribution of maize on dry land based on climate and soil conditions and compared the results with the actual spatial distribution (Estes et al. 2013). In addition, based on the sensitivity of wheat to water and temperature, the effects of climate change on rainfed wheat yield were predicted under the IPCC SRES A1FI and B1 climate change scenarios (Mosammam et al. 2016).

Few studies have investigated the relationship between climate change and planting structure in cold regions and main grain-producing areas. With climate change, especially change in temperature, the cold area cropping system and planting structure have undergone great changes, and these changes are more extensive in major grain-producing areas than in other areas. This paper uses Heilongjiang Province as an example because it has the highest latitude, the highest grain yield and the largest area of arable land in China. With precipitation, annual mean temperature, annual mean minimum temperature and annual mean maximum temperature as climatic factors, the proportions of rice, wheat, corn and soybean planting area indicated the planting structure. The change trend and abrupt changes in climate and agricultural planting structure in Heilongjiang Province were analysed; the relationship between climate change and the change of agricultural planting structure in Heilongjiang Province was investigated. The results can provide a reference for the changes in regional climate and planting structure as well as the effects of climate change on the planting structure in cold regions.

MATERIALS AND METHODS

Study area

Heilongjiang Province is located in northeastern China and is the highest-latitude province in China. Heilongjiang Province is located between north latitude 43°26′ and 53°33′ and between east longitude 121°11′ and 135°05′. Its eastern and northern border is with Russia, the western border is with Inner Mongolia, and the southern border is with Jilin Province. The total land area of 473 thousand square kilometres ranks sixth in China. The climate is a cold temperate zone and temperate continental monsoon climate, from the south to the north, and the study area can be divided into temperate and cold temperate according to the temperature index. From the east to the west, the province can be divided into humid areas, semi-humid areas and semi-arid areas according to the degree of drying. The climate characteristics of the whole province are the low temperatures and drought in the spring, warm and rainy conditions in the summer, early frost in the autumn, and a long cold winter, with a short frostless period. The province's average annual rainfall is 400–800 mm, and the annual average temperature is −4–6 °C. Heilongjiang Province is the largest area of cultivated land in China and is one of the world's three most famous black soil belts. The grain-sown area was 14.775 million hectares in 2014, ranking first in the country, of which, 3.997 million hectares was rice, 0.123 million hectares was wheat, 6.642 million hectares was corn and 3.146 million hectares was soybeans, together accounting for 94% of the total sown area. The grain yield of Heilongjiang was 62.442 million tons in 2014, ranking first in the country, and the province is an important commodity grain base in China. Regional profiles are shown in Figure 1.
Figure 1

Regional profiles of Heilongjiang Province.

Figure 1

Regional profiles of Heilongjiang Province.

Data

This study investigated 12 cities in Heilongjiang Province and one administrative office, for a total of 13 prefecture-level cities, as the study area, as shown in Figure 1. Daily precipitation data and temperature data from 13 meteorological stations covering the period of 1961–2010 were collected. For analysis, the daily precipitation data were summed for each year to obtain the total annual precipitation, and the mean daily temperature data were averaged for each year to obtain the average annual temperature. The planting area of rice, wheat, corn and soybean was obtained from the Heilongjiang Statistical Yearbook of 1987–2015. Planting structure was measured as the proportion of crop planting area to total sown area. In the following analysis, the time scale of the data sequences is annual.

Methods

Variation trend

The Mann–Kendall test is recommended by the World Meteorological Organization and has been widely used to study variation trends in hydrological and meteorological sequence data (Mann 1945; Qiu et al. 2016). The method is suitable for non-normal, incomplete and a few abnormal data sequences and does not require the sample to follow a certain distribution. The Mann–Kendall test is calculated as follows: 
formula
1
where and are the j-th and k-th values, respectively; j > k; and n is the length of the sequence: 
formula
2
The standardized test statistic was calculated as follows: 
formula
3
where , the confidence level is α, and indicates a significant change in the sequence; otherwise, there is no significant trend in the sequence. The change trend is determined using Sen's method (Ghosh & Sen 1968): 
formula
4
where indicates that the sequence has a rising trend, and indicates that the sequence has a downward trend.

Detection of change points

To avoid the possibility of biased results, this paper uses the detection method for seven types of change points (Lei et al. 2007; Zhang et al. 2015a). The Mann–Kendall test, the moving F test, the moving T test, the Lee Hamelin test, R/S test, the Brown–Forsythe test, and the sequential cluster test were used for the detection of change points. The mode of seven results was used as the final change point. Brief descriptions of the seven methods follow.

The Mann–Kendall test: The following statistic is defined: 
formula
5
where , , , and . By taking the sequence in reverse order and repeating the above process, we obtain: 
formula
6
If the two curves intersect and this intersection lies between the critical lines, then the intersection point indicates a change point.
The moving F test: and are two sequences before and after the change points. The F statistic is defined as: 
formula
7
where , , , are the sample variances. By moving the point, we obtain a series of F values, and the boundary point that yields the largest F value is the change point.
The moving T test: This test is similar to the F test. The T statistic is defined as: 
formula
8
where are the sample means. By moving the point, we obtain a series of T values, and the boundary point that yields the largest T value is the change point.
The Lee Hamelin test: A posteriori probability density function is defined as follows: 
formula
9
where . The boundary point that yields the largest value is the change point.
The R/S test: The following statistic is defined: 
formula
10
where , , H is the Hearst index. and are two sequences before and after the change points. For the two sequences before and after the dividing point, the and values are calculated: . The boundary point that yields the highest value is the change point.
The Brown–Forsythe test: The following statistic is defined: 
formula
11
where is the sample mean, and is sample variance. The boundary point that yields the highest F value is the change point.
The sequential cluster test: The following statistic is defined: 
formula
12

The boundary point that yields the highest value is the change point.

Correlation

Mutual information represents the amount of information shared between two or more variables, which can then be used to measure the correlation between the two variables. The greater the information shared by the two variables, the stronger the correlation between the variables. The greater amount of mutual information represents a stronger correlation. Furthermore, in addition to measuring a linear relationship, mutual information can also measure a nonlinear relationship, which is a major advantage of mutual information relative to other statistical indicators (Granger & Lin 1994). In addition, it can measure the complex nonlinear relationship between climate and agriculture (Chen et al. 2016). This is also the reason mutual information is used as a measure of the relationship between climate and planting structure in this study. In this paper, the calculation method of mutual information is proposed by using Sharma (2000). The mutual information between X and Y is as follows: 
formula
13
where , and are the estimates of the probability density according to the kernel function method (Basso et al. 2005).

RESULTS AND DISCUSSION

Variation trends of climate and planting structure

Variation trend of climate

The method described in the section ‘Variation trend’ was used to test the trends of precipitation, average temperature, minimum temperature and maximum temperature of 13 cities in Heilongjiang. The results are shown in Figure 2.
Figure 2

Climate variation trend of Heilongjiang Province: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature; ↑ represents a rising trend; ↓ represents a downward trend; filled arrows represent significant trends, and open arrows represent non-significant trends; α = 0.01.

Figure 2

Climate variation trend of Heilongjiang Province: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature; ↑ represents a rising trend; ↓ represents a downward trend; filled arrows represent significant trends, and open arrows represent non-significant trends; α = 0.01.

As shown in Figure 2, in terms of precipitation, with the exception of the trend in Daxinganling, the precipitation trends are not significant. In terms of temperature, the average temperature, the minimum temperature and the maximum temperature of each region exhibited a significant increasing trend. The increases in the average temperature and the minimum temperature were higher than the maximum temperature; in particular, the increase in the minimum temperature was clear. From the regional differences, the rising trend of temperature in the southeast region is higher than in other regions.

The annual average rainfall in each area of Heilongjiang Province is between 400 and 800 mm, and the general pattern is more precipitation in the central and southern parts and less in the east, with the smallest amount in the west and north. There is an uneven distribution of rainfall during the year, mainly concentrated in 6–9 months, accounting for 65%–85% of the full year. The interannual variability in precipitation is greater, and the regularity is not significant. In the temperature change, with the increase in global temperature, the temperature in cold regions also increases, and at the latitude of 45 degrees, the global warming rate is higher than the global average (Meng et al. 2013). The temperature change in Heilongjiang Province is consistent with this general pattern, and the rising trend of minimum temperature is particularly obvious.

Variation trend of planting structure

The planting structure trends of rice, wheat, corn and soybean in different regions of Heilongjiang are shown in Figure 3.
Figure 3

Planting structures in Heilongjiang Province: (a) rice; (b) wheat; (c) maize; (d) soybean; ↑ represents a rising trend; ↓ represents a downward trend; filled arrows represent significant trends, and open arrows represent non-significant trends; α = 0.01.

Figure 3

Planting structures in Heilongjiang Province: (a) rice; (b) wheat; (c) maize; (d) soybean; ↑ represents a rising trend; ↓ represents a downward trend; filled arrows represent significant trends, and open arrows represent non-significant trends; α = 0.01.

As shown in Figure 3, the proportions of rice and maize had a significant increasing trend in the province, whereas the proportion of wheat showed a significant downward trend. For soybean, the planting area in the northern regions of Daxinganling, Heihe, Yichun and Qiqihar significantly increased, while in other areas, the change trends were not significant. In most parts of the province, although there was an increasing trend of rice and maize, the increasing trend of rice was significantly greater than that of maize.

Climate change is one of the factors that has affected the change of planting structure in this region. The increase of temperature caused the agricultural climate belt to shift northward. The temperature of cold regions generally increased, such that rice became increasingly suitable for planting; however, the increase in temperature was not favourable for wheat production. Due to the prolonged frost-free period, high-yield maize varieties become suitable for planting, and the sown area of maize continuously increased. Other factors also affected the planting structure. The economic benefits of rice are higher than those of maize; therefore, the increasing trend of rice is greater than of maize, driven by economic interests. For soybean, due to its lower production capacity and the role of the government in driving soybean cultivation, the change in soybean is more complex and awaits further study.

Abrupt changes in climate and planting structure

Abrupt changes in climate

The method described in the section ‘Detection of change points’ was used to detect the abrupt change point of the climate data sequence, as shown in Figure 4.
Figure 4

Distribution of climate change points in Heilongjiang Province: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature. MK, Mann–Kendall test; MF, moving F test; MT, moving T test; LH, Lee Hamelin test; RS, R/S test; BF, Brown–Forsythe test; SC, sequential cluster test.

Figure 4

Distribution of climate change points in Heilongjiang Province: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature. MK, Mann–Kendall test; MF, moving F test; MT, moving T test; LH, Lee Hamelin test; RS, R/S test; BF, Brown–Forsythe test; SC, sequential cluster test.

As shown in Figure 4, in the southern region, most of the precipitation change points focus around the 1960s, while those in the northern region are mainly concentrated in 2010; with respect to temperature, the change points in each region occur in the 1980s. The average temperature and the minimum temperature variation trends are similar. At the highest temperature, the southern region of Heilongjiang Province, the change is mainly concentrated in the early 1980s, and in other areas, the change is concentrated in the late 1980s.

Before and after the change point, the trend of climate change is shown in Figure 5.
Figure 5

Trends of climate change before and after change points: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature; ↑ represents a rising trend; ↓ represents a downward trend; solid arrows represent significant trends, and empty arrows represent non-significant trends; α = 0.01.

Figure 5

Trends of climate change before and after change points: (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature; ↑ represents a rising trend; ↓ represents a downward trend; solid arrows represent significant trends, and empty arrows represent non-significant trends; α = 0.01.

As shown in Figure 5, in terms of precipitation, the variation trends before and after the change points were not significant in the whole area. The minimum temperature maintained a more significant increase before and after the change points. The average temperature in the northern region increased significantly before the abrupt point, and it was not significant in the other regions. In most parts of the province, the maximum temperature changes before and after the change points were not significant. It is noteworthy that after the change points, the average temperature and the maximum temperature in some areas exhibited a downward trend, but the trend was not significant.

From Figures 4 and 5, it is apparent that there is substantial variation in the precipitation change points and the change trends before and after the change points are not significant. These results indicate that the factors driving precipitation patterns are complex. It is possible that a larger spatial scale and smaller time period than considered here should be used to understand the factors underlying precipitation change. In contrast, the pattern of temperature change is more obvious, change points are more concentrated, and the trends before and after the change points are consistent. These changes are related to global warming. Especially at the lowest temperature, the change trends before and after the point both show strong and significant upward trends, whereas at the maximum temperature, although the trends are increasing ones, they are weaker than those of minimum temperature. These findings indicate that minimum temperature is more sensitive to climate change than is maximum temperature.

To observe the abrupt behaviour and change trend of climate change in the whole study area, the data of various regions of the whole province were integrated and the climate change characteristics of Heilongjiang Province were drawn, as shown in Figure 6.
Figure 6

Characteristics of climate change in Heilongjiang Province (calculation method: centralized data sequence; (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature).

Figure 6

Characteristics of climate change in Heilongjiang Province (calculation method: centralized data sequence; (a) precipitation; (b) average temperature; (c) minimum temperature; (d) maximum temperature).

As shown in Figures 2, 4 and 5, the trend of precipitation was not significant, with an irregular distribution of change points. However, as evident in Figure 6(a), the precipitation in the whole province exhibited periodic variation, which indicates that a shorter study period is more appropriate for studying abrupt changes in precipitation and change trends to identify patterns of change. The average temperature, minimum temperature and maximum temperature showed increasing trends, and abrupt changes were obvious (Figure 6(b)6(d)).

Abrupt changes in planting structure

The method described in the section ‘Detection of change points’ was used to detect the abrupt change point of the planting structure data sequence, as shown in Figure 7.
Figure 7

Distribution of planting structure change points in Heilongjiang Province: (a) rice; (b) wheat; (c) maize; (d) soybean. MK, Mann–Kendall test; MF, moving F test; MT, moving T test; LH, Lee Hamelin test; RS, R/S test; BF, Brown–Forsythe test; SC, sequential cluster test.

Figure 7

Distribution of planting structure change points in Heilongjiang Province: (a) rice; (b) wheat; (c) maize; (d) soybean. MK, Mann–Kendall test; MF, moving F test; MT, moving T test; LH, Lee Hamelin test; RS, R/S test; BF, Brown–Forsythe test; SC, sequential cluster test.

As shown in Figure 7, the rice planting area showed more complexity in the pattern of change points than did the other types of planting areas, and the change in the southern area was mainly concentrated in the 1990s, whereas in the central region, the change was mainly concentrated in approximately 2010. Wheat changes were concentrated in the middle and late 1990s. The change point of maize was mainly concentrated in 2010. Soybean changes in the eastern region were concentrated in 1995, while in other regions, they were around 2010.

The trends of planting structure change before and after the change points are shown in Figure 8.
Figure 8

Trends of planting structure change before and after the change points: (a) rice; (b) wheat; (c) maize; (d) soybean; ↑ represents a rising trend; ↓ represents a downward trend; solid arrows represent significant trends, and open arrows represents non-significant trends; α = 0.01.

Figure 8

Trends of planting structure change before and after the change points: (a) rice; (b) wheat; (c) maize; (d) soybean; ↑ represents a rising trend; ↓ represents a downward trend; solid arrows represent significant trends, and open arrows represents non-significant trends; α = 0.01.

As shown in Figure 8, the change trend before and after the change points and the trend of the total change were consistent. Before and after the change points, rice maintained a strong increasing trend. In addition, the trend of the southern area after the change points was stronger than that before the change point. Wheat maintained a downward trend before and after the change points. Maize also maintained an increasing trend, but its intensity was significantly lower than that of rice. The pattern of soybean before and after the change points was not strong, and most of the changes in the trend were not significant.

From Figures 7 and 8, it is apparent that the change points of rice were not consistent among different regions; however, there were significant increasing trends before and after the change points. These results are due to climate change, which increased the suitability of the cold region for planting rice. In addition, the higher economic benefit of rice contributed to the increase in the rice planting area. For maize, the change trends of some areas after the change points are not significant; excess production and lower prices compared with those of rice are possible reasons. In wheat, there were large downward trends before and after the change points due to the increase of temperature, which was disadvantageous to wheat growth and yield. The change trends of soybean were complex, and the soybean production capacity is low. However, due to concerns over genetically modified food and other issues, soybean is more involved in food security issues. Soybean planting in many areas is controlled by the government, and policy issues are a main factor affecting soybean planting.

To observe the abrupt behaviour and change trend of the planting structure change from the whole study area, the data of various regions of the whole province were integrated, and the climate change characteristics of Heilongjiang Province were plotted, as shown in Figure 9.
Figure 9

Characteristics of planting structure change in Heilongjiang Province (calculation method: centralized data sequence; (a) rice; (b) wheat; (c) maize; (d) soybean).

Figure 9

Characteristics of planting structure change in Heilongjiang Province (calculation method: centralized data sequence; (a) rice; (b) wheat; (c) maize; (d) soybean).

As shown in Figure 9, the province's rice and maize planting areas showed significant increasing trends, which is consistent with the change trend in the province's temperature. The change trend of rice is more pronounced than that of maize, which is consistent with the change in each region. The wheat planting area was obviously decreased, in contrast to the change in temperature. For soybean, the change trend is not strong and a downward trend is evident in recent years, which is consistent with the results of the previous analysis. In contrast, the pattern of change in planting structure is more obvious than that of climate change, which indicates the change of planting structure is dominated by human subjectivity, and the regularity of human activities is more significant.

Relationship between climate and planting structure

The method described in the section ‘Correlation’ was used to analyse the relationship between climate and planting structure, as shown in Figure 10.
Figure 10

Relationship between climate and planting structure in areas of Heilongjiang Province (darker shades represent stronger relationships).

Figure 10

Relationship between climate and planting structure in areas of Heilongjiang Province (darker shades represent stronger relationships).

As shown in Figure 10, with respect to climate, the relationships between each of precipitation and rice, maize and soybean were strongest in the eastern part of Heilongjiang Province, and the relationship between precipitation and wheat was strongest in northern Daxinganling and Heihe. The eastern part of Heilongjiang Province lies within a relatively humid region, and the regional agriculture is more dependent on water resources. The relationship between the average temperature, the minimum temperature and the planting structure was stronger than that with the maximum temperature, and the relationship between temperatures and planting structure in the south was stronger than that in the north, which shows that the temperature, especially the lowest temperature, is the main factor affecting the planting structure in Heilongjiang. For agricultural production, the increase in temperature is not a favourable factor (Tack et al. 2014). However, for a cold area, the increase in temperature can prolong the frost-free period and lengthen the growth period, and high-yield varieties become suitable for planting due to technological advantages that offset the adverse effects of climate change (Meng et al. 2013).

From the perspective of planting structure, the relationships between rice, wheat, maize and climate are stronger than those involving soybean. The effect of temperature on rice was stronger than that of precipitation, and the effects of the average temperature and the minimum temperature on rice were stronger than the effect of maximum temperature. The increasing of temperature is the main factor affecting rice cultivation in the highest-latitude province in China. With the increase in temperature, the rice planting belt in China has moved to the north significantly, making the northern cold region increasingly suitable for rice cultivation, and the results of this paper confirm this trend (Liu et al. 2013b). Wheat planting showed a significant downward trend because increases in temperature are not conducive to the production of wheat (Yang et al. 2013). The changes in maize were closely related to temperature, and the increase in temperature prolonged the growth period of maize, but the increase in corn production only indirectly benefitted from the change in temperature (Meng et al. 2013). Unlike the other three crops, the relationship between soybean and climate is more complex. In reality, the soybean production capacity is low. China relies to a large extent on imports for soybean, and soybean is more closely related to national food security (Yan et al. 2016). Therefore, the change in soybean is more complex, and it is worth studying in depth.

Temperature is the main factor that influences the change in planting structure (Figure 10), and it is generally believed that the rise in temperature is unfavourable to agricultural production but is beneficial in cold regions (Meng et al. 2013; Zhou et al. 2013). Therefore, determining how to use the benefits of climate warming to improve food production in cold regions is a problem that is worth studying. However, the temperature rise will bring increased evaporation and drought risk and aggravate pest problems (Bhagirath et al. 2014; Zhang et al. 2015b). Therefore, the combination of natural factors, such as precipitation, and the threshold effect of temperature on local agriculture is also a direction for future research.

CONCLUSIONS

Using Heilongjiang Province in a cold region of China as a case study, the variation trend, abrupt behaviour and the relationship between climate and planting structure were analysed, and the conclusions are as follows:

  1. With the exception of precipitation change in Daxinganling, precipitation changes in the areas of Heilongjiang Province were not significant. The temperature showed a significant upward trend in the whole area, and the minimum temperature and average temperature rise are higher than the maximum temperature rise; in particular, minimum temperature increases to a great extent. Overall, the rising trend of temperature in the southeast region is higher than that in other regions.

  2. The proportions of rice and maize planting areas showed a significant upward trend, and the trend of rice was the most obvious. Wheat showed a significant downward trend. Soybean in the northern region exhibited a significant upward trend, while the trend was not obvious in other areas.

  3. Change points of precipitation have large regional differences, and the pattern is not obvious. The change points in temperature occurred in the 1980s. The change trend was not obvious in rice and soybean, while the change point in wheat was concentrated in the late 1990s and that in maize was probably around 2010. Before and after the change points, the trends of climate and planting structure change were generally consistent.

  4. The relationship between temperature and planting structure was stronger than the relationship between precipitation and planting structure, and the relationships between planting structure and each of minimum temperature and average temperature were stronger than the relationship between planting structure and maximum temperature. These results reflect the influence of temperature, especially that of minimum temperature, on agricultural production in cold regions. With respect to crops, the relationship between rice and climate change was the closest, showing a significant increasing trend, which indicates that climate change makes cold regions increasingly suitable for rice cultivation.

ACKNOWLEDGEMENTS

The authors thank the National Natural Science Foundation of China (No. 51279031, 51479032, 51579044 and 51609039), the Province Natural Science Foundation of Heilongjiang (No. E201241), the Yangtze River Scholars Support Program of Colleges and Universities in Heilongjiang Province, the Heilongjiang Province Water Conservancy Science and Technology Project (No. 201318, 201503), and the Prominent Young Person of Heilongjiang Province (No. JC201402).

REFERENCES

REFERENCES
Basso
K.
Margolin
A. A.
Stolovitzky
G.
Klein
U.
Dalla-Favera
R.
Califano
A.
2005
Reverse engineering of regulatory networks in human B cells
.
Nature Genetics
37
(
4
),
382
390
.
doi:10.1038/ng1532
.
Bhagirath
S. C.
Kaur
P.
Mahajan
G.
Randhawa
R. K.
Singh
H.
Kang
M. S.
2014
Global warming and its possible impact on agriculture in India
. In:
Advances in Agronomy
,
Vol. 123
(
Sparks
D. L.
, ed.).
Academic Press
,
Burlington, MA
, pp.
65
121
.
Chen
S. A.
Chen
X. G.
Xu
J. T.
2016
Impacts of climate change on agriculture: evidence from China
.
Journal of Environmental Economics and Management
76
,
105
124
.
doi:10.1016/j.jeem.2015.01.005
.
Estes
L. D.
Bradley
B. A.
Beukes
H.
Hole
D. G.
Lau
M.
Oppenheimer
M. G.
Schulze
R.
Tadross
M. A.
Turner
W. R.
2013
Comparing mechanistic and empirical model projections of crop suitability and productivity: implications for ecological forecasting
.
Global Ecology and Biogeography
22
(
8
),
1007
1018
.
doi:10.1111/geb.12034
.
Ghosh
M.
Sen
P. K.
1968
Estimates of the regression coefficient based on Kendall's tau
.
Journal of the American Statistical Association
63
(
324
),
1379
1389
.
Granger
C.
Lin
J. L.
1994
Using the mutual information coefficient to identify lags in nonlinear models
.
Journal of Time Series Analysis
15
(
4
),
371
384
.
doi:10.1111/j.1467-9892.1994.tb00200.x
.
Iizumi
T.
Ramankutty
N.
2016
Changes in yield variability of major crops for 1981–2010 explained by climate change
.
Environmental Research Letters
11
(
3
),
034003
.
doi:10.1088/1748-9326/11/3/034003
.
Jalota
S. K.
Kaur
H.
Kaur
S.
Vashisht
B. B.
2013
Impact of climate change scenarios on yield, water and nitrogen-balance and-use efficiency of rice–wheat cropping system
.
Agricultural Water Management
116
,
29
38
.
doi:10.1016/j.agwat.2012.10.010
.
Lei
H.
Xie
P.
Chen
G.
Li
J.
2007
Comparison and analysis on the performance of hydrological time series change-point testing methods
.
Water Resources and Power
4
,
36
40
(in Chinese)
.
Li
Y.
Wang
C.
2010
Impacts of climate change on crop planting structure in China
.
Advances in Climate Change Research
6
(
2
),
123
129
(in Chinese)
.
Lin
Y.
Wu
W.
Ge
Q.
2015
CERES-Maize model-based simulation of climate change impacts on maize yields and potential adaptive measures in Heilongjiang Province, China
.
The Journal of the Science of Food and Agriculture
95
(
14
),
2838
2849
.
doi:10.1002/jsfa.7024
.
Liu
Z.
Li
Z.
Tang
P.
Li
Z.
Wu
W.
Yang
P.
You
L.
Tang
H.
2013a
Change analysis of rice area and production in China during the past three decades
.
Journal of Geographical Sciences
23
(
6
),
1005
1018
.
doi:10.1007/s11442-013-1059-x
.
Liu
Z.
Zhengguo
L. I.
Tang
P.
Zhipeng
L. I.
Wenbin
W. U.
Yang
P.
You
L.
Tang
H.
2013b
Spatial-temporal changes of rice area and production in China during 1980–2010
.
Acta Geographica Sinica
68
(
5
),
680
693
(in Chinese)
.
Mann
H. B.
1945
Nonparametric tests against trend
.
Econometrica
13
(
3
),
245
259
.
doi:10.2307/1907187
.
Meng
Q.
Hou
P.
Lobell
D. B.
Wang
H.
Cui
Z.
Zhang
F.
Chen
X.
2013
The benefits of recent warming for maize production in high latitude China
.
Climatic Change
122
(
1
),
341
349
.
doi:10.1007/s10584-013-1009-8
.
Mosammam
H. M.
Mosammam
A. M.
Sarrafi
M.
Nia
J. T.
Esmaeilzadeh
H.
2016
Analyzing the potential impacts of climate change on rainfed wheat production in Hamedan Province, Iran, via generalized additive models
.
Journal of Water and Climate Change
7
(
1
),
212
223
.
doi:10.2166/wcc.2015.153
.
Qiu
L. H.
Peng
D. Z.
Xu
Z. X.
Liu
W. F.
2016
Identification of the impacts of climate changes and human activities on runoff in the upper and middle reaches of the Heihe River Basin, China
.
Journal of Water and Climate Change
7
(
1
),
251
262
.
doi:10.2166/wcc.2015.115
.
Sumathi
A.
Mohandass
S.
Ramasamy
S.
2014
Impact of climate change scenario on rice production in two planting methods: a simulation study
.
International Journal of Environmental Science and Technology
12
(
5
),
1539
1548
.
doi:10.1007/s13762-014-0517-z
.
Sun
Q.
Miao
C.
Duan
Q.
2016
Extreme climate events and agricultural climate indices in China: CMIP5 model evaluation and projections
.
International Journal of Climatology
36
(
1
),
43
61
.
doi:10.1002/joc.4328
.
Tack
J.
Barkley
A.
Nalley
L. L.
2014
Heterogeneous effects of warming and drought on selected wheat variety yields
.
Climatic Change
125
(
3
),
489
500
.
doi:10.1007/s10584-014-1185-1
.
Tan
J.
Li
Z.
Yang
P.
Yu
Q.
Zhang
L.
Wu
W.
Tang
P.
Liu
Z.
You
L.
2014
Spatial evaluation of crop maps by the spatial production allocation model in China
.
Journal of Applied Remote Sensing
8
(
1
),
085197
.
doi:10.1117/1.jrs.8.085197
.
Wang
J.
Wang
E.
Yang
X.
Zhang
F.
Yin
H.
2012
Increased yield potential of wheat-maize cropping system in the North China Plain by climate change adaptation
.
Climatic Change
113
(
3
),
825
840
.
doi:10.1007/s10584-011-0385-1
.
Yan
H. R.
Chen
Y. Y.
Ku
H. B.
2016
China's soybean crisis: the logic of modernization and its discontents
.
Journal of Peasant Studies
43
(
2
),
373
395
.
doi:10.1080/03066150.2015.1132205
.
Yang
X.
Tang
X.
Chen
B.
Tian
Z.
Zhong
H.
2013
Impacts of heat stress on wheat yield due to climatic warming in China
.
Progress in Geography
32
(
12
),
1771
1779
(in Chinese)
.
Zhang
Z. Y.
Ma
H. Y.
Li
Q. G.
Wang
X.
Feng
G. X.
2013
Agricultural planting structure optimization and agricultural water resources optimal allocation of Yellow River irrigation area in Shandong Province
.
Desalination and Water Treatment
52
(
13–15
),
2750
2756
.
doi:10.1080/19443994.2013.819167
.
Zhang
Q.
Gu
X.
Singh
V. P.
Kong
D.
Chen
X.
2015a
Spatiotemporal behavior of floods and droughts and their impacts on agriculture in China
.
Global and Planetary Change
131
,
63
72
.
doi:10.1016/j.gloplacha.2015.05.007
.
Zhang
Z.
Song
X.
Tao
F.
Zhang
S.
Shi
W.
2015b
Climate trends and crop production in China at county scale, 1980 to 2008
.
Theoretical and Applied Climatology
123
(
1
),
291
302
.
doi:10.1007/s00704-014-1343-4
.
Zhang
S.
Tao
F.
Zhang
Z.
2016
Changes in extreme temperatures and their impacts on rice yields in southern China from 1981 to 2009
.
Field Crops Research
189
,
43
50
.
doi:10.1016/j.fcr.2016.02.008
.
Zhao
J.
Guo
J.
Xu
Y.
Mu
J.
2015
Effects of climate change on cultivation patterns of spring maize and its climatic suitability in Northeast China
.
Agriculture, Ecosystems and Environment
202
,
178
187
.
doi:10.1016/j.agee.2015.01.013
.
Zhou
Y.
Li
N.
Dong
G.
Wu
W.
2013
Impact assessment of recent climate change on rice yields in the Heilongjiang reclamation area of Northeast China
.
Journal of the Science of Food and Agriculture
93
(
11
),
2698
2706
.
doi:10.1002/jsfa.6087
.