Elsevier

Field Crops Research

Volume 264, 1 May 2021, 108098
Field Crops Research

Analysis of spatio-temporal variation of crop yield in China using stepwise multiple linear regression

https://doi.org/10.1016/j.fcr.2021.108098Get rights and content

Highlights

  • Aggregated yield over all crops were higher in the eastern coastal provinces of China and in south China than those in the inland provinces and west and north China.

  • The average aggregate yield increased in all provinces but the average annual growth rates varied.

  • Most of the regression models explained more than 60 % of the yield variance, except for rice, potato and cotton models.

  • Agricultural management practices (AMP), soil and economic covariates were the most important factors in all models.

  • The model performance for the aggregate yield was different for each province in individual years.

Abstract

With increasing discrepancies between population growth and food production in China, the monitoring of crop yield is essential to support food security policies. However, current studies about spatio-temporal variation of yield mainly focus on the influence of climatic factors on grain crops, and do not explore the contributions of agricultural, environmental and economic factors on crop yield in China. In this study, a large yield dataset, covering 31 provinces and a 38-year period from 1978 to 2015, and related explanatory variables were collected for analyzing the spatio-temporal variation of different yield aggregations using stepwise multiple linear regression. At the national scale, the average aggregate yield increased from 3.04 Mg ha−1 in 1978 to 10.04 Mg ha−1 in 2015. Overall, the average aggregate yield increased in all provinces but the average annual growth rates varied: it was smaller than 2.5 % in Heilongjiang, Guizhou, Beijing, Qinghai and Jilin, more than 4.0 % in Hainan, Guangxi, Ningxia, Hebei and Shaanxi, and between 2.5 % and 4.0 % in other provinces. The spatial patterns of the average yield from 1978 to 2015 were different for different crop aggregations. Most of the regression models explained more than 60 % of the yield variance, except for rice, potato and cotton models. Agricultural management practices, soil and economic covariates were important explanatory variables in all models. Topography and climatic covariates were also important for some of the crop models. The regression model of the aggregate yield for all crops explained 95 % of the yield variance, which was mainly explained by planting area index of vegetables (20 %), followed by farmer income (14 %), planting area index of other crops (orchards 11 %, melons 8 %, sugar 6 %, cereals 6 %), and density of agricultural diesel engines (5 %). Although the regression residual of the aggregate yield model was zero on average, the trends were different in different provinces: most provinces demonstrated a small negative or positive residual; the yield was substantially lower (< -0.20 Mg ha−1 y−1) than predicted by the regression model in three provinces in central China (Hebei, Shaanxi and Anhui) and substantially higher (> 0.20 Mg ha−1 y−1) in four provinces (Shanxi, Shandong, Sichuan and Guangdong). These systematic over- and underpredictions may be caused by other factors, such as plagues, pests, natural hazards, market structures (such as competition for labor or impediments to market access) and farmer’s management skills. With the increasing population and limited agricultural land resources, enhancing economic growth might be an adequate solution to meet the growing demand for food. It can also promote agricultural efficiency in China, certainly when combined with better management practices, crop composition, breeding and planting technologies.

Introduction

China must make strategic decisions on enhancing food production and ensuring food security for 1.4 billion people, and those decisions will have a large effect on agriculture and land use. Over the past 60 years, crop yield in China has increased dramatically, on average by 125 kg ha−1 y−1 (FAO, 2019). For example, the average annual yield increases are 2.5, 1.7 and 3.1 % for maize, rice and wheat, respectively (National Bureau of Statistics of China, 2019). However, there are large spatio-temporal variations of crop yields in China, among others due to climate and soil variation (Chen et al., 2011). Overall, the grain production in eastern China was higher than that in the west of China (Wang et al., 2018a). Therefore, understanding the temporal and spatial variation of crop yield across China is crucial for national food availability and food security.

Many studies were conducted on analyzing the spatio-temporal variation of crop yields in China. High-yield maize was mainly produced in Heilongjiang, Jilin and Liaoning provinces in northeast China, with average yields of 1.07, 1.38 and 1.52 Mg ha−1 in 1961 and increasing rates of 54 kg ha−1 y−1, 81 kg ha−1 y−1 and 67 kg ha−1 y−1 over the past 60 years, respectively (Guo et al., 2017). Rice yield increased by 23 kg ha−1 y−1 since 2000. The highest rice yield was produced in central China with 7.07 Mg ha−1 in 2015; while the lowest rice yield was in south China, with 5.85 Mg ha−1 in 2015 (Wang et al., 2018b). Wheat yield varied from 6.33 Mg ha−1 in north China to 14.80 Mg ha−1in south China. It also increased from west to east along the same latitude (Lv et al., 2017). Sunflower yield was steady from 1985 to 2008 and increased slowly afterwards; in 2015, productive sunflower was mainly distributed in north and central China (Gansu 3.67 Mg ha−1, Hebei 2.82 Mg ha−1, Xinjiang 2.81 Mg ha−1, Inner Mongolia 2.74 Mg ha−1, Ningxia 2.56 Mg ha−1) (Fu et al., 2019).

In order to improve crop yields, exploring the causes of spatio-temporal yield variation is helpful to design policies. Important factors of agricultural production and its variation include technology, genetics, climate, soil, field management practices and associated decisions such as fertilizer application, tillage and crop hybrid selection, irrigation management, row spacing, planting date and depth (Kukal and Irmak, 2018). On a global scale, over 21 % of yield variation could be explained by agro-climatic variation (Iizumi and Ramankutty, 2016). Climatic factors, especially their annual variation, exhibit a stronger overall linkage to changes in late paddy rice yield of China than technological factors (Wang et al., 2016). The warming trend increased rice yield in northeast China and soybean in north and northeast China; however, it decreased maize yield in seven provinces in the central and northeast regions and wheat yield in three provinces (central and northeast China) (Tao et al., 2008). Thus, the relationship between climatic factors and yield is scale-, location- and crop-dependent. Large-scale statistical data and regional climate datasets are important for investigating general response patterns of crop yields to climate change and variation.

Most studies on yield variation addressed either one specific crop (Deng et al., 2019) or main grain crops (Iizumi and Ramankutty, 2016). In such case, the result does not reflect the integral productivity for all crops. Moreover, existing research mostly focused on the influence of climatic factors on yield (e.g. Kukal and Irmak, 2018). Few studies aimed at explaining the yield variation using agricultural, environmental and economic factors, let alone explore the importance of these factors. As a consequence, a better understanding of the contributions of agricultural, environmental and economic factors on yield and yield variation is needed. A provincial-scale evaluation of the spatio-temporal variation for aggregate yield over a long-term period could provide a general overview and the required high-level information for decision makers.

A disadvantage of analysis of aggregate yield variation is that different crops may have very different yields. For instance, the average yield of tomatoes, rice and maize per unit area is quite different and only evaluating the aggregate yield would not recognize these differences. To account for this, it is also useful to run separate analyses for different crop categories and for individual crops. Such analyses do not suffer from the problem that yields from different crops are aggregated, but an important drawback is that the analysis has to be done for many different crop categories, individual crops and perhaps even for different crop varieties. This may yield too detailed and too much information, thus obscuring general patterns. Decision and policy makers in particular need integrated information that show general patterns and trends.

In this research, we collected agricultural, environmental and economic factors that may influence the spatio-temporal variation of yield, and analyzed their correlation with the aggregate yield of all crops in 31 provinces in China, from 1978 to 2015. This study aims at: a) analyzing the space-time patterns of yield aggregations at multiple levels in China with large yield datasets; b) constructing empirical models of different yield aggregations using stepwise linear regression; c) exploring the major explanatory variables of spatio-temporal variation of different yield aggregations and their relative importance; d) analyzing the temporal and spatial patterns of the regression residual (difference between observations and predictions) for the models for yield aggregated over all crops, staples crops, cash crops and three individual crops (maize, rice and wheat).

Section snippets

Yield and explanatory variables

We collected yield data from 1978 to 2015 of 31 provinces of China (excluding Hong Kong, Macao and Taiwan; Chongqing started from 1997). For potatoes, the data were from 1982 to 2015, while watermelon data were from 1996 to 2012. To assess the effect of fertilization policies on temporal variation of yield at provincial scale, we distinguished three fertilization periods: (a) high-yield fertilization (1978–1995), (b) balanced fertilization (1996–2005), (c) soil-test based fertilization

Results

Because of space limitations, this section mainly focuses on the results of the first level modelling (aggregate yield over all crops), and presents fewer results for the level 2 (staples and cash crop categories) and level 3 (individual crops) analyses. More detailed results for levels 2 and 3 are provided in Appendix B.

Spatial and temporal variation of yield

The marked increase of aggregate yield in China between 1978 and 2015 (Fig. 2a) is likely caused by improved crop varieties and increased fertilizer application. However, the average annual growth rates in Heilongjiang, Guizhou, Beijing, Qinghai and Jilin were smaller than those in other provinces. In Heilongjiang and Jilin, constraints from shallow topsoil resulting from long-term continuous cropping, severe black soil loss due to soil erosion and insufficient organic matter input led to

Conclusions

This study analyzed temporal and spatial variation in crop yield aggregations at provincial level in China from 1978 to 2015. Stepwise multiple linear regression was used to explore the relationships between crop yield and agricultural, environmental and economic explanatory variables. The temporal and spatial patterns of yields were different for different levels of crop aggregations. Most of the models had an R-square larger than 0.6, except for rice, potato and cotton. AMP, soil and economic

Funding

This research was supported by the National Key Research & Development Program of China (No. 2016YFD0200101), the National Natural Science Foundation of China (No. 31972515) and the Earmarked fund for the China Agriculture Research System (No. CARS-09-P31).

CRediT authorship contribution statement

Yingxia Liu: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Writing - review & editing, Visualization. Gerard B.M. Heuvelink: Methodology, Software, Resources, Data curation, Writing - review & editing, Visualization, Supervision, Project administration, Funding acquisition. Zhanguo Bai: Conceptualization, Software, Investigation, Resources, Data curation, Writing - review & editing, Visualization, Supervision,

Declaration of Competing Interest

None.

Acknowledgements

The authors thank Luis de Sousa (ISRIC-World Soil Information, The Netherlands) for kindly explaining the SoilGrids parameters and two anonymous reviewers for constructive comments that greatly improved this work.

References (55)

  • Y. Sun et al.

    Changes in the yield and associated photosynthetic traits of dry-land winter wheat (Triticum aestivum L.) from the 1940s to the 2010s in Shaanxi Province of China

    Field Crops Res.

    (2014)
  • D. Wang et al.

    Synthesized use of VisNIR DRS and PXRF for soil characterization: total carbon and total nitrogen

    Geoderma

    (2015)
  • X. Xu et al.

    Quantification of yield gap and nutrient use efficiency of irrigated rice in China

    Field Crops Res.

    (2016)
  • X. Xu et al.

    Spatial variation of attainable yield and fertilizer requirements for maize at the regional scale in China

    Field Crops Res.

    (2017)
  • C. Chen et al.

    Characterizing spatial and temporal variability of crop yield caused by climate and irrigation in the North China Plain

    Theor. Appl. Climatol.

    (2011)
  • Climatic Research Unit. University of East Anglia, Norwich, UK (1978-2015)....
  • T.D. Cook et al.

    Experimental and Quasi-experimental Designs for Generalized Causal Inference

    (2002)
  • N. Deng et al.

    Closing yield gaps for rice self-sufficiency in China

    Nat. Commun.

    (2019)
  • FAO

    Food and Agriculture Organization of the United Nations

    (2019)
  • A.S. Fotheringham et al.

    Geographically Weighted Regression: the Analysis of Spatially Varying Relationships

    (2003)
  • M. Fu et al.

    Temporal and spatial changes of sunflower production in China from 1985 to 2015

    J. Henan Agric. Univ.

    (2019)
  • I. Harris et al.

    Updated high-resolution grids of monthly climatic observations – the CRU TS3.10 Dataset

    Int. J. Climatol.

    (2014)
  • T. Hengl et al.

    SoilGrids250m: global gridded soil information based on machine learning

    PLoS One

    (2017)
  • B. Huang et al.

    Geographically and temporally weighted regression for modeling spatio-temporal variation in house prices

    Int. J. Geogr. Inf. Sci.

    (2010)
  • J.X. Huang et al.

    Standardized cultivation technique of hami melon in hainan greenhouse

    China Vegetables

    (2018)
  • T. Iizumi et al.

    Changes in yield variability of major crops for 1981–2010 explained by climate change

    Environ. Res. Lett.

    (2016)
  • P. Jiang et al.

    Effect of soil and topographic properties on crop yield in a North‐Central corn–soybean cropping system

    Agron. J.

    (2004)
  • Cited by (36)

    View all citing articles on Scopus
    View full text