The study on diversion in grain trade dependence and potential trade relationship prediction under climate shocks

doi:10.3969/j.issn.1004-9479.2026.01.20241054

Abstract

Abstract:

The fluctuation of grain production and supply caused by climate shocks affects the pattern of bilateral trade dependence. Based on the grain trade data of 155 economies from 2001 to 2022, this paper uses the time-varying index random graph model and its separation model to quantitatively analyze the impact of climate shocks on the evolution of grain trade dependence patterns. Furthermore, the stochastic forest link prediction model is used to explore potential future grain trade dependence. The findings are as follows: Grain trade dependency networks are characterized by clear heterogeneous networks and "small worlds", with an overall fragmentation into polycentric, cross-regional trade dependency clusters, with a tendency of alternating growth and decline. Long-term climate change deepens the degree of trade dependence by improving the supply capacity of exporting countries, which is conducive to maintaining the existing dependence relationships. Short-term meteorological shocks will make importing countries tend to stabilize trade links and expand import sources, and the impact is more significant. The effects of long-term climate change have led to crop heterogeneity. Maize and rice have an evolutionary pattern of export expansion and import contraction depending on the network. Resource allocation index is the link motivation for grain trade countries to establish trade dependency links; the future potential food trade dependence mainly relies on the existing export power and form the market expansion effect; China may strengthen trade ties with Ukraine, Argentina and other countries in wheat and corn, expand its rice export markets, and make full use of the potential for soybean export expansion in Paraguay and other countries.

Key words: climate shocks, grain trade dependence network, diversion effect, mechanism of evolution, link prediction

摘要：

气候冲击引发的粮食生产与供给波动影响双边贸易的依赖格局。本研究基于2001—2022年155个经济体的粮食贸易数据，利用时变指数随机图模型及其分离模型，定量分析气候冲击对粮食贸易依赖格局演化的影响，并进一步利用随机森林链路预测模型探究未来潜在的粮食贸易依赖关系。研究发现：①粮食贸易依赖网络呈现明显的异配特征与“小世界”特征，整体分化为多中心、跨区域的贸易依赖联系组团，各组团间存在此消彼长的态势；②长期气候变化通过提升出口国供给能力而加深贸易依赖程度，且有利于维持现有贸易依赖关系，而短期气象冲击则会使进口国趋向于稳固贸易联系并拓宽进口来源，其冲击效果更为显著；③粮食贸易依赖网络长期受气候变化影响，存在作物异质性，玉米和稻米依赖网络分别形成出口扩张，进口收缩的演化格局；④共同贸易伙伴关系的建立对贸易联系的创造作用明显；未来潜在的粮食贸易依赖关系主要依托现有的出口大国，以形成市场扩张效应；中国可同乌克兰、阿根廷等国加强小麦、玉米贸易联系，扩展稻米出口市场，同时充分利用巴拉圭等国的大豆出口潜力拓展市场空间。

关键词: 气候冲击, 粮食贸易依赖网络, 转移效应, 演化机制, 链路预测

Guangjun HUO, Siyuan ZHANG, Qingping ZHANG. The study on diversion in grain trade dependence and potential trade relationship prediction under climate shocks[J]. World Regional Studies, 2026, 35(1): 77-94.

霍广军, 张思源, 张庆萍. 气候冲击下世界粮食贸易依赖转移与潜在贸易关系预测研究[J]. 世界地理研究, 2026, 35(1): 77-94.

Figures/Tables 11

Tab. 1 Control variable description and data source

变量类型	变量名称	描述	数据来源
内生结构变量	edges	连边特征：网络节点间贸易依赖程度	模型计算获得
	mutual	互惠性：网络中两节点间存在双向联系的可能，反映了贸易伙伴间互相作为进口来源国并形成双向进口依赖的程度
	gwidegree	偏好依附性：网络中单个节点i接收多节点j发出的连边而形成的结构，即网络中某一经济体作为进口国拥有广泛进口来源
	gwesp	传递性：一对网络节点间存在一个或若干中间节点而形成贸易依赖关系，并受第三方中介影响还可能会形成直接贸易依赖关系
节点属性变量	Hurs	相对湿度：研究期内相对湿度值，作为气候冲击控制变量	ERA5数据集
	WSDI	暖日持续日数：每年至少连续6天达到日最高气温>90%分位数的天数，作为气候冲击控制变量	ERA5数据集
	PGDP	网络中各经济体人均国内生产总值	World Bank数据库
	PLand	网络中各经济体人均耕地规模
	PWater	网络中各经济体人均可再生淡水资源量
网络协变量	Dist	经济体间的地理距离	CEPII数据库
	Comlang	经济体间语言相近性
	Colony	经济体间历史殖民关系

Tab. 2 Growth characteristics of grain trade dependency network

年份	层次性		匹配性		传输性		集聚性
年份	入度分布	出度分布	入度关联	出度关联	平均路径	连边数量	全局集聚系数	平均集聚系数
2001	-3.369 9	-7.469 9	0.004 8	-0.116 0	2.296 8	1 825	0.221 1	0.533 1
2004	-4.831 6	-8.060 5	0.023 6	-0.123 7	2.176 7	2 049	0.225 7	0.501 0
2007	-4.765 9	-1.737 8	-0.014 8	-0.139 7	2.131 9	2 317	0.243 9	0.549 8
2010	-3.981 7	-2.753 6	-0.039 4	-0.147 6	2.113 9	2 659	0.259 8	0.547 2
2013	-4.400 0	-6.510 1	-0.077 5	-0.171 4	2.012 7	2 873	0.276 4	0.565 1
2016	-4.266 3	-6.768 8	-0.088 3	-0.179 4	1.984 6	3 045	0.283 0	0.580 4
2019	-4.477 5	-7.088 9	-0.075 0	-0.174 8	2.044 6	3 064	0.292 6	0.591 9
2022	-3.701 7	-6.173 9	-0.093 2	-0.189 2	1.889 3	2 712	0.345 5	0.594 8

Fig. 1 Evolution of the communities in the grain trade dependency network in 2001， 2011 and 2021

Tab. 3 Baseline regression

变量		长期				短期
变量		（1）	（2）	（3）	（4）	（5）	（6）	（7）	（8）
网络结构特征	edges	-8.989 1^***	-8.342 3^***	-6.425 1^***	-5.974 5^***	-9.122 0^***	-8.229 8^***	-6.597 4^***	-5.969 2^***
	edges	(0.201 1)	(0.208 0)	(0.216 1)	(0.225 2)	(0.156 0)	(0.160 7)	(0.167 1)	(0.174 3)
	mutual		1.597 6^***	1.467 9^***	1.095 7^***		1.577 5^***	1.449 1^***	1.062 0^***
	mutual		(0.020 2)	(0.020 3)	(0.021 0)		(0.020 0)	(0.020 1)	(0.020 9)
	gwidegree			-6.699 4^***	-6.586 3^***			-6.451 9^***	-6.271 5^***
	gwidegree			(0.415 6)	(0.537 0)			(0.396 5)	(0.525 7)
	gwesp				1.678 3^***				1.670 0^***
	gwesp				(0.020 9)				(0.020 5)
接收者效应	$t a s i t$	-0.010 9^***	-0.009 2^***	-0.012 6^***	-0.009 7^***	-0.011 4	0.083 1^***	0.072 6^***	0.079 9^***
	$t a s i t$	(0.001 4)	(0.001 5)	(0.001 5)	(0.001 6)	(0.012 3)	(0.012 9)	(0.012 9)	(0.013 3)
	$p r i t$	0.000 1^**	-0.000 0	0.000 0	0.000 0	-0.003 9^***	-0.003 7^***	-0.005 3^***	-0.003 5^***
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 6)	(0.000 6)	(0.000 6)	(0.000 6)
	$h u r s i t$	-0.004 7^***	-0.003 7^***	-0.006 5^***	-0.004 3^***	-0.031 8^***	-0.047 0^***	-0.035 8^***	-0.034 5^***
		(0.001 0)	(0.001 1)	(0.001 1)	(0.001 2)	(0.006 0)	(0.006 3)	(0.006 5)	(0.006 7)
	$w s d i i t$	-0.048 9^***	-0.068 8^***	-0.075 5^***	-0.078 3^***	0.001 6	0.000 3	0.000 3	-0.000 2
	$w s d i i t$	(0.004 1)	(0.004 2)	(0.004 3)	(0.004 4)	(0.000 8)	(0.000 9)	(0.000 9)	(0.000 9)
	$l n p g d p i t$	0.336 5^***	0.240 3^***	0.157 4^***	0.137 6^***	0.359 5^***	0.253 0^***	0.175 9^***	0.154 3^***
	$l n p g d p i t$	(0.007 1)	(0.007 3)	(0.007 6)	(0.007 9)	(0.006 0)	(0.006 3)	(0.006 4)	(0.006 7)
	$l n p w a t e r i t$	0.023 3^***	0.012 6^**	0.002 3	0.004 6	0.019 5^***	0.004 9	-0.002 2	-0.002 0
	$l n p w a t e r i t$	(0.004 3)	(0.004 5)	(0.004 5)	(0.004 7)	(0.004 3)	(0.004 5)	(0.004 5)	(0.004 7)
	$l n p l a n d i t$	0.188 8^***	0.090 3^***	-0.019 0	-0.029 7^**	0.195 8^***	0.078 5^***	-0.031 8^***	-0.050 1^***
	$l n p l a n d i t$	(0.008 5)	(0.008 6)	(0.010 1)	(0.010 5)	(0.008 0)	(0.008 1)	(0.009 5)	(0.009 9)
发送者效应	$t a s j t$	0.001 5	0.004 0^**	0.005 0^**	0.005 3^**	-0.397 9^***	-0.426 5^***	-0.427 0^***	-0.414 9^***
	$t a s j t$	(0.001 4)	(0.001 5)	(0.001 5)	(0.001 6)	(0.012 5)	(0.013 1)	(0.013 2)	(0.013 4)
	$p r j t$	0.000 3^***	0.000 3^***	0.000 3^***	0.000 4^***	-0.005 2^***	-0.005 3^***	-0.005 2^***	-0.012 0^***
	$p r j t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 6)	(0.000 6)	(0.000 6)	(0.000 7)
	$h u r s j t$	-0.007 1^***	-0.007 5^***	-0.008 3^***	-0.021 6^***	0.032 6^***	0.041 8^***	0.047 9^***	0.029 5^***
	$h u r s j t$	(0.001 0)	(0.001 1)	(0.001 1)	(0.001 2)	(0.005 8)	(0.006 0)	(0.006 0)	(0.006 2)
	$w s d i j t$	0.128 8^***	0.142 2^***	0.158 5^***	0.139 7^***	0.010 1^***	0.010 5^***	0.010 9^***	0.011 2^***
	$w s d i j t$	(0.003 9)	(0.004 0)	(0.004 1)	(0.004 3)	(0.000 8)	(0.000 9)	(0.000 9)	(0.001 0)
	$l n p g d p j t$	0.471 2^***	0.419 8^***	0.437 9^***	0.345 1^***	0.504 7^***	0.446 7^***	0.461 3^***	0.305 9^***
	$l n p g d p j t$	(0.007 4)	(0.007 6)	(0.007 8)	(0.008 2)	(0.006 1)	(0.006 3)	(0.006 4)	(0.006 7)
	$l n p w a t e r j t$	0.052 5^***	0.050 2^***	0.049 6^***	0.047 5^***	0.056 4^***	0.055 4^***	0.055 1^***	0.054 4^***
	$l n p w a t e r j t$	(0.004 4)	(0.004 6)	(0.004 6)	(0.004 8)	(0.004 4)	(0.004 6)	(0.004 6)	(0.004 8)
	$l n p l a n d j t$	0.487 7^***	0.481 0^***	0.489 6^***	0.436 8^***	0.525 0^***	0.516 9^***	0.524 7^***	0.430 5^***
	$l n p l a n d j t$	(0.009 8)	(0.010 3)	(0.010 4)	(0.010 7)	(0.009 1)	(0.009 6)	(0.009 7)	(0.010 0)
网络协变量	dist	-0.696 7^***	-0.551 2^***	-0.566 1^***	-0.557 2^***	-0.723 7^***	-0.575 0^***	-0.583 3^***	-0.553 0^***
	dist	(0.009 5)	(0.010 1)	(0.010 3)	(0.010 7)	(0.009 3)	(0.009 9)	(0.010 0)	(0.010 3)
	comlang	0.752 8^***	0.604 3^***	0.589 8^***	0.463 1^***	0.626 1^***	0.505 0^***	0.492 2^***	0.413 8^***
	comlang	(0.023 3)	(0.024 3)	(0.024 7)	(0.026 1)	(0.022 7)	(0.023 7)	(0.024 0)	(0.025 5)
	colony	1.163 3^***	0.899 4^***	0.989 8^***	0.957 9^***	1.146 3^***	0.875 9^***	0.959 4^***	0.898 5^***
	colony	(0.048 5)	(0.051 2)	(0.052 9)	(0.055 7)	(0.048 1)	(0.050 9)	(0.052 5)	(0.055 1)

Tab. 3 Baseline regression

变量		长期				短期
变量		（1）	（2）	（3）	（4）	（5）	（6）	（7）	（8）
网络结构特征	edges	-8.989 1^***	-8.342 3^***	-6.425 1^***	-5.974 5^***	-9.122 0^***	-8.229 8^***	-6.597 4^***	-5.969 2^***
	edges	(0.201 1)	(0.208 0)	(0.216 1)	(0.225 2)	(0.156 0)	(0.160 7)	(0.167 1)	(0.174 3)
	mutual		1.597 6^***	1.467 9^***	1.095 7^***		1.577 5^***	1.449 1^***	1.062 0^***
	mutual		(0.020 2)	(0.020 3)	(0.021 0)		(0.020 0)	(0.020 1)	(0.020 9)
	gwidegree			-6.699 4^***	-6.586 3^***			-6.451 9^***	-6.271 5^***
	gwidegree			(0.415 6)	(0.537 0)			(0.396 5)	(0.525 7)
	gwesp				1.678 3^***				1.670 0^***
	gwesp				(0.020 9)				(0.020 5)
接收者效应	$t a s i t$	-0.010 9^***	-0.009 2^***	-0.012 6^***	-0.009 7^***	-0.011 4	0.083 1^***	0.072 6^***	0.079 9^***
	$t a s i t$	(0.001 4)	(0.001 5)	(0.001 5)	(0.001 6)	(0.012 3)	(0.012 9)	(0.012 9)	(0.013 3)
	$p r i t$	0.000 1^**	-0.000 0	0.000 0	0.000 0	-0.003 9^***	-0.003 7^***	-0.005 3^***	-0.003 5^***
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 6)	(0.000 6)	(0.000 6)	(0.000 6)
	$h u r s i t$	-0.004 7^***	-0.003 7^***	-0.006 5^***	-0.004 3^***	-0.031 8^***	-0.047 0^***	-0.035 8^***	-0.034 5^***
		(0.001 0)	(0.001 1)	(0.001 1)	(0.001 2)	(0.006 0)	(0.006 3)	(0.006 5)	(0.006 7)
	$w s d i i t$	-0.048 9^***	-0.068 8^***	-0.075 5^***	-0.078 3^***	0.001 6	0.000 3	0.000 3	-0.000 2
	$w s d i i t$	(0.004 1)	(0.004 2)	(0.004 3)	(0.004 4)	(0.000 8)	(0.000 9)	(0.000 9)	(0.000 9)
	$l n p g d p i t$	0.336 5^***	0.240 3^***	0.157 4^***	0.137 6^***	0.359 5^***	0.253 0^***	0.175 9^***	0.154 3^***
	$l n p g d p i t$	(0.007 1)	(0.007 3)	(0.007 6)	(0.007 9)	(0.006 0)	(0.006 3)	(0.006 4)	(0.006 7)
	$l n p w a t e r i t$	0.023 3^***	0.012 6^**	0.002 3	0.004 6	0.019 5^***	0.004 9	-0.002 2	-0.002 0
	$l n p w a t e r i t$	(0.004 3)	(0.004 5)	(0.004 5)	(0.004 7)	(0.004 3)	(0.004 5)	(0.004 5)	(0.004 7)
	$l n p l a n d i t$	0.188 8^***	0.090 3^***	-0.019 0	-0.029 7^**	0.195 8^***	0.078 5^***	-0.031 8^***	-0.050 1^***
	$l n p l a n d i t$	(0.008 5)	(0.008 6)	(0.010 1)	(0.010 5)	(0.008 0)	(0.008 1)	(0.009 5)	(0.009 9)
发送者效应	$t a s j t$	0.001 5	0.004 0^**	0.005 0^**	0.005 3^**	-0.397 9^***	-0.426 5^***	-0.427 0^***	-0.414 9^***
	$t a s j t$	(0.001 4)	(0.001 5)	(0.001 5)	(0.001 6)	(0.012 5)	(0.013 1)	(0.013 2)	(0.013 4)
	$p r j t$	0.000 3^***	0.000 3^***	0.000 3^***	0.000 4^***	-0.005 2^***	-0.005 3^***	-0.005 2^***	-0.012 0^***
	$p r j t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 6)	(0.000 6)	(0.000 6)	(0.000 7)
	$h u r s j t$	-0.007 1^***	-0.007 5^***	-0.008 3^***	-0.021 6^***	0.032 6^***	0.041 8^***	0.047 9^***	0.029 5^***
	$h u r s j t$	(0.001 0)	(0.001 1)	(0.001 1)	(0.001 2)	(0.005 8)	(0.006 0)	(0.006 0)	(0.006 2)
	$w s d i j t$	0.128 8^***	0.142 2^***	0.158 5^***	0.139 7^***	0.010 1^***	0.010 5^***	0.010 9^***	0.011 2^***
	$w s d i j t$	(0.003 9)	(0.004 0)	(0.004 1)	(0.004 3)	(0.000 8)	(0.000 9)	(0.000 9)	(0.001 0)
	$l n p g d p j t$	0.471 2^***	0.419 8^***	0.437 9^***	0.345 1^***	0.504 7^***	0.446 7^***	0.461 3^***	0.305 9^***
	$l n p g d p j t$	(0.007 4)	(0.007 6)	(0.007 8)	(0.008 2)	(0.006 1)	(0.006 3)	(0.006 4)	(0.006 7)
	$l n p w a t e r j t$	0.052 5^***	0.050 2^***	0.049 6^***	0.047 5^***	0.056 4^***	0.055 4^***	0.055 1^***	0.054 4^***
	$l n p w a t e r j t$	(0.004 4)	(0.004 6)	(0.004 6)	(0.004 8)	(0.004 4)	(0.004 6)	(0.004 6)	(0.004 8)
	$l n p l a n d j t$	0.487 7^***	0.481 0^***	0.489 6^***	0.436 8^***	0.525 0^***	0.516 9^***	0.524 7^***	0.430 5^***
	$l n p l a n d j t$	(0.009 8)	(0.010 3)	(0.010 4)	(0.010 7)	(0.009 1)	(0.009 6)	(0.009 7)	(0.010 0)
网络协变量	dist	-0.696 7^***	-0.551 2^***	-0.566 1^***	-0.557 2^***	-0.723 7^***	-0.575 0^***	-0.583 3^***	-0.553 0^***
	dist	(0.009 5)	(0.010 1)	(0.010 3)	(0.010 7)	(0.009 3)	(0.009 9)	(0.010 0)	(0.010 3)
	comlang	0.752 8^***	0.604 3^***	0.589 8^***	0.463 1^***	0.626 1^***	0.505 0^***	0.492 2^***	0.413 8^***
	comlang	(0.023 3)	(0.024 3)	(0.024 7)	(0.026 1)	(0.022 7)	(0.023 7)	(0.024 0)	(0.025 5)
	colony	1.163 3^***	0.899 4^***	0.989 8^***	0.957 9^***	1.146 3^***	0.875 9^***	0.959 4^***	0.898 5^***
	colony	(0.048 5)	(0.051 2)	(0.052 9)	(0.055 7)	(0.048 1)	(0.050 9)	(0.052 5)	(0.055 1)

Tab. 4 STERGM of long-term impact

变量		形成模型				解除（或持续）模型
变量		(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
网络结构特征	$e d g e s$	-6.362 4^***(0.364 6)	—5.904 3^***(0.355 0)	—4.256 7^*** (0.361 8)	—5.051 2^*** (0.350 0)	—3.428 3^*** (0.482 5)	—3.279 2^*** (0.483 0)	—2.484 1^*** (0.459 3)	—3.510 3^*** (0.477 9)
	$m u t u a l$		1.104 5^***	1.032 5^***	0.665 4^***		0.975 8^***	0.961 8^***	0.867 0^***
	$m u t u a l$		(0.040 2)	(0.038 5)	(0.040 6)		(0.050 4)	(0.051 5)	(0.052 0)
	$g w i d e g r e e$			—4.265 2^***	—2.348 7^***			—4.413 2^***	—3.705 3^***
	$g w i d e g r e e$			(0.169 4)	(0.169 0)			(0.183 7)	(0.191 3)
	$g w e s p$				2.082 7^***				0.988 5^***
	$g w e s p$				(0.078 2)				(0.081 7)
接收者效应	$t a s i t$	—0.004 5	—0.002 2	—0.006 7^**	—0.003 0	—0.003 0	—0.002 1	—0.005 4	—0.005 5
	$t a s i t$	(0.002 6)	(0.002 6)	(0.002 5)	(0.002 5)	(0.003 5)	(0.003 6)	(0.003 4)	(0.003 4)
	$p r i t$	0.000 1^**	0.000 1	0.000 1^*	0.000 1^*	—0.000 1^*	—0.000 2^***	—0.000 1^*	—0.000 1^*
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)
发送者效应	$t a s j t$	—0.020 2^***	—0.019 4^***	—0.019 5^***	—0.018 4^***	0.022 1^***	0.024 9^***	0.026 4^***	0.025 9^***
	$t a s j t$	(0.002 5)	(0.002 6)	(0.002 6)	(0.002 3)	(0.003 6)	(0.003 7)	(0.003 7)	(0.003 6)
	$p r j t$	0.000 2^***	0.000 2^***	0.000 2^***	0.000 2^***	0.000 0	—0.000 0	0.000 0	0.000 0
	$p r j t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 1)	(0.000 1)	(0.000 1)	(0.000 1)

Tab. 4 STERGM of long-term impact

变量		形成模型				解除（或持续）模型
变量		(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
网络结构特征	$e d g e s$	-6.362 4^***(0.364 6)	—5.904 3^***(0.355 0)	—4.256 7^*** (0.361 8)	—5.051 2^*** (0.350 0)	—3.428 3^*** (0.482 5)	—3.279 2^*** (0.483 0)	—2.484 1^*** (0.459 3)	—3.510 3^*** (0.477 9)
	$m u t u a l$		1.104 5^***	1.032 5^***	0.665 4^***		0.975 8^***	0.961 8^***	0.867 0^***
	$m u t u a l$		(0.040 2)	(0.038 5)	(0.040 6)		(0.050 4)	(0.051 5)	(0.052 0)
	$g w i d e g r e e$			—4.265 2^***	—2.348 7^***			—4.413 2^***	—3.705 3^***
	$g w i d e g r e e$			(0.169 4)	(0.169 0)			(0.183 7)	(0.191 3)
	$g w e s p$				2.082 7^***				0.988 5^***
	$g w e s p$				(0.078 2)				(0.081 7)
接收者效应	$t a s i t$	—0.004 5	—0.002 2	—0.006 7^**	—0.003 0	—0.003 0	—0.002 1	—0.005 4	—0.005 5
	$t a s i t$	(0.002 6)	(0.002 6)	(0.002 5)	(0.002 5)	(0.003 5)	(0.003 6)	(0.003 4)	(0.003 4)
	$p r i t$	0.000 1^**	0.000 1	0.000 1^*	0.000 1^*	—0.000 1^*	—0.000 2^***	—0.000 1^*	—0.000 1^*
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)
发送者效应	$t a s j t$	—0.020 2^***	—0.019 4^***	—0.019 5^***	—0.018 4^***	0.022 1^***	0.024 9^***	0.026 4^***	0.025 9^***
	$t a s j t$	(0.002 5)	(0.002 6)	(0.002 6)	(0.002 3)	(0.003 6)	(0.003 7)	(0.003 7)	(0.003 6)
	$p r j t$	0.000 2^***	0.000 2^***	0.000 2^***	0.000 2^***	0.000 0	—0.000 0	0.000 0	0.000 0
	$p r j t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 1)	(0.000 1)	(0.000 1)	(0.000 1)

Tab. 5 STERGM of short-term impact

变量		形成模型				解除（或持续）模型
变量		(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
网络结构特征	$e d g e s$	-7.972 6^***	-7.345 7^***	-5.987 1^***	-6.429 1^***	-1.816 7^***	-1.336 2^***	-0.733 6^*	-1.814 3^***
	$e d g e s$	(0.290 4)	(0.286 5)	(0.285 4)	(0.288 3)	(0.353 6)	(0.355 3)	(0.356 5)	(0.360 3)
	$m u t u a l$		1.111 8^***	1.041 1^***	0.678 1^***		0.995 0^***	0.982 1^***	0.887 6^***
	$m u t u a l$		(0.040 3)	(0.039 2)	(0.040 0)		(0.053 2)	(0.052 3)	(0.052 3)
	$g w i - d e g r e e$			-4.319 9^***	-2.408 0^***			-4.562 8^***	-3.856 6^***
	$g w i - d e g r e e$			(0.167 6)	(0.171 7)			(0.181 7)	(0.196 1)
	$g w e s p$				2.053 9^***				0.977 9^***
	$g w e s p$				(0.078 2)				(0.082 5)
接收者效应	$t a s i t$	-0.045 1	0.017 8	0.023 1	0.051 5^*	0.037 4	0.114 0^***	0.097 3^***	0.104 6^***
	$t a s i t$	(0.023 3)	(0.023 4)	(0.022 9)	(0.022 0)	(0.029 6)	(0.030 5)	(0.029 6)	(0.029 9)
	$p r i t$	-0.007 3^***	-0.006 9^***	-0.007 0^***	-0.003 8^***	0.006 0^***	0.006 1^***	0.007 8^***	0.007 3^***
	$p r i t$	(0.001 1)	(0.001 1)	(0.001 1)	(0.001 0)	(0.001 4)	(0.001 4)	(0.001 4)	(0.001 4)
发送者效应	$t a s j t$	-0.219 4^***	-0.233 4^***	-0.241 9^***	-0.205 4^***	-0.331 1^***	-0.346 4^***	-0.367 4^***	-0.356 8^***
	$t a s j t$	(0.022 8)	(0.022 9)	(0.022 9)	(0.021 8)	(0.028 7)	(0.029 2)	(0.029 4)	(0.029 5)
	$p r j t$	-0.002 4^*	-0.002 0	-0.002 3^*	-0.001 0	-0.005 9^***	-0.006 4^***	-0.006 5^***	-0.005 7^***
	$p r j t$	(0.001 0)	(0.001 0)	(0.001 1)	(0.001 0)	(0.001 6)	(0.001 6)	(0.001 6)	(0.001 6)

Tab. 5 STERGM of short-term impact

变量		形成模型				解除（或持续）模型
变量		(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
网络结构特征	$e d g e s$	-7.972 6^***	-7.345 7^***	-5.987 1^***	-6.429 1^***	-1.816 7^***	-1.336 2^***	-0.733 6^*	-1.814 3^***
	$e d g e s$	(0.290 4)	(0.286 5)	(0.285 4)	(0.288 3)	(0.353 6)	(0.355 3)	(0.356 5)	(0.360 3)
	$m u t u a l$		1.111 8^***	1.041 1^***	0.678 1^***		0.995 0^***	0.982 1^***	0.887 6^***
	$m u t u a l$		(0.040 3)	(0.039 2)	(0.040 0)		(0.053 2)	(0.052 3)	(0.052 3)
	$g w i - d e g r e e$			-4.319 9^***	-2.408 0^***			-4.562 8^***	-3.856 6^***
	$g w i - d e g r e e$			(0.167 6)	(0.171 7)			(0.181 7)	(0.196 1)
	$g w e s p$				2.053 9^***				0.977 9^***
	$g w e s p$				(0.078 2)				(0.082 5)
接收者效应	$t a s i t$	-0.045 1	0.017 8	0.023 1	0.051 5^*	0.037 4	0.114 0^***	0.097 3^***	0.104 6^***
	$t a s i t$	(0.023 3)	(0.023 4)	(0.022 9)	(0.022 0)	(0.029 6)	(0.030 5)	(0.029 6)	(0.029 9)
	$p r i t$	-0.007 3^***	-0.006 9^***	-0.007 0^***	-0.003 8^***	0.006 0^***	0.006 1^***	0.007 8^***	0.007 3^***
	$p r i t$	(0.001 1)	(0.001 1)	(0.001 1)	(0.001 0)	(0.001 4)	(0.001 4)	(0.001 4)	(0.001 4)
发送者效应	$t a s j t$	-0.219 4^***	-0.233 4^***	-0.241 9^***	-0.205 4^***	-0.331 1^***	-0.346 4^***	-0.367 4^***	-0.356 8^***
	$t a s j t$	(0.022 8)	(0.022 9)	(0.022 9)	(0.021 8)	(0.028 7)	(0.029 2)	(0.029 4)	(0.029 5)
	$p r j t$	-0.002 4^*	-0.002 0	-0.002 3^*	-0.001 0	-0.005 9^***	-0.006 4^***	-0.006 5^***	-0.005 7^***
	$p r j t$	(0.001 0)	(0.001 0)	(0.001 1)	(0.001 0)	(0.001 6)	(0.001 6)	(0.001 6)	(0.001 6)

Tab. 6 Heterogeneity analysis

变量		长期				短期
变量		小麦	玉米	稻米	大豆	小麦	玉米	稻米	大豆
接受者效应	$t a s i t$	-0.013 4^***	-0.025 2^***	-0.002 6	-0.001 3	-0.158 9^***	0.145 3^***	0.110 0^***	0.097 3^**
	$t a s i t$	(0.003 6)	(0.002 6)	(0.002 5)	(0.004 0)	(0.023 8)	(0.020 7)	(0.019 0)	(0.029 6)
	$p r i t$	0.000 4^***	0.000 3^***	-0.000 2^***	0.000 1^*	-0.000 7	0.003 6^***	-0.004 0^***	0.007 8^***
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 1)	(0.001 4)	(0.001 0)	(0.000 9)	(0.001 5)
发送者效应	$t a s j t$	-0.030 7^***	0.013 3^***	0.024 9^***	-0.023 5^***	-0.122 8^***	-0.536 5^***	-0.466 6^***	-0.414 0^***
	$t a s j t$	(0.003 5)	(0.002 7)	(0.002 5)	(0.003 7)	(0.021 6)	(0.020 8)	(0.020 7)	(0.030 5)
	$p r j t$	-0.000 0	0.000 6^***	0.000 5^***	0.000 6^***	-0.018 3^***	-0.014 8^***	-0.005 9^***	-0.024 2^***
	$p r j t$	(0.000 1)	(0.000 0)	(0.000 0)	(0.000 1)	(0.001 7)	(0.001 2)	(0.000 9)	(0.001 8)

Tab. 6 Heterogeneity analysis

变量		长期				短期
变量		小麦	玉米	稻米	大豆	小麦	玉米	稻米	大豆
接受者效应	$t a s i t$	-0.013 4^***	-0.025 2^***	-0.002 6	-0.001 3	-0.158 9^***	0.145 3^***	0.110 0^***	0.097 3^**
	$t a s i t$	(0.003 6)	(0.002 6)	(0.002 5)	(0.004 0)	(0.023 8)	(0.020 7)	(0.019 0)	(0.029 6)
	$p r i t$	0.000 4^***	0.000 3^***	-0.000 2^***	0.000 1^*	-0.000 7	0.003 6^***	-0.004 0^***	0.007 8^***
	$p r i t$	(0.000 0)	(0.000 0)	(0.000 0)	(0.000 1)	(0.001 4)	(0.001 0)	(0.000 9)	(0.001 5)
发送者效应	$t a s j t$	-0.030 7^***	0.013 3^***	0.024 9^***	-0.023 5^***	-0.122 8^***	-0.536 5^***	-0.466 6^***	-0.414 0^***
	$t a s j t$	(0.003 5)	(0.002 7)	(0.002 5)	(0.003 7)	(0.021 6)	(0.020 8)	(0.020 7)	(0.030 5)
	$p r j t$	-0.000 0	0.000 6^***	0.000 5^***	0.000 6^***	-0.018 3^***	-0.014 8^***	-0.005 9^***	-0.024 2^***
	$p r j t$	(0.000 1)	(0.000 0)	(0.000 0)	(0.000 1)	(0.001 7)	(0.001 2)	(0.000 9)	(0.001 8)

Fig.2 The AUC value of link prediction algorithms， 2001-2022

Fig. 3 Top 10 potential trade dependency country（region） pairs， 2013-2022

Tab. 7 Potential trade dependency country(region) pairs grouped by crops

位次	小麦			玉米			稻米			大豆
位次	出口国	进口国	概率	出口国	进口国	概率	出口国	进口国	概率	出口国	进口国	概率
1	澳大利亚	西班牙	0.98	乌克兰	墨西哥	0.99	坦桑尼亚^*	南非	0.95	巴拉圭	中国	0.96
2	乌克兰^*	阿尔及利亚	0.98	美国^*	越南	0.98	巴基斯坦^*	乌克兰	0.94	美国	阿根廷	0.96
3	波兰	意大利	0.98	阿根廷^*	中国	0.98	美国	哈萨克斯坦	0.94	加拿大	阿根廷	0.96
4	波兰^*	阿尔及利亚	0.98	阿根廷	墨西哥	0.98	巴基斯坦	印度	0.94	巴拉圭	墨西哥	0.94
5	保加利亚	德国	0.97	阿根廷	西班牙	0.97	中国^*	德国	0.94	巴西	德国	0.94
6	澳大利亚^*	意大利	0.97	乌克兰	越南	0.97	中国^*	西班牙	0.94	乌克兰	阿根廷	0.94
7	保加利亚	荷兰	0.97	乌克兰	哥伦比亚	0.97	美国	保加利亚	0.94	尼日利亚	阿根廷	0.94
8	保加利亚	摩洛哥	0.97	美国^*	埃及	0.96	中国^*	南非	0.94	巴西	印尼	0.93
9	乌克兰	比利时	0.96	乌克兰^*	阿尔及利亚	0.95	印度	斯洛伐克	0.93	美国	伊朗	0.93
10	乌克兰	中国	0.96	乌克兰^*	伊朗	0.94	缅甸^*	日本	0.93	巴西	印度	0.92

Tab. 8 Other potential trade dependency with China

作物	位次	出口国	进口国	概率	标记
小麦	25	保加利亚	中国	0.93
	26	波兰	中国	0.93
	30	德国	中国	0.92
稻米	13	中国	法国	0.93	可加强
	14	中国	荷兰	0.93	可加强
	15	中国	英国	0.93	可加强
	18	中国	葡萄牙	0.93
	20	中国	新加坡	0.93	可加强
	24	中国	加拿大	0.93	可加强
	25	中国	马来西亚	0.93
	28	中国	比利时	0.92	可加强
大豆	22	乌克兰	中国	0.88
大豆	26	尼日利亚	中国	0.87

References 47

[1]	顾善松, 赵将. 俄乌冲突与全球和中国粮食安全:冲击与修复. 农业经济问题, 2024(7): 73-89.
	GU S, ZHAO J. The impact of Russia-Ukraine conflict on world and China's food security: A study of the impact and self-repair mechanism. Issues in Agricultural Economy, 2024(7): 73-89.
[2]	周波涛,钱进.IPCC AR6报告解读:极端天气气候事件变化.气候变化研究进展,2021,17(6):713-718.
	ZHOU B, QIAN J. Changes of weather and climate extremes in the IPCC AR6.Advances in Climate Change Research,2021,17(6): 713-718.
[3]	赵敏娟, 姚柳杨, 赵明恩, 等. 气候变化对中国粮食安全的影响:理论逻辑和应对措施. 农业经济问题, 2024(10): 34-43.
	ZHAO M, YAO L, ZHAO M, et al. The impact of climate change on China' s food security: Theoretical logic and response strategies issues in agricultural economy, 2024(10): 34-43.
[4]	XIE W, HUANG J, WANG J, et al. Climate change impacts on China's agriculture: The responses from market and trade. China Economic Review, 2020, 62: 101256.
[5]	刘东, 冯晓龙, 司伟. 中国粮食生产的气候变化适应水平及其机制研究. 经济学(季刊), 2024, 24(5): 1516-1532.
	LIU D, FENG X, SI W. The adaptation level and mechanism of grain production to climate change in China.China Economic Quarterly, 2024(5): 1516-1532.
[6]	陈帅, 徐晋涛, 张海鹏. 气候变化对中国粮食生产的影响——基于县级面板数据的实证分析. 中国农村经济, 2016(5): 2-15.
	CHEN S, XU J, ZHANG H. The impact of climate change on China's food production: An empirical analysis based on county-level panel data.2016(5): 2-15.
[7]	陈志钢, 胡霜. 气候变化对全球粮食安全的影响与应对策略. 农业经济问题, 2024(10): 44-56.
	CHEN Z, HU S. Impact of climate change on global food security and coping measures.Issues in Agricultural Economy,2024(10): 44-56.
[8]	黄萌田, 周佰铨, 翟盘茂. 极端天气气候事件变化对荒漠化、土地退化和粮食安全的影响. 气候变化研究进展, 2020, 16(1): 17-27.
	HUANG M, ZHOU B, ZHAI P. Impacts of extreme weather and climate events on desertification, land degradation and food security. Advances in Climate Change Research, 2020, 16(1): 17-27.
[9]	于飞, 崔惠娟, 葛全胜. "一带一路"沿线国家的自主贡献中水资源相关适应措施评估. 气候变化研究进展, 2022, 18(1): 70-80.
	YU F, CUI H, GE Q. Evaluation of water-related adaptation measures in nationally determined contributions of belt and road countries. Climate Change Research, 2022, 18(1): 70-80.
[10]	许彩艳,何爱平,安梦天.自然灾害如何影响农户人力资本投资.农业技术经济,2024(6):18-37.
	XU C, HE A, AN M.How do natural disasters affect farmers' human capital investment.Journal of Agrotechnical Economics,2024(6): 18-37.
[11]	AHMED MD R. Climate shocks' impact on agricultural income and household food security in Bangladesh: An implication of the food insecurity experience scale. Heliyon, 2024, 10(4): e25687.
[12]	覃朝晖, 范振楠, 余思明. 气温变化对粮食绿色生产效率的影响效应与传导机制. 中国农业资源与区划, 2024, 45(11): 81-94.
	QIN Z, FAN Z, YU S. The impact and transmission mechanism of temperature changes on green production efficiency of grain. Chinese Journal of Agricultural Resources and Regional Planning, 2024, 45(11): 81-94.
[13]	孔锋, 王一飞, 吕丽莉, 等. 互联互通背景下巨灾对经济影响的全球性和复杂性的进展与展望. 华中师范大学学报(自然科学版), 2018, 52(6): 871-882.
	KONG F, WANG Y, LYU L, et al. Progress and prospect of global and complex economic impacts of catastrophe under the background of interoperability. Journal of Central China Normal University(Natural Sciences), 2018, 52(6): 871-882.
[14]	KARLSSON J. Temperature and exports: Evidence from the United States. Environmental and Resource Economics, 2021, 80(2): 311-337.
[15]	OSBERGHAUS D. The effects of natural disasters and weather variations on international trade and financial flows: A review of the empirical literature. Economics of Disasters and Climate Change, 2019, 3: 305-325.
[16]	GOUEL C, LABORDE D. The crucial role of domestic and international market-mediated adaptation to climate change. Journal of Environmental Economics and Management, 2021, 106: 102408.
[17]	卞雨晨, 顾海峰. 气候风险对跨境资本流动的时变冲击——基于TVP-VAR模型的实证研究. 国际商务(对外经济贸易大学学报), 2024(1): 98-116.
	BIAN Y, GU H. Time-varying shock to cross-border capital flows from climate risk — Empirical research based on the TVP-VAR model. International Business, 2024(1): 98-116.
[18]	魏艳骄, 郁淮娣, 朱晶. 经济政策不确定性、区域经济合作与全球粮食贸易增长. 世界农业, 2024(10): 43-55.
	WEI Y, YU H, ZHU J. Economic policy uncertainty, regional economic cooperation and global food trade increase world agriculture, 2024(10): 43-55.
[19]	GUO K, LI Y, ZHANG Y, et al. How are climate risk shocks connected to agricultural markets?. Journal of Commodity Markets, 2023, 32: 100367.
[20]	OH C H, REUVENY R. Climatic natural disasters, political risk, and international trade. Global Environmental Change, 2010, 20(2): 243-254.
[21]	INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE IPCC). Climate Change 2021–The physical science basis: Working group I contribution to the sixth assessment report of the intergovernmental panel on climate change. 1 edition. Cambridge University Press, 2023[2025-04-10].
[22]	BALDOS U, HERTEL T, MOORE F. Understanding the spatial distribution of welfare impacts of global warming on agriculture and its drivers. American Journal of Agricultural Economics, 2019, 101(5): 1455-1472.
[23]	朴英姬. 气候变化下的全球粮食安全:传导机制与系统转型. 世界农业, 2023(10): 16-26.
	PIAO Y. Global food security under climate change: Transmission mechanism and system transformation.World Agriculture, 2023(10): 16-26.
[24]	朴英姬. 气候变化对非洲粮食安全的多维冲击及其治理. 西亚非洲, 2024(5): 66-87, 153-154.
	PIAO Y. Multidimensional impacts of climate change on food security in africa and its governance.West Asia and Africa, 2024(5): 66-87.
[25]	余淼杰,林雨晨.气候变化与国际贸易研究综述.长安大学学报(社会科学版),2019, 21(1):9-15.
	YU M, LIN Y.Literature review of international trade and climate change.Journal of Chang'an University(Social Science Edition),2019,21(1): 9-15.
[26]	BOZZOLA M, LAMONACA E, SANTERAMO F. Impacts of climate change on global agri-food trade. Ecological Indicators, 2023, 154: 110680.
[27]	GÖTZ L, GLAUBEN T, BRÜMMER B. Wheat export restrictions and domestic market effects in Russia and Ukraine during the food crisis. Food Policy, 2013, 38: 214-226.
[28]	刘立涛, 刘晓洁, 伦飞, 等. 全球气候变化下的中国粮食安全问题研究. 自然资源学报, 2018, 33(6): 927-939.
	LIU L, LIU X, LUN F, et al. Research on China's food security under global climate change background. Journal of Natural Resources, 2018, 33(6): 927-939.
[29]	彭俊杰. 气候变化对全球粮食产量的影响综述. 世界农业, 2017(5): 19-24, .
	PENG J. An overview of the impact of climate change on global food production. World Agriculture, 2017(5): 19-24.
[30]	DING C, XIA Y, SU Y,et al.Study on the impact of climate change on China's import trade of major agricultural products and adaptation strategies. International Journal of Environmental Research and Public Health, 2022, 19(21): 14374.
[31]	LESK C, ROWHANI P, RAMANKUTTY N. Influence of extreme weather disasters on global crop production. Nature, 2016, 529(7584): 84-87.
[32]	吴高艺. 国际粮食价格波动趋势及典型特征研究:基于1990年以来四轮波动周期表征分析. 价格理论与实践, 2024(3): 125-131.
	WU G. Research on the fluctuation trends and typical characteristics of international grain prices —Based on the characterization analysis of four round wave cycles since 1990. Price:Theory & Practice, 2024(3): 125-131.
[33]	魏艳骄, 张慧艳, 朱晶. 新发展格局下中国大豆进口依赖性风险及市场布局优化分析. 中国农村经济, 2021(12): 66-86.
	WEI Y, ZHANG H, ZHU J. An analysis of dependence risk and market layout optimization for soybean import of China under the new development pattern. Chinese Rural Economy, 2021(12): 66-86.
[34]	刘凯, 王欢, 穆月英. 中国大豆进口风险分散及进口来源结构优化——基于替代性与依赖性视角. 中国油脂, 2025, 50(2): 1-7, 22.
	LIU K, WANG H, MU Y. Risk dispersion and optimization of soybean imports for China: Based on substitution and dependence risk. CHINA OILS AND FATS, 2025, 50(2): 1-7, 22.
[35]	谭用, 周洺竹, 綦建红. 不确定性与中国粮食分散进口:结构估计与反事实研究. 经济学(季刊), 2024, 24(2): 570-587.
	TAN Y, ZHOU M, QI J. Uncertainty and China's decentralized food import: Structure estimation and counter factual research.China Economic Quarterly, 2024, 24(2): 570-587.
[36]	COSTINOT A, KOMUNJER I. What goods do countries trade? New ricardian predictions.National Bureau of Economic Research, 2007.
[37]	GOUEL C, LABORDE D. The crucial role of domestic and international market-mediated adaptation to climate change. Journal of Environmental Economics and Management, 2021, 106: 102408.
[38]	付清.经济相互依赖武器化与美国主导权护持战略.世界经济与政治,2024(7):125-152.
	FU Q.Weaponized economic interdependence and the maintenance strategy of US dominance.World Economics and Politics,2024(7): 125-152.
[39]	D'ODORICO P, CARR J, LAIO F, et al. Feeding humanity through global food trade.Earth's Future,2014,2(9): 458-469.
[40]	刘东, 陈景帅, 冯晓龙, 等. 气候变化对农户农地流转行为的影响——来自全国农村固定观察点的证据. 中国农村经济, 2024(5): 40-61.
	LIU D, CHEN J, FENG X, et al. The impact of climate change on farmland transfer behavior: Evidence from the national rural fixed observation points survey.Chinese Rural Economy, 2024(5): 40-61.
[41]	HUANG K, ZHAO H, HUANG J, et al. The impact of climate change on the labor allocation: Empirical evidence from China. Journal of Environmental Economics and Management, 2020, 104: 102376.
[42]	CUI X. Climate change and adaptation in agriculture: Evidence from US cropping patterns. Journal of Environmental Economics and Management, 2020, 101: 102306.
[43]	KRIVITSKY P, HANDCOCK M. A separable model for dynamic networks. Journal of the Royal Statistical Society Series B: Statistical Methodology, 2014, 76(1): 29-46.
[44]	GUO K, JI Q, ZHANG D. A dataset to measure global climate physical risk. Data in Brief, 2024, 54: 110502.
[45]	郝晓燕, 李雪. 基于"口粮绝对安全"的小麦和稻谷多元化进口策略分析. 华南农业大学学报(社会科学版), 2022, 21(4): 67-78.
	HAO X, LI X. An analysis on the diversified import strategy of wheat and rice based on "Absolute Security of Staple Food". Journal of South China Agricultural University(Social Science Edition, 2022,21(4): 67-78.
[46]	刘靖文,侯丽薇,杨艳涛. 中国玉米供需平衡及国际市场可利用性分析. 中国农业资源与区划, 2021, 42(4): 126-133.
	LIU J, HOU L, YANG Y. Analysis on the balance between supply and demand of corn market in China and the availability of international market. Chinese Journal of Agricultural Resources and Regional Planning, 2021, 42(4): 126-133.
[47]	苏丹华, 倪国华, 鲍勤. "一带一路"沿线国家大豆生产潜力及其对中国大豆贸易主导权的影响研究. 管理评论, 2023: 1-10.
	SU D, NI G, BAO Q. A study on the soybean production potential of countries along the Belt and Road and its impact on China's soybean trade dominance. Management Review, 2023: 1-10.