WO2023245399A1 - 基于土地系统和气候变化耦合的水稻生产潜力模拟方法 - Google Patents

基于土地系统和气候变化耦合的水稻生产潜力模拟方法 Download PDF

Info

Publication number
WO2023245399A1
WO2023245399A1 PCT/CN2022/100006 CN2022100006W WO2023245399A1 WO 2023245399 A1 WO2023245399 A1 WO 2023245399A1 CN 2022100006 W CN2022100006 W CN 2022100006W WO 2023245399 A1 WO2023245399 A1 WO 2023245399A1
Authority
WO
WIPO (PCT)
Prior art keywords
land
production potential
model
rice production
climate
Prior art date
Application number
PCT/CN2022/100006
Other languages
English (en)
French (fr)
Inventor
李满春
姜朋辉
杜聪
Original Assignee
南京大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京大学 filed Critical 南京大学
Priority to PCT/CN2022/100006 priority Critical patent/WO2023245399A1/zh
Priority to LU503105A priority patent/LU503105B1/en
Publication of WO2023245399A1 publication Critical patent/WO2023245399A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/20Design optimisation, verification or simulation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02Agriculture; Fishing; Forestry; Mining
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00Details relating to CAD techniques
    • G06F2111/10Numerical modelling
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/22Yield analysis or yield optimisation

Definitions

  • the invention relates to the field of geographic information technology, and in particular to a rice production potential simulation method based on the coupling of land systems and climate change.
  • Food is an important strategic material related to the national economy and people's death.
  • Food security is an important foundation for achieving economic development, social stability, and national security. Affected by global population expansion and declining food production capacity, ensuring food security is an eternal issue. Land and climate are the two basic factors that affect food production.
  • Empirical method This method considers factors related to grain production and is obtained by establishing functional equations to fit the trend of grain production series.
  • the main models used include Miami model, wageningen model, etc. This type of model can only be used for Grain yields are roughly estimated in study areas with relatively uniform climatic conditions, especially in areas with huge differences in climatic conditions between different regions.
  • Crop growth model method This method is based on crop growth dynamics and considers crop photosynthesis, physiological and ecological characteristics, and the external environment to simulate the flow of water, carbon, and nitrogen in the farmland production system, crop growth, and crop yield.
  • Commonly used models include DSSAT, ORYZA, SWAP, EPIC, WOFOST, AquaCrop, APSIM, etc. This type of model requires many parameters and has high computational cost. It can simulate grain yield in small-scale research areas with higher accuracy;
  • the third category machine/deep learning model. This type of method is more expensive to build models and requires a large number of reliable training samples to train a more stable and accurate model;
  • Category 4 climate production potential model method, which is recognized at home and abroad as the most basic method for simulating food production potential.
  • Representative models include the step-by-step correction model, the agricultural ecological zone (AEZ) model, the GAEZ model, etc.
  • AEZ agricultural ecological zone
  • GAEZ GAEZ model
  • This type of method Comprehensive consideration of changes in climate factors and land elements, grain production is obtained by gradually calculating photosynthetic production potential, light and temperature production potential, climate production potential, and land production potential, which can more scientifically reflect the grain production potential under the influence of natural and human factors. As a result, the data required for the model are easy to obtain, and the simulation results at the macro scale are more consistent with actual production conditions.
  • the present invention proposes a rice production potential simulation method based on the coupling of land systems and climate change to overcome the above technical problems existing in existing related technologies.
  • the method includes the following steps:
  • the construction of the GM-FLUS model to simulate land system changes includes the following steps:
  • collecting original data and predicting the number of future land use types based on system change patterns includes the following steps:
  • S112. Process the number of each land use type, find the system change rules, generate a regular data sequence, and establish a corresponding differential equation model to predict the number of future land use types.
  • constructing a gray prediction model and using the number of land use changes predicted by it as the input of the FLUS model includes the following steps:
  • X (0) ⁇ x (0) (1),(x) (0) (2),...,x (0) (n) ⁇
  • X (1) ⁇ x (1) (1),(x) (1) (2),...,x (1) (n) ⁇
  • n the number of original sequences
  • a represents the development gray level
  • b represents the endogenous control gray level
  • a and b satisfy:
  • the accuracy test and evaluation of the established gray prediction model includes the following steps:
  • q (0) represents the residual sequence
  • q represents the mean of the residual sequence
  • S 1 represents the standard deviation of the original sequence
  • S 2 represents the standard deviation of the residual sequence
  • a represents the development gray scale
  • b represents the internal generation control grayscale
  • t represents time.
  • the adaptive inertial competition mechanism based on roulette selection is used to simulate future land changes and deal with the uncertainty and complexity when multiple land use types transform into each other.
  • sp(p,k,t) is the suitability probability of the k-th land type on grid p and time t;
  • ⁇ j,k is the weight between the hidden layer and the output layer
  • sigmoid() is the activation function from the hidden layer to the output layer
  • net j (p,t) is the signal received by the j-th hidden layer grid p at time t.
  • the adaptive inertia competition mechanism based on roulette selection simulates future land changes and handles the uncertainty and complexity when multiple land use types transform into each other, including the following steps:
  • sc c ⁇ k represents the cost of converting land type c into k
  • ⁇ k represents the weight of the domain influence degree of each land use type.
  • the use of the GAEZ model to estimate rice production potential includes the following steps:
  • step-by-step restriction method is used to calculate rice production potential.
  • the production potential includes photosynthetic production potential, light and temperature production potential, light and temperature water production potential, climate production potential and rice production potential in order.
  • the beneficial effects of the present invention are: simulating land system changes by constructing a GM-FLUS model, combined with SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 in the shared socio-economic road under the CAS-ESM2 climate scenario model comparison plan
  • Four future climate scenarios use the GAEZ model to comprehensively consider climate, soil, terrain, land system and other factors to estimate rice production potential, which can provide technical support for coping with future climate change, rational use of cultivated land resources, and ensuring food security.
  • Figure 1 is a flow chart of a rice production potential simulation method based on the coupling of land systems and climate change according to an embodiment of the present invention
  • Figure 2 is an overall technical roadmap of a rice production potential simulation method based on land system and climate change coupling according to an embodiment of the present invention
  • Figure 3 is a structural diagram of the GAEZ model in the rice production potential simulation method based on the coupling of land system and climate change according to an embodiment of the present invention
  • Figure 4 is a spatial distribution diagram of rice production potential in 2020 in the rice production potential simulation method based on coupling of land system and climate change according to an embodiment of the present invention
  • Figure 5 is a schematic diagram of the correlation verification of the rice production potential simulation method based on the coupling of land system and climate change according to an embodiment of the present invention.
  • a rice production potential simulation method based on coupling of land systems and climate change is provided.
  • a rice production potential simulation method based on the coupling of land systems and climate change is provided.
  • the method includes the following steps :
  • S112. Process the number of each land use type, find the system change rules, generate a regular data sequence, and establish a corresponding differential equation model to predict the number of future land use types.
  • This invention uses a total of seven periods of land use remote sensing monitoring data in 1990, 1995, 2000, 2005, 2010, 2015, and 2020 as original data to process the quantity of each land use type and find the system change rules. , generate a data sequence with strong regularity, and then establish a corresponding differential equation model to predict the number of land use types in 2030, 2040, 2050, and 2060.
  • X (0) ⁇ x (0) (1),(x) (0) (2),...,x (0) (n) ⁇
  • X (1) ⁇ x (1) (1),(x) (1) (2),...,x (1) (n) ⁇
  • n the number of original sequences
  • a represents the development gray level
  • b represents the endogenous control gray level
  • a and b satisfy:
  • q (0) represents the residual sequence
  • q represents the mean of the residual sequence
  • S 1 represents the standard deviation of the original sequence
  • S 2 represents the standard deviation of the residual sequence
  • a represents the development gray scale
  • b represents the internal generation control grayscale
  • t represents time.
  • sp(p,k,t) is the suitability probability of the k-th land type on grid p and time t;
  • ⁇ j,k is the weight between the hidden layer and the output layer
  • sigmoid() is the activation function from the hidden layer to the output layer
  • net j (p,t) is the signal received by the j-th hidden layer grid p at time t.
  • the adaptive inertial competition mechanism based on roulette selection is used to simulate future land changes and deal with the uncertainty and complexity when multiple land use types transform into each other, including the following steps:
  • sc c ⁇ k represents the cost of converting land type c into k
  • ⁇ k represents the weight of the domain influence degree of each land use type.
  • step-by-step restriction method is used to calculate rice production potential.
  • the production potential includes photosynthetic production potential, light and temperature production potential, light and temperature water production potential, climate production potential and rice production potential in order.
  • This invention uses the GAEZ model to estimate rice production potential. First, it evaluates the spatial distribution of rice based on temperature (daily average temperature, daily maximum temperature, daily minimum temperature, accumulated temperature) and precipitation (precipitation amount, relative humidity, precipitation intensity, precipitation variability) conditions. According to the climate suitability of the rice planting area, the step-by-step restriction method is used to calculate the rice production potential, that is, according to photosynthetic production potential (only light limitation) - light and temperature production potential (light and temperature limitation) - light, temperature and water production potential (light, temperature and water restrictions) - climate production potential (agricultural climate disaster restrictions) - rice production potential (soil and various management measures restrictions) are carried out step by step, as shown in Figure 3.
  • the GAEZ model can simulate two scenarios: rainfed and irrigated. Under rainfed conditions, only the yield-reducing effect of precipitation on rice production potential is considered; while under irrigated conditions, it is assumed that water is sufficient and the impact of water on rice production potential is not considered.
  • This invention assumes that irrigation technology will reach a higher level in the future, and therefore directly uses the grain production potential under irrigation scenarios.
  • the calculation process combines cultivated land distribution data and potential ripening data. Rice is only grown on paddy fields, and multiple cropping systems are considered to obtain the maximum rice production potential.
  • This invention estimates the rice production potential of paddy fields in 2020. Taking each province (municipality) across the country as the basic statistical unit, the total rice production potential of each province and municipality in 2020 is calculated, and compared with the 2020 national 31 provinces (municipalities) (excluding Hong Kong Special Administrative Region, Taiwan province, and Macao Special Administrative Region) released by the National Bureau of Statistics. (External) rice yield data for correlation verification.
  • the red dotted line in the figure represents the correlation trend line between the estimated simulated potential rice production in 2020 and the actual rice production in 2020 released by the National Bureau of Statistics.
  • the correlation coefficient between the two is 0.82, indicating that there is a strong correlation between the two, that is Changes in potential rice yield can largely reflect the changing trend of actual rice yield.
  • the correlation coefficient between the total rice production potential estimated by this invention and the actual production in 2020 is 0.82, and the total rice production potential is 1.27 times the national actual rice production statistical value. This is consistent with Liu Luo (2014) et al.
  • the correlation verification results between the estimated total grain production potential in 2010 and the actual grain production are basically consistent.
  • the total rice potential estimated by the present invention is closer to the actual statistical value.
  • the land system changes are simulated by constructing the GM-FLUS model, combined with the SSP1-2.6, SSP2-4.5, and SSP2-4.5 of the shared socio-economic road under the CAS-ESM2 climate scenario model comparison plan.
  • SSP3-7.0 and SSP5-8.5 use the GAEZ model to comprehensively consider climate, soil, terrain, land system and other factors to estimate rice production potential, which can provide solutions for future climate change, rational utilization of cultivated land resources, and security Food security provides technical support.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mining & Mineral Resources (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Agronomy & Crop Science (AREA)
  • Animal Husbandry (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种基于土地系统和气候变化耦合的水稻生产潜力模拟方法,包括以下步骤:构建GM-FLUS模型模拟土地系统变化(S1),采用GAEZ模型估算水稻生产潜力(S2),分析不同情景下水稻生产潜力的空间分布特征和时间变化趋势(S3)。通过构建GM-FLUS模型模拟土地系统变化,结合CAS-ESM2气候情景模式比较计划下的共享社会经济路中的SSP1-2.6、SSP2-4.5、SSP3-7.0和SSP5-8.5四个未来气候情景,采用GAEZ模型综合考虑气候、土壤、地形、土地系统等多方面因素,估算了水稻生产潜力,为应对未来气候变化、合理利用耕地资源、保障粮食安全提供了技术支撑。

Description

基于土地系统和气候变化耦合的水稻生产潜力模拟方法 技术领域
本发明涉及地理信息技术领域,尤其涉及基于土地系统和气候变化耦合的水稻生产潜力模拟方法。
背景技术
粮食是关系国计民生的重要战略物资,粮食安全是实现经济发展、社会稳定、国家安全的重要基础。受全球性人口膨胀和粮食生产能力下降的影响,保障粮食安全是一个永恒的课题。而土地和气候正是影响粮食产量的两大基本因素。
据政府间气候变化专门委员会(IPCC)发布的第六次评估报告(AR6)指出,自19世纪以来,全球气温已经平均上升了1.1℃,如果全球温度继续上升,会引发干旱、暴雨、洪水、寒潮等极端天气频繁发生,改变粮食作物生长发育条件,进而导致产量的减产。
同时,受到工业化、城市化等进程影响,以及退耕还林还草等政策驱动,发展中国家耕地的数量有所减少、质量有所下降以及空间分布发生了变化,进而对粮食产量造成深远的影响,多次引发全球性的粮食危机。因此,科学估测土地系统和气候变化对于粮食产量和安全问题的影响,探索在未来土地利用情况、不同气候条件下的粮食产量,对政策制定和国民经济可持续发展具有重要意义。
对粮食生产潜力的计算,国内外学者采用不同的方法开展了大量的研究。常用的方法可以概括为四类:
第一类:经验法,这类方法考虑反映与粮食产量相关的因子,通过建立函数方程拟合粮食产量序列趋势得到,主要用到的模型有Miami模型、wageningen模型等,这类模型只能对气候条件较为均一的研究区粮食产量进行粗略地估算,尤其在不同区域气候条件差异巨大的区域效果较差;
第二类:作物生长模型法,这类方法以作物生长动力学为基础,考虑了作物光合作用、生理生态特性以及外部环境模拟农田生产系统水、碳、氮的流动、作物生长以及作物产量,常用的模型有DSSAT、ORYZA、SWAP、EPIC、WOFOST、AquaCrop、APSIM等,这类模型所需要的参数较多,计算成本较高,对小尺度的研究区进行粮食产量模拟的精度较高;
第三类:机器/深度学习模型,该类方法构建模型的成本较高,需要大量可靠的训练 样本,才能训练出较为稳定精确的模型;
第四类:气候生产潜力模型法,是国内外公认为是模拟粮食生产潜力最基本的方法,具有代表性的模型有逐级订正模型、农业生态区位(AEZ)模型、GAEZ模型等,这类方法综合考虑了气候因素和土地要素的变化,通过逐步计算光合生产潜力、光温生产潜力、气候生产潜力、土地生产潜力得到粮食产量,能较为科学地反映在自然和人文要素影响下粮食生产潜力的结果,且模型所需的数据易获取,在宏观尺度的模拟结果更符合实际生产情况。
然而,上述模型均基于现状或历史数据开展耕地产量模拟,并未考虑未来自然与人文发展不确定性对耕地产量的影响,因此对于远期耕地保护或者农业生产布局等空间决策的指导价值较低。
发明内容
针对相关技术中的问题,本发明提出基于土地系统和气候变化耦合的水稻生产潜力模拟方法,以克服现有相关技术所存在的上述技术问题。
为此,本发明采用的具体技术方案如下:该方法包括以下步骤:
S1、构建GM-FLUS模型模拟土地系统变化;
S2、采用GAEZ模型估算水稻生产潜力;
S3、分析不同情景下水稻生产潜力的空间分布特征和时间变化趋势。
进一步的,所述构建GM-FLUS模型模拟土地系统变化,包括以下步骤:
S11、采集原始数据,并根据系统变动规律预测未来土地利用类型的数量;
S12、构建灰色预测模型,并将其预测出的土地利用变化的数量作为FLUS模型的输入;
S13、对建立的灰色预测模型进行精度检验与评估;
S14、利用所述FLUS模型未来土地系统变化。
进一步的,所述采集原始数据,并根据系统变动规律预测未来土地利用类型的数量,包括以下步骤:
S111、选取多组历史土地利用遥感监测数据作为原始数据;
S112、对每种土地利用类型数量进行处理,寻找系统变动规律,生成有规律性的数据序列,并建立相应的微分方程模型,预测未来土地利用类型的数量。
进一步的,所述构建灰色预测模型,并将其预测出的土地利用变化的数量作为FLUS 模型的输入,包括以下步骤:
S121、设每一类土地利用数据原始非负序列为X (0),其表达式为:
X (0)={x (0)(1),(x) (0)(2),…,x (0)(n)}
S122、对X (0)构造一次累加序列为X (1),其表达式为:
X (1)={x (1)(1),(x) (1)(2),…,x (1)(n)}
Figure PCTCN2022100006-appb-000001
式中,i=1,…,k,k=1,2,…,n,n表示原始序列的个数;
S123、构建一阶灰色预测模型,其白化方程为:
Figure PCTCN2022100006-appb-000002
式中,a表示发展灰度,b表示内生成控制灰度,且a、b满足:
Figure PCTCN2022100006-appb-000003
z (1)(k)=0.5x (1)(k)+0.5x (1)(k-1)
式中,z (1)表示X (1)紧邻均值生成序列,k=2,3,…,n,T表示转置运算符号。
进一步的,其特征在于,所述对建立的灰色预测模型进行精度检验与评估,包括以下步骤:
S131、依据后验差比值c与小误差概率p两个指标对灰色预测模型进行检验,其表达式分别为:
Figure PCTCN2022100006-appb-000004
p=P{|q (0)(k)-q|<0.6745s 1}
式中,q (0)表示残差序列;
q表示残差序列的均值;
S 1表示原始序列的标准差;
S 2表示残差序列的标准差;
S132、若所述灰色预测模型的精度不符合要求,则使用残差序列建立灰色预测模型对原模型进行修正,提高精度;
S133、若所述灰色预测模型的精度符合要求,则还原数据与预测值,其运算表达式为:
Figure PCTCN2022100006-appb-000005
Figure PCTCN2022100006-appb-000006
式中,a表示发展灰度;
b表示内生成控制灰度;
x (0)表示原始非负序列;
Figure PCTCN2022100006-appb-000007
表示原始非负序列的预测值;
Figure PCTCN2022100006-appb-000008
表示一次累加序列的预测值;
e表示自然常数;
t表示时间。
进一步的,利用所述FLUS模型未来土地系统变化,包括以下步骤:
S141、采用人工神经网络从原始土地利用数据与驱动因子获取各类用地类型的适宜性概率;
S142、基于轮盘赌选择的自适应惯性竞争机制进行未来土地变化的模拟,处理多种土地利用类型相互转化时的不确定性与复杂性。
进一步的,所述采用人工神经网络从原始土地利用数据与驱动因子获取各类用地类型的适宜性概率的运算表达式为:
Figure PCTCN2022100006-appb-000009
式中,sp(p,k,t)为第k种地类在栅格p和时间t上的适宜性概率;
ω j,k为隐藏层与输出层间的权重;
sigmoid()为隐藏层到输出层的激励函数;
net j(p,t)为第j个隐藏层栅格p在时间t上所接收到的信号。
进一步的,所述基于轮盘赌选择的自适应惯性竞争机制进行未来土地变化的模拟,处理多种土地利用类型相互转化时的不确定性与复杂性,包括以下步骤:
S1421、计算第k种地类在t时刻的自适应惯性系数
Figure PCTCN2022100006-appb-000010
其表达式为:
Figure PCTCN2022100006-appb-000011
式中,
Figure PCTCN2022100006-appb-000012
分别为t-1、t-2时刻第k种类型用地的需求数量与栅格数量的差值;
S1422、通过元胞自动机模型的迭代功能,将栅格的概率分配到各用地类型,计算t时刻栅格p转化为k类型用地的概率
Figure PCTCN2022100006-appb-000013
其表达式为:
Figure PCTCN2022100006-appb-000014
Figure PCTCN2022100006-appb-000015
式中,sc c→k表示用地类型c转化为k的成本,
1-sc c→k表示转化的难易程度;
Figure PCTCN2022100006-appb-000016
表示领域影响程度;
Figure PCTCN2022100006-appb-000017
表示在N阶Moore领域窗口内第k种用地类型的栅格总数;
ω k表示各用地类型的领域影响程度的权重。
进一步的,所述采用GAEZ模型估算水稻生产潜力,包括以下步骤:
S21、根据温度和降水条件来评价水稻空间分布的气候适宜性;
S22、针对水稻适宜种植区域,采用逐级限制法来计算水稻生产潜力。
进一步的,所述生产潜力依次包括光合生产潜力、光温生产潜力、光温水生产潜力、气候生产潜力及水稻生产潜力。
本发明的有益效果为:通过构建GM-FLUS模型模拟土地系统变化,结合CAS-ESM2气候情景模式比较计划下的共享社会经济路中的SSP1-2.6、SSP2-4.5、SSP3-7.0和SSP5-8.5四个未来气候情景,采用GAEZ模型综合考虑气候、土壤、地形、土地系统等多方面因素,估算了水稻生产潜力,可以为应对未来气候变化、合理利用耕地资源、保障粮食安全提供了技术支撑。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是根据本发明实施例的基于土地系统和气候变化耦合的水稻生产潜力模拟方法的流程图;
图2是根据本发明实施例的基于土地系统和气候变化耦合的水稻生产潜力模拟方法的总体技术路线图;
图3是根据本发明实施例的基于土地系统和气候变化耦合的水稻生产潜力模拟方法中GAEZ模型结构图;
图4是根据本发明实施例的基于土地系统和气候变化耦合的水稻生产潜力模拟方法中2020年水稻生产潜力空间分布图;
图5是根据本发明实施例的基于土地系统和气候变化耦合的水稻生产潜力模拟方法的相关性验证示意图。
具体实施方式
为进一步说明各实施例,本发明提供有附图,这些附图为本发明揭露内容的一部分,其主要用以说明实施例,并可配合说明书的相关描述来解释实施例的运作原理,配合参考这些内容,本领域普通技术人员应能理解其他可能的实施方式以及本发明的优点,图中的组件并未按比例绘制,而类似的组件符号通常用来表示类似的组件。
根据本发明的实施例,提供了基于土地系统和气候变化耦合的水稻生产潜力模拟方法。
现结合附图和具体实施方式对本发明进一步说明,如图1-7所示,根据本发明的一个实施例,提供了基于土地系统和气候变化耦合的水稻生产潜力模拟方法,该方法包括以下步骤:
S1、构建GM-FLUS模型模拟土地系统变化,包括以下步骤:
S11、采集原始数据,并根据系统变动规律预测未来土地利用类型的数量,包括以下步骤:
S111、选取多组历史土地利用遥感监测数据作为原始数据;
S112、对每种土地利用类型数量进行处理,寻找系统变动规律,生成有规律性的数据序列,并建立相应的微分方程模型,预测未来土地利用类型的数量。
本发明使用1990年、1995年、2000年、2005年、2010年、2015年、2020年共七期土地利用遥感监测数据作为原始数据,对每一种土地利用类型数量进行处理,寻找系统变动规律,生成有较强规律性的数据序列,然后建立相应的微分方程模型,从而预测2030年、2040年、2050年、2060年的土地利用类型的数量。
S12、构建灰色预测模型,并将其预测出的土地利用变化的数量作为FLUS模型的输入,包括以下步骤:
S121、设每一类土地利用数据原始非负序列为X (0),其表达式为:
X (0)={x (0)(1),(x) (0)(2),…,x (0)(n)}
S122、对X (0)构造一次累加序列为X (1),其表达式为:
X (1)={x (1)(1),(x) (1)(2),…,x (1)(n)}
Figure PCTCN2022100006-appb-000018
式中,i=1,…,k,k=1,2,…,n,n表示原始序列的个数;
S123、构建一阶灰色预测模型,其白化方程为:
Figure PCTCN2022100006-appb-000019
式中,a表示发展灰度,b表示内生成控制灰度,且a、b满足:
Figure PCTCN2022100006-appb-000020
z (1)(k)=0.5x (1)(k)+0.5x (1)(k-1)
式中,z (1)表示X (1)紧邻均值生成序列,k=2,3,…,n,T表示转置运算符号。
S13、对建立的灰色预测模型进行精度检验与评估,包括以下步骤:
S131、依据后验差比值c与小误差概率p两个指标对灰色预测模型进行检验,其表达式分别为:
Figure PCTCN2022100006-appb-000021
p=P{|q (0)(k)-q|<0.6745s 1}
式中,q (0)表示残差序列;
q表示残差序列的均值;
S 1表示原始序列的标准差;
S 2表示残差序列的标准差;
S132、若所述灰色预测模型的精度不符合要求,则使用残差序列建立灰色预测模型对原模型进行修正,提高精度;
S133、若所述灰色预测模型的精度符合要求,则还原数据与预测值,其运算表达式为:
Figure PCTCN2022100006-appb-000022
Figure PCTCN2022100006-appb-000023
式中,a表示发展灰度;
b表示内生成控制灰度;
x (0)表示原始非负序列;
Figure PCTCN2022100006-appb-000024
表示原始非负序列的预测值;
Figure PCTCN2022100006-appb-000025
表示一次累加序列的预测值;
e表示自然常数;
t表示时间。
S14、利用所述FLUS模型未来土地系统变化,包括以下步骤:
S141、采用人工神经网络从原始土地利用数据与驱动因子获取各类用地类型的适宜性概率,其运算表达式为:
Figure PCTCN2022100006-appb-000026
式中,sp(p,k,t)为第k种地类在栅格p和时间t上的适宜性概率;
ω j,k为隐藏层与输出层间的权重;
sigmoid()为隐藏层到输出层的激励函数;
net j(p,t)为第j个隐藏层栅格p在时间t上所接收到的信号。
S142、基于轮盘赌选择的自适应惯性竞争机制进行未来土地变化的模拟,处理多种土地利用类型相互转化时的不确定性与复杂性,包括以下步骤:
S1421、计算第k种地类在t时刻的自适应惯性系数
Figure PCTCN2022100006-appb-000027
其表达式为:
Figure PCTCN2022100006-appb-000028
式中,
Figure PCTCN2022100006-appb-000029
分别为t-1、t-2时刻第k种类型用地的需求数量与栅格数量的差值;
S1422、通过元胞自动机(CA)模型的迭代功能,将栅格的概率分配到各用地类型, 计算t时刻栅格p转化为k类型用地的概率
Figure PCTCN2022100006-appb-000030
其表达式为:
Figure PCTCN2022100006-appb-000031
Figure PCTCN2022100006-appb-000032
式中,sc c→k表示用地类型c转化为k的成本,
1-sc c→k表示转化的难易程度;
Figure PCTCN2022100006-appb-000033
表示领域影响程度;
Figure PCTCN2022100006-appb-000034
表示在N阶Moore领域窗口内第k种用地类型的栅格总数,本发明选择N=3;ω_k为各用地类型的邻域影响程度的权重;
ω k表示各用地类型的领域影响程度的权重。
S2、采用GAEZ模型估算水稻生产潜力,包括以下步骤:
S21、根据温度和降水条件来评价水稻空间分布的气候适宜性;
S22、针对水稻适宜种植区域,采用逐级限制法来计算水稻生产潜力。
其中,所述生产潜力依次包括光合生产潜力、光温生产潜力、光温水生产潜力、气候生产潜力及水稻生产潜力。
本发明使用GAEZ模型估算水稻生产潜力,首先根据温度(日均温、日最高温度、日最低温度、积温)和降水(降水量、相对湿度、降水强度、降水变率)条件来评价水稻空间分布的气候适宜性,然后针对水稻适宜种植区域,采用逐级限制法来计算水稻生产潜力,即:按光合生产潜力(仅光照限制)—光温生产潜力(光照和温度限制)—光温水生产潜力(光照、温度和水分限制)—气候生产潜力(农业气候灾害限制)—水稻生产潜力(土壤及各种管理措施限制)逐级进行,如图3所示。
GAEZ模型可以模拟雨养和灌溉两种情景,在雨养条件下仅考虑降水对水稻生产潜力的减产效应;而在灌溉条件下,假设水分是充足的,不考虑水分对水稻生产潜力的影响。本发明假设未来灌溉技术达到较高水平,因而直接采用灌溉情景下的粮食生产潜力。计算过程中结合了耕地分布数据和潜在熟制数据,水稻仅在水田上种植,考虑了多熟制来获取最大的水稻生产潜力。
S3、分析不同情景下水稻生产潜力的空间分布特征和时间变化趋势。
本发明估算了2020年水田的水稻生产潜力。以全国各省(市)为基本统计单元,统计各省市2020年水稻生产潜力总产量,并与国家统计局发布的2020年全国31个省(市)(除香港特别行政区、台湾省、澳门特别行政区外)水稻产量数据进行相关性验证。图中红色虚线表示估算出的2020年模拟的潜在水稻产量与国家统计局发布的2020年实际水稻产量的相关趋势线,二者相关性系数为0.82,说明二者存在较强的相关性,即潜在水稻产量的变化在很大程度上能反映实际水稻产量的变化趋势。
从估算结果来看,本发明估算的2020年水稻生产潜力总量与实际产量相关性系数为0.82,水稻生产潜力总量是全国实际水稻产量统计值的1.27倍,与刘洛(2014)等人估算的2010年粮食生产潜力总量与实际粮食产量的相关性验证结果基本一致,本发明估算的水稻潜力总量与实际统计值更为接近。
综上所述,借助于本发明的上述技术方案,通过构建GM-FLUS模型模拟土地系统变化,结合CAS-ESM2气候情景模式比较计划下的共享社会经济路中的SSP1-2.6、SSP2-4.5、SSP3-7.0和SSP5-8.5四个未来气候情景,采用GAEZ模型综合考虑气候、土壤、地形、土地系统等多方面因素,估算了水稻生产潜力,可以为应对未来气候变化、合理利用耕地资源、保障粮食安全提供了技术支撑。
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本发明构思的前提下,还可以做出若干变形和改进,这些都属于本发明的保护范围。因此,本发明专利的保护范围应以所附权利要求为准。

Claims (10)

  1. 基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,该方法包括以下步骤:
    S1、构建GM-FLUS模型模拟土地系统变化;
    S2、采用GAEZ模型估算水稻生产潜力;
    S3、分析不同情景下水稻生产潜力的空间分布特征和时间变化趋势。
  2. 根据权利要求1所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述构建GM-FLUS模型模拟土地系统变化,包括以下步骤:
    S11、采集原始数据,并根据系统变动规律预测未来土地利用类型的数量;
    S12、构建灰色预测模型,并将其预测出的土地利用变化的数量作为FLUS模型的输入;
    S13、对建立的灰色预测模型进行精度检验与评估;
    S14、利用所述FLUS模型未来土地系统变化。
  3. 根据权利要求2所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述采集原始数据,并根据系统变动规律预测未来土地利用类型的数量,包括以下步骤:
    S111、选取多组历史土地利用遥感监测数据作为原始数据;
    S112、对每种土地利用类型数量进行处理,寻找系统变动规律,生成有规律性的数据序列,并建立相应的微分方程模型,预测未来土地利用类型的数量。
  4. 根据权利要求3所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述构建灰色预测模型,并将其预测出的土地利用变化的数量作为FLUS模型的输入,包括以下步骤:
    S121、设每一类土地利用数据原始非负序列为X (0),其表达式为:
    X (0)={x (0)(1),(x) (0)(2),…,x (0)(n)}
    S122、对X (0)构造一次累加序列为X (1),其表达式为:
    X (1)={x (1)(1),(x) (1)(2),…,x (1)(n)}
    Figure PCTCN2022100006-appb-100001
    式中,i=1,…,k,k=1,2,…,n,n表示原始序列的个数;
    S123、构建一阶灰色预测模型,其白化方程为:
    Figure PCTCN2022100006-appb-100002
    式中,a表示发展灰度,b表示内生成控制灰度,且a、b满足:
    Figure PCTCN2022100006-appb-100003
    z (1)(k)=0.5x (1)(k)+0.5x (1)(k-1)
    式中,z (1)表示X (1)紧邻均值生成序列,k=2,3,…,n,T表示转置运算符号。
  5. 根据权利要求4所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述对建立的灰色预测模型进行精度检验与评估,包括以下步骤:
    S131、依据后验差比值c与小误差概率p两个指标对灰色预测模型进行检验,其表达式分别为:
    Figure PCTCN2022100006-appb-100004
    p=P{|q (0)(k)-q|<0.6745s 1}
    式中,q (0)表示残差序列;
    q表示残差序列的均值;
    S 1表示原始序列的标准差;
    S 2表示残差序列的标准差;
    S132、若所述灰色预测模型的精度不符合要求,则使用残差序列建立灰色预测模型对原模型进行修正,提高精度;
    S133、若所述灰色预测模型的精度符合要求,则还原数据与预测值,其运算表达式为:
    Figure PCTCN2022100006-appb-100005
    Figure PCTCN2022100006-appb-100006
    式中,a表示发展灰度;
    b表示内生成控制灰度;
    x (0)表示原始非负序列;
    Figure PCTCN2022100006-appb-100007
    表示原始非负序列的预测值;
    Figure PCTCN2022100006-appb-100008
    表示一次累加序列的预测值;
    e表示自然常数;
    t表示时间。
  6. 根据权利要求5所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,利用所述FLUS模型未来土地系统变化,包括以下步骤:
    S141、采用人工神经网络从原始土地利用数据与驱动因子获取各类用地类型的适宜性概率;
    S142、基于轮盘赌选择的自适应惯性竞争机制进行未来土地变化的模拟,处理多种土地利用类型相互转化时的不确定性与复杂性。
  7. 根据权利要求6所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述采用人工神经网络从原始土地利用数据与驱动因子获取各类用地类型的适宜性概率的运算表达式为:
    Figure PCTCN2022100006-appb-100009
    式中,sp(p,k,t)为第k种地类在栅格p和时间t上的适宜性概率;
    ω j,k为隐藏层与输出层间的权重;
    sigmoid()为隐藏层到输出层的激励函数;
    net j(p,t)为第j个隐藏层栅格p在时间t上所接收到的信号。
  8. 根据权利要求7所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述基于轮盘赌选择的自适应惯性竞争机制进行未来土地变化的模拟,处 理多种土地利用类型相互转化时的不确定性与复杂性,包括以下步骤:
    S1421、计算第k种地类在t时刻的自适应惯性系数
    Figure PCTCN2022100006-appb-100010
    其表达式为:
    Figure PCTCN2022100006-appb-100011
    式中,
    Figure PCTCN2022100006-appb-100012
    分别为t-1、t-2时刻第k种类型用地的需求数量与栅格数量的差值;
    S1422、通过元胞自动机模型的迭代功能,将栅格的概率分配到各用地类型,计算t时刻栅格p转化为k类型用地的概率
    Figure PCTCN2022100006-appb-100013
    其表达式为:
    Figure PCTCN2022100006-appb-100014
    Figure PCTCN2022100006-appb-100015
    式中,sc c→k表示用地类型c转化为k的成本,
    1-sc c→k表示转化的难易程度;
    Figure PCTCN2022100006-appb-100016
    表示领域影响程度;
    Figure PCTCN2022100006-appb-100017
    表示在N阶Moore领域窗口内第k种用地类型的栅格总数;
    ω k表示各用地类型的领域影响程度的权重。
  9. 根据权利要求1所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述采用GAEZ模型估算水稻生产潜力,包括以下步骤:
    S21、根据温度和降水条件来评价水稻空间分布的气候适宜性;
    S22、针对水稻适宜种植区域,采用逐级限制法来计算水稻生产潜力。
  10. 根据权利要求9所述的基于土地系统和气候变化耦合的水稻生产潜力模拟方法,其特征在于,所述生产潜力依次包括光合生产潜力、光温生产潜力、光温水生产潜力、气候生产潜力及水稻生产潜力。
PCT/CN2022/100006 2022-06-21 2022-06-21 基于土地系统和气候变化耦合的水稻生产潜力模拟方法 WO2023245399A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2022/100006 WO2023245399A1 (zh) 2022-06-21 2022-06-21 基于土地系统和气候变化耦合的水稻生产潜力模拟方法
LU503105A LU503105B1 (en) 2022-06-21 2022-11-24 Rice Productive Potential Simulation Method Based on Coupling of Land System and Climate Change

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/100006 WO2023245399A1 (zh) 2022-06-21 2022-06-21 基于土地系统和气候变化耦合的水稻生产潜力模拟方法

Publications (1)

Publication Number Publication Date
WO2023245399A1 true WO2023245399A1 (zh) 2023-12-28

Family

ID=86426699

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/100006 WO2023245399A1 (zh) 2022-06-21 2022-06-21 基于土地系统和气候变化耦合的水稻生产潜力模拟方法

Country Status (2)

Country Link
LU (1) LU503105B1 (zh)
WO (1) WO2023245399A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117933477A (zh) * 2024-01-26 2024-04-26 中国科学院西北生态环境资源研究院 一种青藏高原多年冻土区植被特性时间变化趋势预测方法
CN117933477B (zh) * 2024-01-26 2024-06-07 中国科学院西北生态环境资源研究院 一种青藏高原多年冻土区植被特性时间变化趋势预测方法

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101916337A (zh) * 2010-08-23 2010-12-15 湖南大学 一种基于地理信息系统的水稻生产潜力动态预测方法
CN105447235A (zh) * 2015-11-12 2016-03-30 中山大学 一种土地未来利用情景动态模拟方法
CN110390129A (zh) * 2019-06-11 2019-10-29 同济大学 基于GeoSOS-FLUS的土地利用政策有效性的定量评价方法
AU2021102457A4 (en) * 2021-05-11 2021-07-01 Jiangxi Agricultural University High-resolution coupling simulation system and method for land use and forest landscape process
CN113177345A (zh) * 2021-06-30 2021-07-27 中国科学院地理科学与资源研究所 一种网格化作物种植布局优化方法
CN113222316A (zh) * 2021-01-15 2021-08-06 中山大学 一种基于flus模型和生物多样性模型的变化情景模拟方法
US20210256640A1 (en) * 2020-02-14 2021-08-19 Cibo Technologies, Inc. Method and apparatus for generation and employment of parcel production stability attributes for land parcel valuation
CN113641946A (zh) * 2021-10-15 2021-11-12 中国科学院地理科学与资源研究所 耗水作物种植布局优化方法及装置
CN114357879A (zh) * 2021-12-31 2022-04-15 中山大学 一种未来土地利用模拟方法、装置、存储介质及终端设备

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101916337A (zh) * 2010-08-23 2010-12-15 湖南大学 一种基于地理信息系统的水稻生产潜力动态预测方法
CN105447235A (zh) * 2015-11-12 2016-03-30 中山大学 一种土地未来利用情景动态模拟方法
CN110390129A (zh) * 2019-06-11 2019-10-29 同济大学 基于GeoSOS-FLUS的土地利用政策有效性的定量评价方法
US20210256640A1 (en) * 2020-02-14 2021-08-19 Cibo Technologies, Inc. Method and apparatus for generation and employment of parcel production stability attributes for land parcel valuation
CN113222316A (zh) * 2021-01-15 2021-08-06 中山大学 一种基于flus模型和生物多样性模型的变化情景模拟方法
AU2021102457A4 (en) * 2021-05-11 2021-07-01 Jiangxi Agricultural University High-resolution coupling simulation system and method for land use and forest landscape process
CN113177345A (zh) * 2021-06-30 2021-07-27 中国科学院地理科学与资源研究所 一种网格化作物种植布局优化方法
CN113641946A (zh) * 2021-10-15 2021-11-12 中国科学院地理科学与资源研究所 耗水作物种植布局优化方法及装置
CN114357879A (zh) * 2021-12-31 2022-04-15 中山大学 一种未来土地利用模拟方法、装置、存储介质及终端设备

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117933477A (zh) * 2024-01-26 2024-04-26 中国科学院西北生态环境资源研究院 一种青藏高原多年冻土区植被特性时间变化趋势预测方法
CN117933477B (zh) * 2024-01-26 2024-06-07 中国科学院西北生态环境资源研究院 一种青藏高原多年冻土区植被特性时间变化趋势预测方法

Also Published As

Publication number Publication date
LU503105B1 (en) 2023-05-24

Similar Documents

Publication Publication Date Title
CN110084367B (zh) 一种基于lstm深度学习模型的土壤墒情预测方法
CN108304668B (zh) 一种结合水文过程数据和历史先验数据的洪水预测方法
Waongo et al. Adaptation to climate change: The impacts of optimized planting dates on attainable maize yields under rainfed conditions in Burkina Faso
Matsumura et al. Maize yield forecasting by linear regression and artificial neural networks in Jilin, China
CN111027175B (zh) 基于耦合模型集成模拟的洪水对社会经济影响的评估方法
Cantelaube et al. Seasonal weather forecasts for crop yield modelling in Europe
CN101480143B (zh) 一种预测灌区作物单产量的方法
CN111665575B (zh) 一种基于统计动力的中长期降雨分级耦合预报方法及系统
KR20170005553A (ko) 지능형 정보제공을 위한 홍수, 가뭄정보의 평가 및 예측 기법 개발
CN112288164B (zh) 一种计及空间相关性和修正数值天气预报的风功率组合预测方法
CN106251027B (zh) 基于模糊支持向量分位数回归的电力负荷概率密度预测方法
Sultan et al. Estimating the potential economic value of seasonal forecasts in West Africa: A long-term ex-ante assessment in Senegal
CN113902580B (zh) 一种基于随机森林模型的历史耕地分布重建方法
CN111539597B (zh) 一种网格化的流域社会经济干旱评估方法
Aghelpour et al. Forecasting different types of droughts simultaneously using multivariate standardized precipitation index (MSPI), MLP neural network, and imperialistic competitive algorithm (ICA)
Verma et al. Comparative analysis of CMIP5 and CMIP6 in conjunction with the hydrological processes of reservoir catchment, Chhattisgarh, India
CN110119767A (zh) 一种基于lvq神经网络的黄瓜温室温度智能化检测装置
Ali et al. Modeling wheat yield with data-intelligent algorithms: artificial neural network versus genetic programming and minimax probability machine regression
CN117909888B (zh) 智慧人工智能气候预测方法
Li et al. An Integrated Artificial Neural Network-based Precipitation Revision Model.
CN117787081A (zh) 一种基于Morris和Sobol法的水文模型参数不确定性分析方法
WO2023245399A1 (zh) 基于土地系统和气候变化耦合的水稻生产潜力模拟方法
Nasseri et al. A comparison between direct and indirect frameworks to evaluate impacts of climate change on streamflows: case study of Karkheh River basin in Iran
CN116663404A (zh) 一种耦合人工智能和贝叶斯理论的洪水预报方法及系统
CN115358587A (zh) 区域多部门协同的基础设施规划方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22947180

Country of ref document: EP

Kind code of ref document: A1