Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices

In the context of climate change and the development of sustainable agricultural, crop yield prediction is key to ensuring food security. In this study, long-term vegetation and meteorological indices were obtained from the MOD09A1 product and daily weather data. Three types of time series data were...

Full description

Saved in:
Bibliographic Details
Main Authors: Hao Xu, Hongfei Yin, Jia Liu, Lei Wang, Wenjie Feng, Hualu Song, Yangyang Fan, Kangkang Qi, Zhichao Liang, WenJie Li, Xiaohu Zhang, Rongjuan Zhang, Shuai Wang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/15/5/1114
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327769048055808
author Hao Xu
Hongfei Yin
Jia Liu
Lei Wang
Wenjie Feng
Hualu Song
Yangyang Fan
Kangkang Qi
Zhichao Liang
WenJie Li
Xiaohu Zhang
Rongjuan Zhang
Shuai Wang
author_facet Hao Xu
Hongfei Yin
Jia Liu
Lei Wang
Wenjie Feng
Hualu Song
Yangyang Fan
Kangkang Qi
Zhichao Liang
WenJie Li
Xiaohu Zhang
Rongjuan Zhang
Shuai Wang
author_sort Hao Xu
collection DOAJ
description In the context of climate change and the development of sustainable agricultural, crop yield prediction is key to ensuring food security. In this study, long-term vegetation and meteorological indices were obtained from the MOD09A1 product and daily weather data. Three types of time series data were constructed by aggregating data from an 8-day period (DP), 9-month period (MP), and six growth periods (GP). And we developed the yield prediction model by using random forest (RF) and long short-term memory (LSTM) networks. Results showed that the average root mean squared error (RMSE) of the RF model in each province was 0.5 Mg/ha lower than that of the LSTM model. Both the RF and LSTM prediction accuracies increased with the later growth stages data. Partial dependence plots showed that the influence degree of DVI on yield was above 2 Mg/ha. When the time length of the feature variables was shortened to MP or GP, the growing degree days (GDD), average minimum temperature (AveTmin), and effective precipitation (EP) showed stronger nonlinear relationships with the statistical yields.
format Article
id doaj-art-05472ac262cb4a14a2a1d10ff88ab04e
institution Kabale University
issn 2073-4395
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-05472ac262cb4a14a2a1d10ff88ab04e2025-08-20T03:47:48ZengMDPI AGAgronomy2073-43952025-04-01155111410.3390/agronomy15051114Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological IndicesHao Xu0Hongfei Yin1Jia Liu2Lei Wang3Wenjie Feng4Hualu Song5Yangyang Fan6Kangkang Qi7Zhichao Liang8WenJie Li9Xiaohu Zhang10Rongjuan Zhang11Shuai Wang12Shandong Academy of Agricultural Sciences, Jinan 250100, ChinaSchool of Finance and Taxation, Shandong University of Finance and Economics, Jinan 250014, ChinaChinese Academy of Agricultural Sciences, Beijing 100081, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaQilu Aerospace Information Research Institute, Jinan 100094, ChinaNational Engineering and Technology Center for Information Agriculture, Nanjing Agricultural University, Nanjing 210095, ChinaDongying Academy of Agricultural Sciences, Dongying 257091, ChinaShandong Academy of Agricultural Sciences, Jinan 250100, ChinaIn the context of climate change and the development of sustainable agricultural, crop yield prediction is key to ensuring food security. In this study, long-term vegetation and meteorological indices were obtained from the MOD09A1 product and daily weather data. Three types of time series data were constructed by aggregating data from an 8-day period (DP), 9-month period (MP), and six growth periods (GP). And we developed the yield prediction model by using random forest (RF) and long short-term memory (LSTM) networks. Results showed that the average root mean squared error (RMSE) of the RF model in each province was 0.5 Mg/ha lower than that of the LSTM model. Both the RF and LSTM prediction accuracies increased with the later growth stages data. Partial dependence plots showed that the influence degree of DVI on yield was above 2 Mg/ha. When the time length of the feature variables was shortened to MP or GP, the growing degree days (GDD), average minimum temperature (AveTmin), and effective precipitation (EP) showed stronger nonlinear relationships with the statistical yields.https://www.mdpi.com/2073-4395/15/5/1114crop yieldphenologyrandom forestlong short-term memory
spellingShingle Hao Xu
Hongfei Yin
Jia Liu
Lei Wang
Wenjie Feng
Hualu Song
Yangyang Fan
Kangkang Qi
Zhichao Liang
WenJie Li
Xiaohu Zhang
Rongjuan Zhang
Shuai Wang
Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
Agronomy
crop yield
phenology
random forest
long short-term memory
title Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
title_full Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
title_fullStr Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
title_full_unstemmed Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
title_short Prediction of Spatial Winter Wheat Yield by Combining Multiscale Time Series of Vegetation and Meteorological Indices
title_sort prediction of spatial winter wheat yield by combining multiscale time series of vegetation and meteorological indices
topic crop yield
phenology
random forest
long short-term memory
url https://www.mdpi.com/2073-4395/15/5/1114
work_keys_str_mv AT haoxu predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT hongfeiyin predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT jialiu predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT leiwang predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT wenjiefeng predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT hualusong predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT yangyangfan predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT kangkangqi predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT zhichaoliang predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT wenjieli predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT xiaohuzhang predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT rongjuanzhang predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices
AT shuaiwang predictionofspatialwinterwheatyieldbycombiningmultiscaletimeseriesofvegetationandmeteorologicalindices