Improvement of HLM Modeling for Winter Wheat Yield Estimation Under Drought Conditions

doi:10.12133/j.smartag.SA202408009

Abstract

Abstract:

[Objective] Winter wheat yield is crucial for national food security and the standard of living of the population. Existing crop yield prediction models often show low accuracy under disaster-prone climatic conditions. This study proposed an improved hierarchical linear model (IHLM) based on a drought weather index reduction rate, aiming to enhance the accuracy of crop yield estimation under drought conditions. [Methods] HLM was constructed using the maximum enhanced vegetation index-2 (EVI2max), meteorological data (precipitation, radiation, and temperature from March to May), and observed winter wheat yield data from 160 agricultural survey stations in Shandong province (2018－2021). To validate the model's accuracy, 70% of the data from Shandong province was randomly selected for model construction, and the remaining data was used to validate the accuracy of the yield model. HLM considered the variation in meteorological factors as a key obstacle affecting crop growth and improved the model by calculating the relative meteorological factors. The calculation of relative meteorological factors helped reduce the impact of inter-annual differences in meteorological data. The accuracy of the HLM model was compared with that of the random forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoost) models. The HLM model provided more intuitive interpretation, especially suitable for processing hierarchical data, which helped capture the variability of winter wheat yield data under drought conditions. Therefore, a drought weather index reduction rate model from the agricultural insurance industry was introduced to further optimize the HLM model, resulting in the construction of the IHLM model. The IHLM model was designed to improve crop yield prediction accuracy under drought conditions. Since the precipitation differences between Henan and Shandong provinces were small, to test the transferability of the IHLM model, Henan province sample data was processed in the same way as in Shandong, and the IHLM model was applied to Henan province to evaluate its performance under different geographical conditions. [Results and Discussions] The accuracy of the HLM model, improved based on relative meteorological factors (rMF), was higher than that of RF, SVR, and XGBoost. The validation accuracy showed a Pearson correlation coefficient (r) of 0.76, a root mean squared error (RMSE) of 0.60 t/hm², and a normalized RMSE (nRMSE) of 11.21%. In the drought conditions dataset, the model was further improved by incorporating the relationship between the winter wheat drought weather index and the reduction rate of winter wheat yield. After the improvement, the RMSE decreased by 0.48 t/hm², and the nRMSE decreased by 28.64 percentage points, significantly enhancing the accuracy of the IHLM model under drought conditions. The IHLM model also demonstrated good applicability when transferred to Henan province. [Conclusions] The IHLM model developed in this study improved the accuracy and stability of crop yield predictions, especially under drought conditions. Compared to RF, SVR, and XGBoost models, the IHLM model was more suitable for predicting winter wheat yield. This research can be widely applied in the agricultural insurance field, playing a significant role in the design of agricultural insurance products, rate setting, and risk management. It enables more accurate predictions of winter wheat yield under drought conditions, with results that are closer to actual outcomes.

Key words: winter wheat, yield prediction, HLM model, drought conditions, mechine learning

CLC Number:

S-3
TP181

ZHAO Peiqin, LIU Changbin, ZHENG Jie, MENG Yang, MEI Xin, TAO Ting, ZHAO Qian, MEI Guangyuan, YANG Xiaodong. Improvement of HLM Modeling for Winter Wheat Yield Estimation Under Drought Conditions[J]. Smart Agriculture, 2025, 7(2): 106-116.

Figures/Tables 13

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Table 1

Table 2

Fig. 6

Table 3

Table 4

Fig.7

Fig. 8

Fig.9

References 34

1	WANG L G, TIAN Y C, YAO X, et al. Predicting grain yield and protein content in wheat by fusing multi-sensor and multi-temporal remote-sensing images[J]. Field crops research, 2014, 164: 178-188.
2	陶惠林, 徐良骥, 冯海宽, 等. 基于无人机高光谱遥感数据的冬小麦产量估算[J]. 农业机械学报, 2020, 51(7): 146-155.
	TAO H L, XU L J, FENG H K, et al. Winter wheat yield estimation based on UAV hyperspectral remote sensing data[J]. Transactions of the Chinese society for agricultural machinery, 2020, 51(7): 146-155.
3	CUI K, SHOEMAKER S P. A look at food security in China[J]. NPJ science of food, 2018, 2: 4.
4	BENAMI E, JIN Z N, CARTER M R, et al. Uniting remote sensing, crop modelling and economics for agricultural risk management[J]. Nature reviews earth & environment, 2021, 2: 140-159.
5	王妍, 张晓龙, 石嘉丽, 等. 中国冬小麦主产区气候变化及其对小麦产量影响研究[J]. 中国生态农业学报(中英文), 2022, 30(5): 723-734.
	WANG Y, ZHANG X L, SHI J L, et al. Climate change and its effect on winter wheat yield in the main winter wheat production areas of China[J]. Chinese journal of eco-agriculture, 2022, 30(5): 723-734.
6	AJILOGBA C F, WALKER S. Modeling climate change impact on dryland wheat production for increased crop yield in the Free State, South Africa, using GCM projections and the DSSAT model[J]. Frontiers in environmental science, 2023, 11: ID 1067008.
7	MIRGOL B, NAZARI M, ETEGHADIPOUR M. Modelling climate change impact on irrigation water requirement and yield of winter wheat (triticum aestivum L.), barley (hordeum vulgare L.), and fodder maize (zea mays L.) in the Semi-arid qazvin plateau, Iran[J]. Agriculture, 2020, 10(3): ID 60.
8	BRACHO-MUJICA G, RÖTTER R P, HAAKANA M, et al. Effects of changes in climatic means, variability, and agro-technologies on future wheat and maize yields at 10 sites across the globe[J]. Agricultural and forest meteorology, 2024, 346: ID 109887.
9	杨艳颖, 毛克彪, 韩秀珍, 等. 1949—2016年中国旱灾规律及其对粮食产量的影响[J]. 中国农业信息, 2018, 30(5): 76-90.
	YANG Y Y, MAO K B, HAN X Z, et al. Characteristics of drought disaster and its impact on grain production in China from 1949 to 2016[J]. China agricultural informatics, 2018, 30(5): 76-90.
10	NXUMALO G, BASHIR B, ALSAFADI K, et al. Meteorological drought variability and its impact on wheat yields across South Africa[J]. International journal of environmental research and public health, 2022, 19(24): ID 16469.
11	GUAN X K, SONG L, WANG T C, et al. Effect of drought on the gas exchange, chlorophyll fluorescence and yield of six different-era spring wheat cultivars[J]. Journal of agronomy and crop science, 2015, 201(4): 253-266.
12	曹雯, 成林, 杨太明, 等. 河南省冬小麦拔节-抽穗期干旱天气指数保险研究[J]. 气象, 2019, 45(2): 274-281.
	CAO W, CHENG L, YANG T M, et al. Study on weather index insurance of drought damage at jointing-heading stage of winter wheat in Henan province[J]. Meteorological monthly, 2019, 45(2): 274-281.
13	杨太明, 许莹, 孙喜波, 等. 安徽省夏玉米干旱天气指数保险产品设计及应用[J]. 气象, 2016, 42(4): 450-455.
	YANG T M, XU Y, SUN X B, et al. Design and application of the drought weather index insurance of summer corn in Anhui province[J]. Meteorological monthly, 2016, 42(4): 450-455.
14	王靖雯. 基于遥感模型的作物产量差估算体系研究[D]. 北京: 中国科学院大学(中国科学院空天信息创新研究院), 2022.
	WANG J W. A methodological study on estimating crop yield gap based on remote sensing models[D]. Beijing: University of Chinese Academy of Sciences (Aerospace Information Research Institute, Chinese Academy of Sciences), 2022.
15	LI Z H, TAYLOR J, YANG H, et al. A hierarchical interannual wheat yield and grain protein prediction model using spectral vegetative indices and meteorological data[J]. Field crops research, 2020, 248: ID 107711.
16	DEBAEKE P, ATTIA F, CHAMPOLIVIER L, et al. Forecasting sunflower grain yield using remote sensing data and statistical models[J]. European journal of agronomy, 2023, 142: ID 126677.
17	ZHUANG J Y, XU S W, LI G Q, et al. The influence of meteorological factors on wheat and rice yields in China[J]. Crop science, 2018, 58(3): 1440-1445.
18	韩少宇. 基于多平台遥感数据的冬小麦长势监测和产量预测[D]. 郑州: 河南农业大学, 2023.
	HAN S Y. Growth monitoring and yield prediction of winter wheat based on multi-platform remote sensing data[D]. Zhengzhou: Henan Agricultural University, 2023.
19	KAYAD A, RODRIGUES F A, NARANJO S, et al. Radiative transfer model inversion using high-resolution hyperspectral airborne imagery-Retrieving maize LAI to access biomass and grain yield[J]. Field crops research, 2022, 282: ID 108449.
20	BELGIU M, DRĂGUŢ L. Random forest in remote sensing: A review of applications and future directions[J]. ISPRS journal of photogrammetry and remote sensing, 2016, 114: 24-31.
21	CHEN R Q, ZHANG C J, XU B, et al. Predicting individual apple tree yield using UAV multi-source remote sensing data and ensemble learning[J]. Computers and electronics in agriculture, 2022, 201: ID 107275.
22	FEI S P, HASSAN M A, XIAO Y G, et al. UAV-based multi-sensor data fusion and machine learning algorithm for yield prediction in wheat[J]. Precision agriculture, 2023, 24(1): 187-212.
23	李红艳, 徐建强, 许甫金, 等. 气象因素对水稻产量的影响及预测模型的建立[J]. 浙江农业科学, 2018, 59(7): 1104-1107, 1110.
	LI H Y, XU J Q, XU F J, et al. Influences and prediction model of rice yield based on meteorological[J]. Journal of Zhejiang agricultural sciences, 2018, 59(7): 1104-1107, 1110.
24	CARNEIRO F M, DE BRITO FILHO A L, FERREIRA F M, et al. Soil and satellite remote sensing variables importance using machine learning to predict cotton yield[J]. Smart agricultural technology, 2023, 5: ID 100292.
25	王永强. 基于作物模型与遥感数据同化的区域尺度夏玉米生长模拟与灌溉施肥制度优化[D]. 杨凌: 西北农林科技大学, 2023.
	WANG Y Q. Simulation of summer maize growth and optimization of irrigation and fertilization system on regional scale based on crop model and assimilation of remote sensing data[D]. Yangling: Northwest A & F University, 2023.
26	HUANG J X, SEDANO F, HUANG Y B, et al. Assimilating a synthetic Kalman filter leaf area index series into the WOFOST model to improve regional winter wheat yield estimation[J]. Agricultural and forest meteorology, 2016, 216: 188-202.
27	KANG Y H, ÖZDOĞAN M. Field-level crop yield mapping with Landsat using a hierarchical data assimilation approach[J]. Remote sensing of environment, 2019, 228: 144-163.
28	ZHUO W, FANG S B, GAO X R, et al. Crop yield prediction using MODIS LAI, TIGGE weather forecasts and WOFOST model: A case study for winter wheat in Hebei, China during 2009–2013[J]. International journal of applied earth observation and geoinformation, 2022, 106: ID 102668.
29	XU X B, NIE C W, JIN X L, et al. A comprehensive yield evaluation indicator based on an improved fuzzy comprehensive evaluation method and hyperspectral data[J]. Field crops research, 2021, 270: ID 108204.
30	JOHNSON M D, HSIEH W W, CANNON A J, et al. Crop yield forecasting on the Canadian Prairies by remotely sensed vegetation indices and machine learning methods[J]. Agricultural and forest meteorology, 2016, 218: 74-84.
31	YU W G, LI D, ZHENG H B, et al. HIDYM: A high-resolution gross primary productivity and dynamic harvest index based crop yield mapper[J]. Remote sensing of environment, 2024, 311: ID 114301.
32	AHLUWALIA O, SINGH P C, BHATIA R. A review on drought stress in plants: Implications, mitigation and the role of plant growth promoting rhizobacteria[J]. Resources, environment and sustainability, 2021, 5: ID 100032.
33	ZHAO Y, HAN S Y, ZHENG J, et al. ChinaWheatYield30m: A 30 m annual winter wheat yield dataset from 2016 to 2021 in China[J]. Earth system science data, 2023, 15(9): 4047-4063.
34	ZHAO Y, MENG Y, HAN S Y, et al. Should phenological information be applied to predict agronomic traits across growth stages of winter wheat?[J]. The crop journal, 2022, 10(5): 1346-1352.

数据集	样本数量/个	产量/（t/hm²）
数据集	样本数量/个	最小值	最大值	均值	标准差	变异系数
2018	28	5.63	9.50	7.68	1.12	0.15
2019	37	6.41	9.57	8.18	0.78	0.10
2020	37	4.27	9.32	8.00	0.87	0.11
2021	58	4.70	10.07	8.04	0.95	0.12
建模集	112	4.70	10.07	8.04	0.93	0.12
验证集	48	4.27	9.39	7.90	0.96	0.12
总计	160	4.27	10.07	8.00	0.94	0.12

数据集	3月			4月			5月
数据集	Pre/mm	Rad/（MJ/m²）	Tem/℃	Pre/mm	Rad/（MJ/m²）	Tem/℃	Pre/mm	Rad/（MJ/m²）	Tem/℃
2018	0.92	13.97	10.28	1.76	16.0	16.41	2.53	16.56	21.25
2019	0.32	15.47	10.52	1.55	15.21	14.99	0.27	19.26	22.16
2020	0.31	14.53	10.60	1.08	17.82	14.56	1.72	18.34	21.82
2021	0.68	13.15	9.72	1.16	14.58	14.21	1.44	18.30	20.65

模型类别	超参数	超参数范围	最优超参数
RF	子树的数量	100，200，500，1 000，2 000，4 000	2 000
RF	最大生长深度	1，3，5，10，20	3
SVR	惩罚系数	100，500，1 000，2 000，4 000	1 000
SVR	核函数类型	linear，rbf，poly	linear
XGBOOST	子树的数量	50，60，70，80	70
	最大深度	11，13，15，17，19，21	15
	学习率	0.01，0.1，0.2	0.1

模型类别	r	RMSE/（t/hm²）	nRMSE/%
RF	0.71	0.69	13.50
SVR	0.74	0.71	13.90
XGBOOST	0.72	0.67	13.04
HLM	0.72	0.74	14.41
HLM基于rMF改进后	0.75	0.65	12.80

[1]	LIU Ruixuan, ZHANG Fangzhao, ZHANG Jibo, LI Zhenhai, YANG Juntao. Suitable Sowing Date Method of Winter Wheat at the County Level Based on ECMWF Long-Term Reanalysis Data [J]. Smart Agriculture, 2024, 6(5): 51-60.
[2]	WANG Jingyong, ZHANG Mingzhen, LING Huarong, WANG Ziting, GAI Jingyao. A Hyperspectral Image-Based Method for Estimating Water and Chlorophyll Contents in Maize Leaves under Drought Stress [J]. Smart Agriculture, 2023, 5(3): 142-153.
[3]	HU Yanan, LIANG Ju, LIANG Shefang, LI Shijuan, ZHU Yeping, E Yue. Optimum Sowing Date of Winter Wheat in Next 40 Years Based on DSSAT-CERES-Wheat Model [J]. Smart Agriculture, 2021, 3(2): 68-76.
[4]	YANG Feifei, LIU Shengping, ZHU Yeping, LI Shijuan. Identification and Level Discrimination of Waterlogging Stress in Winter Wheat Using Hyperspectral Remote Sensing [J]. Smart Agriculture, 2021, 3(2): 35-44.
[5]	WANG Lin, LIANG Jian, MENG Fanyu, MENG Yang, ZHANG Yongtao, LI Zhenhai. Estimating Grain Protein Content of Winter Wheat in Producing Areas Based on Remote Sensing and Meteorological Data [J]. Smart Agriculture, 2021, 3(2): 15-22.