Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau

As the world’s largest loess deposit region, the Loess Plateau’s vegetation dynamics are crucial for its regional water–heat balance and ecosystem functioning. Leaf Area Index (LAI) serves as a key indicator bridging canopy architecture and plant physiological activities. Existing studies have made...

Full description

Saved in:
Bibliographic Details
Main Authors: Junpo Yu, Yajun Si, Wen Zhao, Zeyu Zhou, Jiming Jin, Wenjun Yan, Xiangyu Shao, Zhixiang Xu, Junwei Gan
Format: Article
Language:English
Published: MDPI AG 2025-08-01
Series:Plants
Subjects:
Online Access:https://www.mdpi.com/2223-7747/14/15/2391
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849406466152202240
author Junpo Yu
Yajun Si
Wen Zhao
Zeyu Zhou
Jiming Jin
Wenjun Yan
Xiangyu Shao
Zhixiang Xu
Junwei Gan
author_facet Junpo Yu
Yajun Si
Wen Zhao
Zeyu Zhou
Jiming Jin
Wenjun Yan
Xiangyu Shao
Zhixiang Xu
Junwei Gan
author_sort Junpo Yu
collection DOAJ
description As the world’s largest loess deposit region, the Loess Plateau’s vegetation dynamics are crucial for its regional water–heat balance and ecosystem functioning. Leaf Area Index (LAI) serves as a key indicator bridging canopy architecture and plant physiological activities. Existing studies have made significant advancements in simulating LAI, yet accurate LAI simulation remains challenging. To address this challenge and gain deeper insights into the environmental controls of LAI, this study aims to accurately simulate LAI in the Loess Plateau using deep learning models and to elucidate the spatiotemporal influence of soil moisture and temperature on LAI dynamics. For this purpose, we used three deep learning models, namely Artificial Neural Network (ANN), Long Short-Term Memory (LSTM), and Interpretable Multivariable (IMV)-LSTM, to simulate LAI in the Loess Plateau, only using soil moisture and temperature as inputs. Results indicated that our approach outperformed traditional models and effectively captured LAI variations across different vegetation types. The attention analysis revealed that soil moisture mainly influenced LAI in the arid northwest and temperature was the predominant effect in the humid southeast. Seasonally, soil moisture was crucial in spring and summer, notably in grasslands and croplands, whereas temperature dominated in autumn and winter. Notably, forests had the longest temperature-sensitive periods. As LAI increased, soil moisture became more influential, and at peak LAI, both factors exerted varying controls on different vegetation types. These findings demonstrated the strength of deep learning for simulating vegetation–climate interactions and provided insights into hydrothermal regulation mechanisms in semiarid regions.
format Article
id doaj-art-0064d489ab274115b539fda48a49e8ca
institution Kabale University
issn 2223-7747
language English
publishDate 2025-08-01
publisher MDPI AG
record_format Article
series Plants
spelling doaj-art-0064d489ab274115b539fda48a49e8ca2025-08-20T03:36:22ZengMDPI AGPlants2223-77472025-08-011415239110.3390/plants14152391Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess PlateauJunpo Yu0Yajun Si1Wen Zhao2Zeyu Zhou3Jiming Jin4Wenjun Yan5Xiangyu Shao6Zhixiang Xu7Junwei Gan8College of Resources and Environment, Yangtze University, Wuhan 430100, ChinaCollege of Water Resources and Architectural Engineering, Northwest A&F University, Yangling 712100, ChinaLanzhou Institute of Arid Meteorology, China Meteorological Administration/Key Laboratory of Arid Climate Change and Reducing Disaster of Gansu Province/Key Open Laboratory of Arid Climate Change and Reducing Disaster, China Meteorological Administration, Lanzhou 730020, ChinaCollege of Resources and Environment, Yangtze University, Wuhan 430100, ChinaCollege of Resources and Environment, Yangtze University, Wuhan 430100, ChinaLanzhou Institute of Arid Meteorology, China Meteorological Administration/Key Laboratory of Arid Climate Change and Reducing Disaster of Gansu Province/Key Open Laboratory of Arid Climate Change and Reducing Disaster, China Meteorological Administration, Lanzhou 730020, ChinaCollege of Resources and Environment, Yangtze University, Wuhan 430100, ChinaCollege of Resources and Environment, Yangtze University, Wuhan 430100, ChinaCollege of Resources and Environment, Yangtze University, Wuhan 430100, ChinaAs the world’s largest loess deposit region, the Loess Plateau’s vegetation dynamics are crucial for its regional water–heat balance and ecosystem functioning. Leaf Area Index (LAI) serves as a key indicator bridging canopy architecture and plant physiological activities. Existing studies have made significant advancements in simulating LAI, yet accurate LAI simulation remains challenging. To address this challenge and gain deeper insights into the environmental controls of LAI, this study aims to accurately simulate LAI in the Loess Plateau using deep learning models and to elucidate the spatiotemporal influence of soil moisture and temperature on LAI dynamics. For this purpose, we used three deep learning models, namely Artificial Neural Network (ANN), Long Short-Term Memory (LSTM), and Interpretable Multivariable (IMV)-LSTM, to simulate LAI in the Loess Plateau, only using soil moisture and temperature as inputs. Results indicated that our approach outperformed traditional models and effectively captured LAI variations across different vegetation types. The attention analysis revealed that soil moisture mainly influenced LAI in the arid northwest and temperature was the predominant effect in the humid southeast. Seasonally, soil moisture was crucial in spring and summer, notably in grasslands and croplands, whereas temperature dominated in autumn and winter. Notably, forests had the longest temperature-sensitive periods. As LAI increased, soil moisture became more influential, and at peak LAI, both factors exerted varying controls on different vegetation types. These findings demonstrated the strength of deep learning for simulating vegetation–climate interactions and provided insights into hydrothermal regulation mechanisms in semiarid regions.https://www.mdpi.com/2223-7747/14/15/2391LAIdeep learningwater–heat couplingvegetation dynamics simulationLoess Plateau
spellingShingle Junpo Yu
Yajun Si
Wen Zhao
Zeyu Zhou
Jiming Jin
Wenjun Yan
Xiangyu Shao
Zhixiang Xu
Junwei Gan
Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
Plants
LAI
deep learning
water–heat coupling
vegetation dynamics simulation
Loess Plateau
title Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
title_full Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
title_fullStr Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
title_full_unstemmed Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
title_short Comparison of Deep Learning Models for LAI Simulation and Interpretable Hydrothermal Coupling in the Loess Plateau
title_sort comparison of deep learning models for lai simulation and interpretable hydrothermal coupling in the loess plateau
topic LAI
deep learning
water–heat coupling
vegetation dynamics simulation
Loess Plateau
url https://www.mdpi.com/2223-7747/14/15/2391
work_keys_str_mv AT junpoyu comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT yajunsi comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT wenzhao comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT zeyuzhou comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT jimingjin comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT wenjunyan comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT xiangyushao comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT zhixiangxu comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau
AT junweigan comparisonofdeeplearningmodelsforlaisimulationandinterpretablehydrothermalcouplingintheloessplateau