Upscaling net ecosystem CO2 exchanges in croplands: The application of integrating object-based image analysis and machine learning approaches

Accurately estimating the net ecosystem exchange of CO2 (NEE) in cropland ecosystems is essential for understanding the impacts of agricultural practices and climate conditions. However, significant uncertainties persist in the estimation of regional cropland NEE due to landscape heterogeneity and v...

Full description

Saved in:
Bibliographic Details
Published in:The Science of the total environment Vol. 944; p. 173887
Main Authors: Gao, Dexiang, Yao, Jingyu, Gao, Zhongming, Yuan, Wenping, He, Yingzhe, Wang, Bojun, Li, Lei, Ma, Yulong, Russell, Eric, Pressley, Shelley N., Zou, Xudong
Format: Journal Article
Language:English
Published: Elsevier B.V 20.09.2024
Subjects:
algorithms
carbon dioxide
Carbon flux upscaling
climate
cropland
Eddy covariance
environment
image analysis
Machine learning
Net ecosystem exchange
Object-oriented image analysis
phenology
prediction
soil water
temperature
Net ecosystem exchange
Eddy covariance
Carbon flux upscaling
Object-oriented image analysis
Machine learning
ISSN:0048-9697, 1879-1026, 1879-1026
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accurately estimating the net ecosystem exchange of CO2 (NEE) in cropland ecosystems is essential for understanding the impacts of agricultural practices and climate conditions. However, significant uncertainties persist in the estimation of regional cropland NEE due to landscape heterogeneity and variations in the efficacy of upscaling models. Here, we applied an integrated approach that combined object-based image analysis (OBIA) techniques with advanced machine learning (ML) approaches to upscale regional cropland NEE. We conducted a thorough evaluation of the upscaling approach across four distinct cropland areas characterized by diverse climate conditions. Our study confirmed that OBIA techniques can efficiently segment cropland objects, thereby enhancing the representation and accuracy of characteristics relevant to cropland features. The sequential least squares programming algorithm, among the three methods used for ML model integration, demonstrated exceptional performance in predicting NEE, with an R2 value exceeding 0.80 across all study areas and peaking at 0.90 in the most successful area. On average, there was an 18 % improvement compared to the poorest-performing ML model and a 6 % enhancement compared to the best-performing ML model. The upscaled regional products exhibited superior performance in characterizing cropland NEE patterns compared to pixel-based products. Additionally, we utilized the SHapley Additive exPlanations (SHAP) to assess driver importance, revealing that phenology and radiation had the greatest influence on prediction accuracy, followed by temperature and soil moisture. This study highlights the potential of integrating OBIA techniques with machine learning approaches for upscaling regional cropland NEE, while concurrently reducing estimation uncertainties. [Display omitted] •Evaluated the performance of an object-based integrated machine learning framework for upscaling CO2 fluxes in croplands•Explored key drivers affecting NEE prediction accuracy in conjunction with interpretable machine learning SHAP models•Generated maps of upscaled NEE and associated uncertainties for croplands in the study areas
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0048-9697
1879-1026
1879-1026
DOI:10.1016/j.scitotenv.2024.173887