Explore the relationship between Bowen ratio and evapotranspiration in wetlands using the maximum entropy production model
(2025) In Journal of Hydrology 661.- Abstract
Wetlands play a critical role in the global water cycle and ecological health. However, understanding and accurately parameterizing water and heat exchanges in these areas is challenging due to the heterogeneous nature of wetland surfaces. This study extends the maximum entropy production (MEP) model to simulate wetland evapotranspiration (ET) by integrating NDVI dynamics, water evaporation, and plant transpiration. The model was applied to six FLUXNET wetland sites, with observation periods ranging from 1 to 8 years at daily temporal resolution. Across all time scales, the Nash-Sutcliffe Efficiency (NSE) improved by up to + 0.30 compared to the original model, with NSE values ranging from 0.67 to 0.83 and coefficients of determination... (More)
Wetlands play a critical role in the global water cycle and ecological health. However, understanding and accurately parameterizing water and heat exchanges in these areas is challenging due to the heterogeneous nature of wetland surfaces. This study extends the maximum entropy production (MEP) model to simulate wetland evapotranspiration (ET) by integrating NDVI dynamics, water evaporation, and plant transpiration. The model was applied to six FLUXNET wetland sites, with observation periods ranging from 1 to 8 years at daily temporal resolution. Across all time scales, the Nash-Sutcliffe Efficiency (NSE) improved by up to + 0.30 compared to the original model, with NSE values ranging from 0.67 to 0.83 and coefficients of determination (R2) between 0.80 and 0.97. Additionally, the root mean square error (RMSE) was reduced to as low as 7.61 W·m−2 for monthly estimates. These results demonstrate the enhanced model's ability to capture the temporal dynamics of ET more accurately. Furthermore, the analysis revealed significant correlations between wetland vegetation, the Bowen ratio, and the transpiration-to-evapotranspiration ratio. To further quantify the contributions of multiple environmental factors to wetland energy flux distribution, a structural equation modeling (SEM) approach was employed. The SEM analysis confirms that vegetation exerts an important influence, while other variables such as temperature and radiation also significantly affect the partitioning between latent and sensible heat fluxes. These findings improve the understanding of wetland ET dynamics and provide a foundation for further exploration of the relationship between NDVI and energy flux partitioning.
(Less)
- author
- Sun, Huaiwei
; Fu, Leying
; Wang, Weiguang
; Xue, Jie
; Wang, Jing
; Liao, Weihong
; Li, Haicheng
; Sun, Xunlai
; Yang, Yong
; Wang, Jingfeng
; Zhang, Hong
; Chen, Fulong
; Zheng, Qiming
; Meng, Changqing
and Zhang, Wenxin
LU
(Less) - publishing date
- 2025-11
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Energy flux partitioning, Evapotranspiration, Maximum entropy production, Wetland
- in
- Journal of Hydrology
- volume
- 661
- article number
- 133586
- publisher
- Elsevier
- external identifiers
-
- scopus:105007356903
- ISSN
- 0022-1694
- DOI
- 10.1016/j.jhydrol.2025.133586
- language
- English
- LU publication?
- no
- additional info
- Publisher Copyright: © 2025
- id
- d5b8aedf-1fe3-46f6-a076-3bc1945456ea
- date added to LUP
- 2025-08-13 12:29:51
- date last changed
- 2025-08-14 03:19:41
@article{d5b8aedf-1fe3-46f6-a076-3bc1945456ea,
abstract = {{<p>Wetlands play a critical role in the global water cycle and ecological health. However, understanding and accurately parameterizing water and heat exchanges in these areas is challenging due to the heterogeneous nature of wetland surfaces. This study extends the maximum entropy production (MEP) model to simulate wetland evapotranspiration (ET) by integrating NDVI dynamics, water evaporation, and plant transpiration. The model was applied to six FLUXNET wetland sites, with observation periods ranging from 1 to 8 years at daily temporal resolution. Across all time scales, the Nash-Sutcliffe Efficiency (NSE) improved by up to + 0.30 compared to the original model, with NSE values ranging from 0.67 to 0.83 and coefficients of determination (R2) between 0.80 and 0.97. Additionally, the root mean square error (RMSE) was reduced to as low as 7.61 W·m<sup>−2</sup> for monthly estimates. These results demonstrate the enhanced model's ability to capture the temporal dynamics of ET more accurately. Furthermore, the analysis revealed significant correlations between wetland vegetation, the Bowen ratio, and the transpiration-to-evapotranspiration ratio. To further quantify the contributions of multiple environmental factors to wetland energy flux distribution, a structural equation modeling (SEM) approach was employed. The SEM analysis confirms that vegetation exerts an important influence, while other variables such as temperature and radiation also significantly affect the partitioning between latent and sensible heat fluxes. These findings improve the understanding of wetland ET dynamics and provide a foundation for further exploration of the relationship between NDVI and energy flux partitioning.</p>}},
author = {{Sun, Huaiwei and Fu, Leying and Wang, Weiguang and Xue, Jie and Wang, Jing and Liao, Weihong and Li, Haicheng and Sun, Xunlai and Yang, Yong and Wang, Jingfeng and Zhang, Hong and Chen, Fulong and Zheng, Qiming and Meng, Changqing and Zhang, Wenxin}},
issn = {{0022-1694}},
keywords = {{Energy flux partitioning; Evapotranspiration; Maximum entropy production; Wetland}},
language = {{eng}},
publisher = {{Elsevier}},
series = {{Journal of Hydrology}},
title = {{Explore the relationship between Bowen ratio and evapotranspiration in wetlands using the maximum entropy production model}},
url = {{http://dx.doi.org/10.1016/j.jhydrol.2025.133586}},
doi = {{10.1016/j.jhydrol.2025.133586}},
volume = {{661}},
year = {{2025}},
}