Publisher：Agricultural Water Management  

Accurate and precise estimation of evapotranspiration (ET) is crucial for understanding the terrestrial carbon, water, and energy cycles. While process-based models of ET, such as the Penman–Monteith model offer robust generalization capabilities, they are limited by the need for detailed parameters (e.g., stomatal conductance,) that are challenging to measure continuously. On the other hand, machine learning models can estimate ET by capturing relationships between ET and environmental variables without experimentally measuring model parameters. However, machine learning models face the challenge of limited generalizability. This issue is particularly significant given the uncertainty introduced by changing climatic conditions, which can restrict the model's predictive performance when it is applied to different environmental contexts. Therefore, we propose a hybrid modeling approach that combines feature engineering using process-based models with machine learning to improve generalizability while maintaining practicality. Our model first converts environmental variables into leaf-scale ET using mechanistic process-based models and then uses these features along with the leaf area index to estimate the canopy-scale ET using an artificial neural network (ANN). We evaluated the generalization of the hybrid model against a pure ANN model using FLUXNET2015 data. Results show that the hybrid model significantly outperformed the pure ANN model, especially when tested on data beyond the range of the training dataset. Furthermore, the estimation accuracy of the hybrid model was stable even when the values of the model parameters in the process-based models used for feature engineering were varied by ±50 %. This indicates that incorporating a mechanistic understanding of plant environmental responses enhances the generalizability and robustness of ET predictions. These findings underscore the potential of hybrid models to combine the strengths of process-based and machine learning approaches.

DOI： 10.1016/j.agwat.2025.109854

Web of Science

Scopus

Publisher：Plant and Soil  

Background and aims: In humid regions, the transpiration rate is determined by transpiration demand because of the sufficiently moist soil. However, inhibition of plant water uptake capacity due to soil waterlogging can significantly constrain the transpiration rate even after drainage. This study aimed to evaluate plant hydraulic conductance during soil waterlogging and subsequent reoxygenation and its impact on whole plant transpiration. Methods: Two experiments were conducted to assess the ecophysiological responses of soybeans during waterlogging (Experiment 1) and reoxygenation (Experiment 2). Transpiration rate, stomatal conductance, leaf water potential, and leaf area were measured. In addition, plant hydraulic conductance was calculated using the root water uptake equation. A simple transpiration model incorporating the response of plant hydraulic conductance to waterlogging was used to evaluate the impact of waterlogging on transpiration estimation. Results: Waterlogging for more than 3 days reduced plant hydraulic conductance, which persisted even during the post-waterlogging reoxygenation period. Furthermore, leaf water potential, stomatal conductance, and transpiration rate in waterlogging treatment exhibited a lower value than those in control during both waterlogging and reoxygenation. The constructed model effectively reproduced the responses of plant hydraulic conductance and transpiration rate, especially during reoxygenation. Conclusion: Soil waterlogging significantly reduce the hydraulic conductance of soybean plants, resulting in leaf water stress and depression of transpiration, even during reoxygenation. Our results highlight the importance of integrating plant hydraulic responses with water dynamics models in the soil-plant-atmosphere system.

DOI： 10.1007/s11104-024-07040-8

Web of Science

Scopus

Language：English   Publisher：The Society of Agricultural Meteorology of Japan  

<p>Soybean production in northern Kyushu (Saga and Fukuoka Prefectures), Japan, has declined since 2008. As major cultivar and cultivation management have hardly changed in recent years, this decrease should be caused by climate variability. Thus, we evaluated the variability of the recent climate in northern Kyushu and its impact on soybean production. Trend analyses and the Lepage tests were conducted on air temperature, solar radiation, and precipitation over the past 30 years. The correlations between these meteorological conditions and soybean yield were assessed. Solar radiation from late August to early September, corresponding to the soybean flowering stage, declined since 2008 and showed a strong positive correlation with yield, suggesting a critical limiting factor. Shifting the flowering stage through early or late sowing could mitigate this effect. Precipitation during the growing season was negatively correlated with soybean yield, indicating that improvements in drainage management are effective. Interestingly, soybean yields remained high under low precipitation conditions, showing that soil drought did not significantly limit production. These findings highlight the need for adaptive strategies, including adjustments of sowing date and improved water management, to enhance soybean production in northern Kyushu.</p>

DOI： 10.2480/agrmet.d-24-00053

Web of Science

Scopus

CiNii Research

Publisher：Paddy and Water Environment  

Efficient irrigation and drainage management are highly required for increasing crop productivity in paddy rice and upland crop rotation. However, conventional management does not sufficiently consider the water status of the plants and soil in the root zone. The aim of this study was to evaluate whether the hydraulic resistance of soil (R<inf>s</inf>) or plant (R<inf>p</inf>) principally controlled transpiration in rotational paddy fields (RPFs) located in humid regions. To achieve this, we conducted field measurements of soil water conditions, evapotranspiration rate, and leaf water potential in RPFs cropped with soybean after the flowering stage and calculated R<inf>s</inf> and R<inf>p</inf> based on the theory of root water uptake. After the flowering stage, the soil was sometimes saturated owing to intermittent precipitation, and thus R<inf>s</inf> was maintained at a low value. By contrast, R<inf>p</inf> gradually increased over time and ranged between 5.1 × 10⁸ and 10.3 × 10⁸ s, which was one to three orders of magnitude higher than R<inf>s</inf>. The ratio of the actual to the potential transpiration rate decreased throughout the investigation period and hardly reached 1.0, even though the soil was sufficiently wet. These results indicate that R<inf>p</inf>, which probably increases with continuous soil saturation, controls crop transpiration in RPFs under humid climates. Our results suggest that drainage systems are essential in RPFs to avoid a change in R<inf>p</inf> and improve crop productivity.

DOI： 10.1007/s10333-022-00923-5

Scopus

Publisher：Paddy and Water Environment  

The copyright holder for this article was incorrectly given as © The International Society of Paddy and Water Environment Engineering 2023 but should have been © The Author(s) 2023. The original article has been corrected.

DOI： 10.1007/s10333-023-00929-7

Scopus