Ming-Che Hu

Accurate flood mapping is essential for rapid impact assessment in agricultural landscapes, where inundation can interrupt production, damage rural infrastructure, and complicate recovery. Recent research has shown that deep learning can support flood delineation from remotely sensed data, but the literature still reports recurring weaknesses in robustness assessment, uncertainty interpretation, and transparent benchmarking against simpler machine learning baselines. In this study, we developed an event aware flood mapping framework for the Arkansas 2025 flood event by integrating terrain, antecedent water conditions, rainfall dynamics, and hydrologic proximity within a spatial learning workflow. We evaluated progressively richer predictor configurations using a U Net based convolutional neural network and assessed robustness through repeated seed experiments under fixed train, validation, and test splits. A Random Forest benchmark was introduced to provide a transparent classical machine learning reference. Mean probability, uncertainty, and agreement maps were also generated to characterize spatial confidence patterns. The results showed that adding rainfall substantially improved performance relative to the more complex advanced deep learning configuration, increasing mean best validation IoU by 7.3%, mean test IoU by 15.8%, and mean test F1 by 12.4%, while also reducing the variability of test IoU, test F1, and threshold selection. The rainfall based U Net achieved a best validation IoU of 0.278 ± 0.015, a test IoU of 0.264 ± 0.014, and a test F1 of 0.417 ± 0.018. The more complex configuration that added distance to river and Focal Tversky optimization did not deliver a consistent gain, with a best validation IoU of 0.259 ± 0.011, a test IoU of 0.228 ± 0.033, and a test F1 of 0.371 ± 0.044. The Random Forest benchmark was unexpectedly strong, reaching a validation IoU of 0.608 at the selected threshold and a test IoU of 0.351 with a test F1 of 0.519, equivalent to gains of 33.0% in test IoU and 24.5% in test F1 relative to the best deep learning configuration. Feature importance analysis identified elevation, slope, rainfall peak, and event total rainfall as dominant contributors. Spatial uncertainty was concentrated along transitional flood margins, while agreement maps highlighted coherent high confidence cores. Cropland overlay analysis further demonstrated that the framework can support first pass agricultural exposure assessment. Rather than promoting model complexity for its own sake, this study shows that event aware predictors, robustness oriented evaluation, and explicit benchmark comparison are central to credible geospatial flood intelligence in agricultural environments.

This study combines the Nash–Cournot competition model and the stochastic optimization model to examine the impact of electric vehicle (EV) quantity fluctuations on microgrid operations, aiming to optimize energy usage in a competitive electricity market. Integrating distributed energy resources and bidirectional charging, microgrids offer a novel approach for energy optimization, aiding in renewable energy generation, peak demand management, and emission reduction. Empirical evidence highlights benefits in Taiwan’s electricity market and net-zero emissions target by 2050, with a case study demonstrating enhanced local renewable energy generation due to EVs and microgrid integration. As the number of EVs increases, electricity sales from microgrids decline, but electricity purchases remain stable. The degree of electricity liberalization also influences the supply and demand dynamics of the electricity market. Microgrids selling electricity only to the main grid increases total power consumption by 65.55 million MWh, reducing the market share of the state-owned utility (Taipower). Conversely, allowing retailers to purchase from microgrids increases total consumption by 30.87 million MWh with a slight market share decrease for Taipower. This study contributes to providing an adaptable and flexible general model for future studies to modify and expand based on different scenarios and variables to shape energy and environmental policies.

This study aims to elucidate the dynamics of fluid topology during drainage and evaporation of wetting and non-wetting fluids via evaluation of their Euler characteristic (EC) and Betti numbers from designed experiments. The results showed that the EC of the non-wetting fluid remained constant at a high wetting fluid saturation and then decreased nonlinearly. However, the EC of the wetting fluid exhibited more complex trends. In the 2-D micromodel, the number of loops decreased monotonously, reached zero, and remained constant for lower saturations, while the number of distinct components fluctuated because of the appearance and disappearance of disconnected clusters. In the 2-D model, the number of distinct components played a significant role; in the 3-D X-ray CT experiment, the number of loops of pendular rings dominated the EC. The findings suggest that the trends observed in the 2-D micromodel may not apply under 3-D conditions.

Climate change might potentially cause extreme weather events to become more frequent and intense. It could also enhance water scarcity and reduce food security. More efficient water management techniques are thus required to ensure a stable food supply and quality. Maintaining proper soil water content and soil temperature is necessary for efficient water management in agricultural practices. The usage of water and fertilizers can be significantly improved with a precise water content prediction tool. In this study, we proposed a new framework that combines weather forecast data, numerical models, and machine learning methods to simulate and predict the soil temperature and volumetric water content in a greenhouse. To test the framework, we performed greenhouse experiments with cherry tomatoes. The numerical models and machine learning methods we selected were Newton’s law of cooling, HYDRUS-1D, the random forest model, and the ICON (inferring connections of networks) model. The measured air temperature, soil temperature, and volumetric water content during the cultivation period were used for model calibration and validation. We compared the performances of the models for soil temperature and volumetric water content predictions. The results showed that the random forest model performed a more accurate prediction than other methods under the limited information provided from greenhouse experiments. This approach provides a framework that can potentially learn best water management practices from experienced farmers and provide intelligent information for smart greenhouse management.

The water–energy–food (WEF) nexus attracts much attention due to the elevated public concern regarding environmental conservation and sustainability. As we head into a new era of civilization, population increase and modernized lifestyles have led to an increasing need for water, energy, and food. However, severe hydrological precipitation significantly impacts agricultural harvest, and such influence becomes more apparent under the influence of climate change. Meanwhile, the major method of electricity generation (i.e., fossil fuel burning) has a negative impact on the environment. These inevitable threats are crucial and have to be dealt with for a society on the road towards sustainability. In the present study, an integrated evaluation of the WEF nexus was conducted for two areas with different levels of urbanization using empirical multiple linear regression in a simultaneous equation model (SEM). By incorporating the collected data into the SEM, the weighting coefficient of each identified variable was obtained, and the nexus implication was assessed in model simulation at different scenarios considering the population growth, agro-technology advancement, energy structure improvement, and available water resources. In the simulated results, three observations were found: (1) the rural area is more sustainable than the urban one; (2) the sustainability for both the investigated areas is significantly subject to their water supply and demand; and (3) food production was found to have a less important effect on the sustainable development of the urban area. This study identified the key factors in the WEF nexus exploration, which are economically and environmentally important for resource allocation. An empirical model was developed to correlate sustainable achievement with WEF management, as well as strategic policies that should be implemented under the pressure of urbanization.

Urban metabolism analyzes the supply and consumption of nutrition, material, energy, and other resources within cities. Food, water, and energy are critical resources for the human society and have complicated cooperative/competitive influences on each other. The management of interactive resources is essential for supply chain analysis. This research analyzes the food-water-energy system of urban metabolism for sustainable resources management. A system dynamics model is established to investigate the urban metabolism of food, water, and energy resources. This study conducts a case study of Shihmen Reservoir system, hydropower generation, paddy rice irrigation of Taoyuan and Shihmen Irrigation Associations, and water consumption in Taoyuan, New Taipei, and Hsinchu cities. The interactive influence of the food-water-energy nexus is quantified in this study; the uncertainty analysis of food, water, and energy nexus is presented.

Dengue fever is a vector-borne infectious disease that is transmitted by contact between vector mosquitoes and susceptible hosts. The literature has addressed the issue on quantifying the effect of individual mobility on dengue transmission. However, there are methodological concerns in the spatial regression model configuration for examining the effect of intercity-scale human mobility on dengue diffusion. The purposes of the study are to investigate the influence of neighborhood structures on intercity epidemic progression from pre-epidemic to epidemic periods and to compare definitions of different neighborhood structures for interpreting the spread of dengue epidemics. We proposed a framework for assessing the effect of model configurations on dengue incidence in 2014 and 2015, which were the most severe outbreaks in 70 years in Taiwan. Compared with the conventional model configuration in spatial regression analysis, our proposed model used a radiation model, which reflects population flow between townships, as a spatial weight to capture the structure of human mobility. The results of our model demonstrate better model fitting performance, indicating that the structure of human mobility has better explanatory power in dengue diffusion than the geometric structure of administration boundaries and geographic distance between centroids of cities. We also identified spatial–temporal hierarchy of dengue diffusion: dengue incidence would be influenced by its immediate neighboring townships during pre-epidemic and epidemic periods, and also with more distant neighbors (based on mobility) in pre-epidemic periods. Our findings suggest that the structure of population mobility could more reasonably capture urban-to-urban interactions, which implies that the hub cities could be a “bridge” for large-scale transmission and make townships that immediately connect to hub cities more vulnerable to dengue epidemics.