Among many Asian countries, Pakistan is also facing an increase in overweight/obesity rate as an epidemic. This research explores how various built environment, personal, religious, cultural, and socioeconomic factors work differently for both sexes to contribute to increasing the Body Mass Index of female adults in all three Socioeconomic Statuses of the megacity Karachi, Pakistan. 210 random samples of 20 years or older male and female adults were selected, and the sample data, from survey questionnaires, was analyzed through Ordinary Least Square Regression in SPSS V.25. The results confirm that low-density neighborhoods, higher fast-food restaurant density, accessibility (walking distance in no. of min) to parks/playgrounds, household size, obesity running in the family, health condition of an individual, and discouraging behavior by others are positively associated with higher BMI. Whereas less junk food consumption in a week and availability of pleasant surroundings where people prefer walking are negatively associated with higher BMI. Although religious factors were not found connected to higher BMI, cultural, built environment, and religious factors combined were responsible for the choice of Physical Activity in both sexes. Walking was the most common whereas cycling was the least common PA in females in all SES.

Wetlands monitoring is essential for sustainable watershed biodiversity conservation, and climate resilience. This study performs a spatiotemporal wetland change analysis for the Sangamon River Watershed, Illinois between 2000 and 2025, using Landsat 5 Thematic Mapper (TM), Sentinel–2 Surface Reflectance (SR), Synthetic Aperture Radar (SAR), Gray–Level Co–occurrence Matrix (GLCM) and terrain data through cloud–based processing in Google Earth Engine (GEE), Google Colab and ArcGIS Pro 3.6.0. We conducted a comparative assessment of deep learning (Deep Neural Network [DNN], U-Net, Attention U-Net, and SegFormer), and machine learning models (Random Forest [RF], Gradient Tree Boosting [GTB], and Support Vector Machines [SVM]) through pixel–based and object–based methods. National Land Cover Database (NLCD) was used for training and validation using stratified random sampling for five categories namely wetlands, forest, agriculture/grassland/barren land, urban/developed and water. A proportion of 54.85% (860.68 km²) of wetlands extent was lost to other land uses, particularly agriculture, urban and forest, along with 46.51% (694.88 km²) forest and 36.80% (66.13 km²) water bodies loss. Agriculture/grassland/barren and urban/developed witnessed increases of 8.56% (820.62 km²) and 72.92% (799.52 km²), respectively. For Landsat–based and Sentinel–based classifications, SegFormer outperformed all ML and DL classifiers with (OA = 94%, Kappa = 0.89, mean F1 = 0.80, mean IoU = 0.70 and OA = 96%, Kappa = 0.92, mean F1 = 0.95, mean IoU = 0.73, respectively) with excellent wetland delineation (PA = 0.99, UA = 0.97, F1 = 0.98, IoU = 0.97 and PA = 0.99, UA = 0.99, F1 = 0.98, IoU = 0.99, respectively). Sentinel–based classifications had improved performance than Landsat, while object–based models consistently outperformed pixel–based methods. The Digital Elevation Model (DEM) and slope were the most influential predictors for RF models, while GLCM and SAR produced negligible influence. The integrated and comparative GeoAI framework provides a robust methodology for watershed–scale wetland monitoring and supports evidence–based conservation, restoration prioritization, climate resilience, and sustainable land–use planning, while offering strong potential for application in other agricultural watersheds following regional validation.

Wetlands monitoring is essential for sustainable watershed and environmental management. This study performs a spatiotemporal wetland change analysis for the Sangamon River Watershed, Illinois between 2000–2025, using Landsat5 TM, Sentinel-2 SR, SAR, GLCM and terrain data through cloud-based processing in GEE, Google Colab and ArcGIS Pro 3.6. We conducted a comparative assessment of DL (DNN, U-Net), and ML models (RF, GTB, and SVM) through pixel-based and object-based methods. NLCD datasets were used for training and validation using stratified random sampling for five categories namely wetlands, forest, agriculture/grassland/barren land, urban/developed and water. 54.85% (860.68 km²) wetlands extent was lost to other land uses particularly agriculture, urban and forest, along with 46.5% (694.88 km²) forest and 36.80% (66.13 km²) water bodies loss. Agriculture/grassland/barren and urban/developed witnessed an increase of 8.56% (820.62 km²) and 72.92% (799.52 km²) respectively. For Landsat-based classification, the object-based GTB model achieved the highest accuracy (OA=91%, Kappa=0.88), with improved wetland delineation (PA=0.99, UA=0.92, F1=0.95). For Sentinel-based classification, U-Net outperformed all ML classifiers with (OA=90%, Kappa =0.89), while object-based GTB, SVM, and RF produced comparable results. Sentinel-based classifications had improved performance than Landsat, while object-based models consistently outperformed pixel-based methods. DEM and slope were the most influential predictors for RF models. The GeoAI framework offers a scalable and transferable approach for watershed-scale wetland monitoring, supporting evidence-based decision-making, restoration prioritization, climate resilience, and sustainable land-use planning in agricultural landscapes.

Introduction: Coastal shorelines are experiencing increased vulnerability from sea-level rise, erosion, and stronger storm activity, highlighting the critical need for adaptive and evidence-based shoreline management approaches that evaluate the site suitability of living shoreline interventions. Traditional hard-armoring solutions, such as seawalls and bulkheads, while protective, often degrade habitats and disrupt natural coastal processes. This study advances a standardized, GIS-based framework for shoreline suitability modeling to support the adoption of living shorelines (LS) and hybrid solutions (HS) as nature-based alternatives. Methods: Focusing on Aberdeen Proving Ground (APG), Maryland, a low-lying coastal military installation along the Chesapeake Bay, we used ArcGIS Pro’s Suitability Modeler (SM) to integrate thirteen physical and ecological variables derived from site assessments and prior studies. The model classified shoreline segments as suitable for LS, HS, or not suitable for living shoreline (NLS) using both weighted and unweighted multi-criteria approaches. Results: Results indicate that the weighted SM classified 30.4% of the shoreline as suitable for LS and 69.5% as suitable for HS, while the unweighted SM increased LS suitability to 33.5% and identified 66.4% as HS, with <1% categorized as NLS in both cases. Weighting increased HS classification by 3.1%, whereas it decreased LS classification by 3.1%. A three-step validation using a confusion matrix, the Living Shoreline Feasibility Model (LSFM), and sensitivity analysis was performed. The weighted model demonstrated stronger agreement beyond chance (Cohen’s Kappa = 0.71) compared to the unweighted model (0.53), indicating improved classification consistency. Introducing weight also improved alignment with LSFM classifications and enhanced differentiation among suitability categories. Sensitivity analysis indicates that classification outcomes are robust to reasonable variations in weights, supporting confidence in the observed LS/HS pattern. Discussion: This study addresses major gaps in prior modeling efforts, specifically the lack of variable standardization, weighting transparency, and multi-step model validation, by offering a transferable, replicable, and data-driven framework. The proposed approach strengthens decision-support capabilities for coastal planners, providing a scientifically robust tool for scaling nature-based shoreline and hybrid protection and advancing coastal resilience.

A correction refers to a change to their article that the author wishes to publish after publication. The publication of this article is subject to Frontiers' editorial approval.• Please read through all the templates before choosing • Pick the most relevant text template(s) from the following page and delete all others.• Edit the text as necessary, ensuring that the original incorrect text is included for the record, please see the below. • Please do not use any extra formatting when editing the templates, and only modify the red text unless absolutely necessary • Submit to Frontiers following the instructions on this page.When the original text contained incorrect information, to preserve the scientific record, please include that text when editing the below templates. For example:There was a mistake in the Funding statement, an incorrect number was used.The correct number is "2015C03Bd051.". The publisher apologizes for this mistake.The original version of this article has been updated.