The large-scale issues of mining and deforestation in the Brazilian Amazon Rainforest have emerged as critical global concerns, particularly due to their impacts on weather patterns and surface temperatures. This study aims to investigate the correlations among mining activities, deforestation area, local and regional surface temperatures and precipitation patterns. We employed the Earthrise Enterprise mining detector model, a lightweight convolutional neural network (CNN), which was trained using hand-labeled data from Sentinel-2 satellite imagery spanning the years 2017 to 2023. This trained model effectively differentiates mining areas from other topographical features with an accuracy of 99.11% at a threshold value of 0.9. Other numerical climate data for the Brazilian Amazonas region from 2017 to 2023 were extracted, including: 1) surface temperature, 2) deforestation area, and 3) precipitation. These data were sourced from Hansan Global Land Discovery and Analysis and ECMWF Reanalysis v5 (ERA5) imagery, utilizing Google Earth Engine for processing. We employed the Pearson correlation analysis to examine the relationships among these variables. Our findings indicate a strong positive correlation between mining and deforestation areas (r = 0.81), a weak correlation between deforestation areas and Earth’s surface temperature (r = 0.27), and a negative correlation between surface temperature and precipitation amount (r = -0.52). This study demonstrates that CNNs can be effective tools for analyzing satellite images to explore mining activities. The correlation between mining and deforestation illustrates the role of mining plays in the deforestation of the Amazon Rainforest. The weak correlation between deforestation and temperature indicates that surface temperature changes may be dynamic and multifactorial. Finally, the negative correlation between surface temperature and precipitation suggests that surface temperature increase can be an important drought-predicting factor. We will also discuss the limitations of this study due to the limited number of data points used for the correlation study, which impacts the strength and accuracy of correlations.

Land management and stewardship teams continue to lack the tools to capture 3D spatiotemporal insights of the ecosystems they oversee. For wildfire management at the wildland-urban interface, teams face challenges in capturing vegetation growth over time after a fuel reduction program and connecting seasonal changes to the vegetation distribution across the treated area. Current approaches rely on triangle meshes or point clouds generated from photogrammetry or LiDAR surveys on drones or hiked traverses. However, the difficulties in optimizing these meshes lead to large triangles that inadequately approximate the bulk vegetation shape, and the point cloud data is often too sparse for local plant-scale understanding. To address this gap, we extend recent machine learning-based computer graphics techniques in 3D Gaussian Splatting (3DGS) to reconstruct scenes at the wildland-urban interface from handheld imagery with sufficient detail to identify species, capture individual leaves, recover plant stature, and disambiguate overhanging plant individuals. We develop a new method to match the 3DGS reconstruction of these scenes across months, associating plant growth across seasons interactively in 3D. To achieve the centimeter-level matching, we adapt the Umeyama algorithm and the iterative closest point algorithm from point cloud maps to the 3DGS scene, leveraging the probabilistic interpretation of the 3D Gaussian data structure and robustly handling visual and geometric changes associated with vegetation phenology over time. We have applied our method to recent pile burns at Stanford’s Jasper Ridge ’Ootchamin ’Ooyakma Biological Preserve at monthly intervals. We demonstrate differences in ecological response where some piles featured the unexpected return of a rare and threatened bushmallow, and others remained more barren. This pile burn microcosm implicates the need for plant-level 3D spatiotemporal models to understand ecosystem recovery to fire mitigation practices.Please visit the project page for the spatiotemporal alignment video and more information: https://danineamati.github.io/burn-ecorecovery.github.io/

Coastal surface currents influence processes like pollutant transport, sediment resuspension, and navigation safety, presenting the need for high-resolution monitoring tools for informed decision-making. We present a remote sensing approach based on unmanned aerial systems (UAS) that estimates surface currents using Doppler analysis of video collected from a continuous UAS flight transect over Freeport Harbor, Texas. This work extends the open-source MATLAB software CopterCurrents, originally developed for UAS videos collected by hovering at fixed station points (Streßer et al., 2017), to linear flights in which the continuous videos are spatially segmented into equivalent hovering videos. The segmentation is achieved by tracking and analyzing the motion of subwindows (240×240 pixels each) over the fixed ocean surface, each representing approximately 10×10 m2. Instead of relying on a fixed hover, each subwindow is tracked as it moves through the camera’s field of view during flight. For subwindows with sufficient temporal overlap (approximately 30 s duration, which is greater than 248 frames in this study), the mean 2D surface velocity field is extracted using three-dimensional fast Fourier transform and Doppler fitting to the linear, deepwater dispersion relation for surface waves (Streßer et al., 2017). Subwindow positions are geo-referenced in UTM coordinates based on UAS flight logs. To remove spurious current vectors, a signal-to-noise ratio threshold was applied to filter out noisy data, along with an upper velocity limit.We apply this method to a 10-minute UAS mission over Freeport Harbor, covering approximately 900 m and producing over 1000 geo-located subwindow measurements. The final output is a spatially continuous current map overlaid on satellite imagery, visualized through a vector field and a current-speed heatmap. This approach enables high-resolution current estimation without the need for hovering, thereby supporting efficient deployments for estuarine dynamics, rapid-response coastal monitoring, and operational nearshore oceanography.ReferencesStreßer, M., Carrasco, R., & Horstmann, J. (2017). Video-Based Estimation of Surface Currents Using a Low-Cost Quadcopter. IEEE Geoscience and Remote Sensing Letters, 14(11), 2027–2031. https://doi.org/10.1109/LGRS.2017.2749120

Water scarcity forces mid-hill households in Nepal to migrate, weakening climate resilience; thus, local adaptation is crucial to combat water stress in the hilly region. Rainwater harvesting (RWH) presents a promising approach to addressing this issue, owing to its cost-effectiveness, self-sufficiency, and strong government support. Among the components of an RWH system, the tank is widely considered the most critical component. This study investigates the reliability and water-saving efficiency (WSE) of RWH at 29 locations in Nepal, considering various climatic scenarios. The average roof area at each location was extracted using Google Earth imagery, and to account for future rooftop growth, the projected values for 2020-2050 were considered. The NEX-GDDP-CMIP6 dataset was used to generate daily rainfall data for the period 2024-2050. Missing values in the historical dataset (1993-2023) were imputed using statistical and machine learning techniques. The performance of RWH was modeled using the Yield After Spillage algorithm. Our investigation revealed that, when using historical data, reliability and WSE of RWH systems range from 43.73% to 87.51%. When projected rainfall was used, performance declined, with reliability and WSE ranging from 40.04% to 82.18% for the SSP585 and SSP245 scenarios, implying that future climate scenarios will negatively impact RWH performance when there is no change in rooftop area. However, when future rooftop area projections were also considered, performance differences gradually reduced over the decades, potentially indicating that increased rooftop areas may offset future climate change-induced performance reductions. Moreover, the difference between the highest and lowest reliability and WSE was generally smallest for 0.5 m3 tanks, increased for intermediate sizes, and decreased again for 8 m3 tanks and larger. This trend suggests that, on average, RWH performance is most consistent for the smallest and largest tank sizes. However, while smaller tanks show smaller differences, larger tanks not only offer similar consistency but also demonstrate higher overall performance. Thus, an RWH system with a larger storage size is expected to be more resilient to future climate scenarios.

The Amazon Basin, one of the most biodiverse regions on Earth, is shaped by complex interactions between climate, hydrology, and forest dynamics. It’s strongly influenced by seasonal flooding, soil type, and forest succession. Deforestation has dominated discussions of land use change in recent decades; However, increasing attention is being given to forest recovery and passive restoration processes. Understanding where and how forests recover is critical for guiding conservation policy and monitoring long-term ecosystem resilience, an approach aligned with the Mosaico do Baixo Rio Negro’s regional conservation framework. This study uses Landsat and Planetscope imagery to explore patterns of vegetation and surface moisture change from 1990 to 2024 along the Amazon River near Manaus, Brazil. Spectral indices derived from Landsat imagery were used to assess vegetation conditions and surface moisture across ten representative assessment sites, establishing baseline environmental patterns. These metrics were then applied across four broader restoration mosaics to evaluate landscape-scale vegetation recovery and ecosystem change. Spatial structure, including fragmentation and edge effects, were also analyzed. This data was combined to identify trends in forest succession, detect areas of passive regeneration, and evaluate ecological stability. PlanetScope imagery supported visual interpretation of finer-scale landscape changes. Trends in spectral index values were extracted, mapped, and categorized across all assessment and restoration zones to visualize temporal and spatial variability over the 34-year period. These patterns provide a quantitative foundation for analyzing vegetation structure, surface moisture dynamics, and regeneration processes across tropical forest systems, supporting long-term monitoring and conservation planning.

The rapid and unexpected nature of flash floods leaves communities with minimal time to respond. Accurate forecasting of such events is crucial, as it allows for timely warnings, early evacuations, and preparations, significantly reducing potential losses. This study applies a transformer-based Deep Learning (DL) with Remote Sensing (RS) validation to predict floods with at different lead times. The Guadalupe River basin in Texas was selected as a case study due to the 2025 catastrophic flash flood, one of the deadliest floods in U.S. history. For forecasting purposes, discharge (Q) data from a downstream USGS station were chosen as the target variables. Precipitation, air temperature and moisture and vegetation indices were extracted for upstream station from RS sources. These RS products were validated against corresponding ground-based observations to ensure quality as dynamic inputs for flood prediction. Input climatic variables were structured based on different times windows (t-1, t-3, t-5, t-7, t-15, and t-30), representing conditions at 1, 3, 5, 7, 15 and 30 days before the current downstream state (t). different input sequences were utilized in a Multivariate Predictive Transformer Network, a DL model designed to make predictions from a temporal multivariate sequential dataset. The flood events were identified by exceedance from the 90th, 95th and 99th percentile threshold for Q within the 2003 to 2025 period. The predictive performance of the model was assessed regarding both the accuracy in predicting the flooding events and timing of impact. Furthermore, the model's capability to forecast floods with annual exceedance probabilities of 2, 5, 10, 25, 50, and 100 was thoroughly evaluated. Prediction results show that developed model can accurately detect the flood occurrence up to 5 days notice (Accuracy: 0.96). While the accuracy drops in next time lags, the metric shows comparable results with similar studies. The outcomes of this study, which integrates advanced transformer-based predictive modeling with remote sensing data, can substantially improve early forecasting and management strategies for floods as a detrimental natural hazard phenomenon.

FjordPhyto is a polar participatory science program committed to public engagement in scientific discovery from the world’s most remote data-limited regions. Launched in 2016, FjordPhyto partners with expedition cruise vessels for gathering environmental data in Antarctica, one of the fastest-warming regions in the world. We have successfully involved more than 10,000 travelers to date; however, our engagement with individuals outside of travel to the poles has remained limited. To expand our reach, from 2023 to present, FjordPhyto partnered with two NASA Science Activation Teams: Infiniscope (through our POLARIS and InfiniteOCEAN initiatives) and OCEANOS (through our three-year PRO OCEANOS initiative in Puerto Rico (PR)). These collaborations significantly broadened FjordPhyto’s outreach to younger learners, as well as to English- and Spanish-speaking audiences via in-person activities and online engagement platforms. We developed hands-on workshops, lecture materials and Augmented and Virtual Reality (AR/VR) assets of 3D phytoplankton and Antarctic environments for in-person engagement reaching 60 students in Puerto Rico. We harnessed Infiniscope’s digital learning platform to develop three interactive ocean theme modules, virtual field trips, and one training course to effectively train 59 tour expedition guides who facilitate FjordPhyto in Antarctica. These efforts provided invaluable opportunities for networking and enhanced proficiency in program, time, and people management, fostering critical leadership skills for Early Career Researchers as Principal Investigators. Our success comes through collaboration with experts in communication and education who provided space and funding for us to expand. Through SCoPE collaborations, we have significantly expanded our reach, engaging more individuals in polar science.

A critical gap in understanding the global methane (CH4) budget is the quantification of emissions from small artificial water bodies. One mechanism that may be driving increased CH4 emissions is an increase in natural emissions from inland aquatic ecosystems, such as small water bodies (<0.1 km²). The number, location, and seasonal extent dynamics of small water bodies—particularly artificial (or man-made) water bodies (i.e., irrigation or stormwater management ponds)—are still relatively unknown, despite their critical importance in CH4 emissions. Data from satellite imagery is a promising way to map the presence and areal extent of small water bodies, as they acquire data systematically over space and time. However, most prior research has used medium-resolution satellite imagery (30 m), which has been shown to underestimate the area of small water bodies. In this study, we employed machine learning (random forest) and deep learning (U-Net) to classify small water bodies in central and eastern North Carolina, using Sentinel-2 (10 m) and PlanetScope (3 m) optical satellite imagery. We then compared these classifications to a Landsat 30 m surface water product. We found that each resolution detected seasonal changes in surface water extent; however, the higher resolution imagery (i.e., PlanetScope) detected larger areas for the same water bodies. We also found that the higher-resolution imagery detected more small water bodies than the moderate-resolution imagery. Finally, we found that the higher resolution surface water extents had a stronger correlation with field-sampled CH4 concentrations from small artificial water bodies (R² 0.68) than the coarser optical imagery (R² 0.16). We conclude that higher resolution imagery can 1) identify more small water bodies than moderate resolution imagery; and 2) be used to provide more accurate estimations of local, regional, and global methane emissions. Such estimates can help countries better plan to meet their Global Methane Pledge.

Accurate hydrological simulation of the Laurentian Great Lakes-the world’s largest surface freshwater system-is critical for effective water resource management, yet remains challenging due to the basin’s vast size, international boundaries, and data heterogeneity. While deep learning models such as long-short-term memory (LSTM) networks have shown promise in hydrological forecasting, a unified, data-driven model for daily streamflow prediction across the entire Great Lakes basin has not been realized. This study addresses this gap by developing and evaluating an Entity-Aware LSTM (EA-LSTM) model, trained on a newly compiled dataset spanning 975 sub-basins across the U.S. and Canada from 1980 to 2023. The dataset integrates daily meteorological forcings, static catchment attributes, and observed streamflow. The EA-LSTM, using a temporal training split, achieved a median Nash-Sutcliffe Efficiency (NSE) of 0.685, significantly outperforming both a standard LSTM (median NSE 0.567) and the operational NOAA National Water Model (median NSE 0.208). Performance analysis reveals that the model’s predictive skill varies by catchment characteristics, with reduced performance observed in very small and highly urbanized basins. Our findings demonstrate that a single, data-driven model can effectively simulate streamflow across a vast, international watershed without basin-specific calibration, offering a robust and unified tool for transboundary water resource management. This work highlights the potential of advanced deep learning models to enhance hydrological prediction and support transboundary water resources management.

Hydrologic forecasting in large, transboundary basins such as the Laurentian Great Lakes remains a significant challenge due to complex hydrological processes, extensive surface water coverage, and widespread ungauged regions. These factors limit the predictive skill of existing models and constrain effective operational water resource management. This study evaluates and compares two modeling approaches for seasonal runoff prediction: a data-driven Long Short-Term Memory (LSTM) neural network model and the process-based Large Basin Runoff Model (LBRM). Using the Climate Forecast System Reanalysis (CFSR) as climate forcing from 1979 to 2023, both LSTM and LBRM models were developed and tested to simulate runoff across the entire Great Lakes basin. Their seasonal forecasting performance was then assessed using archived 9-month forecasts from the Climate Forecast System (CFS) as input. Additionally, total regional runoff was evaluated from a water balance perspective to assess model effectiveness for seasonal water supply forecasting. This research aims to refine hydrologic modeling frameworks using the Great Lakes as a testbed for seasonal-to-annual water supply forecasting. The results offer insights into the comparative strengths of data-driven and process-based models and highlight their potential integration for improving forecast accuracy. Ultimately, this work supports the development of more reliable forecasting tools for operational water management in the Great Lakes and other large transboundary basins worldwide.

Despite growing recognition of climate change impacts on agricultural systems, changes in water constraints on vegetation productivity across India remain poorly understood. Here we conduct a comprehensive evaluation of changes in relationship between vegetation growth and limitations in water availability across India between 1982 and 2020, analysing multiple vegetation indices (Normalized Difference Vegetation Index [NDVI], Leaf Area Index [LAI], Fraction of Photosynthetically Active Radiation [FPAR]) against multi-timescale drought indicators (Standardized Precipitation-Evapotranspiration Index [SPEI] 1-24 months). We document a substantial increase in vegetation water constraint over this period, with water-deficit areas expanding across the country during the primary cropping season of Kharif. Remarkably divergent seasonal trends emerge for the Rabi season, with less pronounced water stress patterns. The increase in water constraints was coincident with significantly decreasing response times, suggesting susceptibility of vegetation to drought events as early as within four months of water scarcity in recent years. These observed changes were found to be primarily attributable to rising atmospheric CO₂, with distinct seasonal patterns - precipitation effects dominating during Kharif while temperature effects become more prominent during Rabi. Our findings highlight the need for incorporation of water constraint in vegetation monitoring and drought assessment frameworks under a warming climate, with implications for food security across India’s diverse agricultural systems.

Recent evidence points to a relatively young age of the Earth’s inner core. Such findings challenge the classical convection-based geodynamo theory that explains the Earth’s magnetic field generation. The interaction of the liquid core with the topographic features at the core-mantle boundary, however, could modify the dynamics in a way that promotes dynamo action. Topography may facilitate increased heat transfer efficiency, propagation of inertial waves, and, as the focus of this work, modified helicity generation. Helicity, an inviscid invariant like momentum and energy, serves as a topological measure of the intertwining of vorticity and velocity in a fluid flow. It plays a central role in the classical picture of geodynamo action. Here, we aim to quantify helicity using three-dimensional flow fields measured by Lagrangian particle tracking in our newly developed apparatus, ToRoCo (Topographic Rotating Convection). We compare helicity generation in experiments with smooth top and bottom boundaries to those with rough boundaries of various length scales. Together with future measurements in liquid metal experiments, we aim to characterize how topography and helicity interact in planetary flows.

Changing magnetic field configurations experienced by Titan as it orbits Saturn can induce currents in the conductive ionosphere of Titan. These induced currents in turn generate ionospheric magnetic fields that deflect the incident plasma and shield out the external magnetic field. By measuring magnetic field perturbations, we can calculate the current densities necessary to create the observed perturbations (with some restrictive assumptions). We determine horizontal currents in the ionosphere of Titan from magnetic field perturbations during Cassini flybys closer than 1500 km, i.e., within the collisional ionosphere. Data from individual flybys with similar trajectories can show significantly different ionospheric current configurations. By grouping the flybys by external magnetic field and plasma conditions and Titan orbital position, we attempt to determine how ionospheric currents change with respect to external conditions. These currents induced in Titan’s ionosphere are but one way Saturn’s magnetospheric environment impacts ionospheric processes and dynamics at Titan.

Ecosystem recovery is intricately linked to human wellbeing. Human wellbeing in this context is an interdisciplinary perspective on what allows humans to thrive in relation to the natural environment and entails values, physical and psychological health, governance, and social, cultural, and economic wellbeing. In the Puget Sound region in Washington state, ecosystem health and recovery include both biophysical and human wellbeing goals, grounded in a socio-ecological systems model. Over the past 20 years, the Puget Sound has developed an existing set of “Human Wellbeing Vital Signs”, which measure progress towards recovery goals through subjective and objective indicators of wellbeing. The existing Human Wellbeing Vital Signs have informed the integration of health and wellbeing, Tribal and cultural knowledge, and environmental justice into ecosystem recovery projects. Despite investments and advances towards integrating human wellbeing into ecosystem restoration in the Puget Sound, there remain crucial gaps and opportunities. These include unequal expertise, funding, and personnel capacity compared to the natural sciences; gaps between social science research, management decisions, and recovery program implementation; challenges to accurately monitor and report on human wellbeing outcomes; and limited involvement of overburdened and vulnerable communities in planning, monitoring, and implementation. To address this last challenge, the Puget Sound Partnership has engaged with various community-based organizations that represent various overburdened and vulnerable communities across the Puget Sound region – including youth, Black communities, Asian American and Pacific Island communities, and immigrant communities – using community-based participatory methods to better understand (1) resonance of existing Human Wellbeing Vital Signs, (2) additional community dimensions of human wellbeing not reflected by the existing Human Wellbeing Vital Signs, and (3) opportunities to revise the Human Wellbeing Vital Signs to improve its resonance with overburdened and vulnerable communities. This presentation will share the results of this community-based participatory research project and its applications for natural resource management decision-making processes.

Transitions between energy-limited, water-limited, and transitional evapotranspiration regimes are central to understanding land–atmosphere interactions and climate impacts. While satellite-derived soil moisture observations have become widely available, most existing approaches rely on modelled evaporative fraction, temperature-based proxies, or curve-fitting techniques. These methods often introduce model dependencies and limit applicability in data-scarce regions. This study presents a novel, observation-only approach using the Matrix Profile—a computationally efficient and parameter-free time series algorithm that detects recurring patterns and regime shifts with exact similarity measures. We applied this method to post-rainfall dry-down sequences from in-situ soil moisture observations in the Yanco region, Australia, using the open-source STUMPY Python library. Regime transitions were detected through shape-based segmentation using FLUSS (Fast Low-Cost Unipotent Semantic Segmentation), and the resulting motifs were clustered into energy-limited, water-limited, and transitional regimes. These classifications were validated against evaporative fraction–soil moisture phase space derived from eddy covariance flux tower data. The same workflow was extended to SMAP Level-3 and Level-4 soil moisture products. Using Matrix Profile–based semantic segmentation of SMAP time series, we generated large-scale spatial maps of hydrological regimes. Remarkably, this was achieved using only soil moisture observations, without relying on auxiliary energy or temperature data. This model-free approach enables robust regime mapping using only soil moisture time series. It supports applications such as identifying climate-sensitive hotspots, monitoring hydrological extremes like flash droughts and heatwaves, assessing climate impacts, and benchmarking land surface models.

Groundwater depletion in northern India, particularly in the intensively cultivated alluvial aquifers of the Indian Ganga Basin (IGB), poses a growing threat to agricultural sustainability and water security. This work aims to present a predictive assessment of groundwater balance and long-term resource sustainability in the IGB using a high-resolution numerical modelling approach. A four-layer aquifer model was developed at a 500-meter spatial resolution to simulate groundwater dynamics from 1997 to 2023. The model integrates refined estimates of recharge, evapotranspiration, groundwater abstraction, and river-aquifer interactions, and is implemented using MODFLOW 6 with a control-volume finite difference scheme. Uncertainties in the estimation of model parameters and resulting uncertainty in water balance components were quantified using the iterative ensemble simulation approach. The model was applied to assess groundwater use under varying climate and abstraction scenarios. Preliminary results showed historical storage depletion consistent with observed groundwater levels and regional trends identified using GRACE-based remote sensing approaches. Areas of significant storage depletion also overlapped with districts that have high estimated groundwater extraction. They also highlight spatial and temporal patterns of groundwater stress, identify critical zones of imbalance, and demonstrate the model’s utility in evaluating policy interventions. This modelling framework provides a robust tool for guiding sustainable groundwater management strategies at regional and sub-basin scales.

Rapid and accurate estimation of magnitude is crucial for earthquake and tsunami early warning systems, particularly for coastal populations closest to the rupture zone of a tsunamigenic event, i.e., a “local” tsunami. Traditional seismic approaches introduce time delays that limit their effectiveness near the source, while geodetic approaches typically require a seismic trigger and are limited to empirical scaling relationships. In this study, we extend a physics-based seismogeodetic approach, which combines near-source high-rate 1 Hz Global Navigation Satellite System (GNSS) and 100 Hz acceleration data, originally developed for thrust earthquakes, to strike-slip and normal fault mechanisms. We demonstrate our expanded approach for rapid seismogeodetic magnitude (Mwg) estimation using a set of 16 historical moderate-to-large earthquakes with strike-slip, normal, and thrust mechanisms. We find that considering the propagation of P and S waves in the intermediate and far fields is critical for accurate assessment of strike-slip events. Applying radiation pattern corrections, which are difficult to compute in real time where fault geometry is not readily available, offers only minimal improvements in magnitude estimation. We further broaden the seismogeodetic approach by interpolating coseismic windows from collocated GNSS and accelerometer stations (or stations with only accelerometers and/or broadband seismometers) to stand-alone GNSS stations, thereby increasing the usable network. We present an integrated workflow for rapid magnitude estimation for the different earthquake mechanisms, which leverages tectonic context (Slab2 geometry) to inform model selection, emphasizing the importance of network geometry and station density. Overall, our extended seismogeodetic approach provides rapid warning time (~2-3 minutes after earthquake initiation) for moderate to large events (Mw > 7) with a robust (IQR) Mwg accuracy of ±0.2 magnitude units as the basis for improved earthquake and tsunami warning systems. Our approach is useful in operational environments where shortened latency is of the essence.

Wetlands are among the world’s most powerful ecosystems. In Michigan, they improve water quality, protect against erosion, and provide habitats for dozens of species; however, Michigan has lost fifty percent of its historic wetlands that existed prior to European settlement. Remote sensing has become a crucial tool to protect Michigan’s remaining wetlands. Wetlands can be seasonal, causing variations in satellite imagery from month to month. Existing studies show that certain months are better for classification, but have been limited to small-scale field studies. However, the impact of using imagery from different seasons over a larger geographic area for wetland classification remains unexplored. This study analyzed the effectiveness of Sentinel-2 and Landsat 9 imagery across seasonal and annual time frames in wetland classification, focusing on the lower peninsula of Michigan. I used a supervised classification approach to build a machine learning model, incorporating standard variables used to identify wetlands such as computed texture, indices, and elevation. I used composited imagery from 2022 and 2023 to assess across four time periods (spring, summer, fall, annual) for wetland classification using accuracy metrics. Notably, the spring composite image attained the highest accuracy among Sentinel-2, while the summer and annual composite images performed best among Landsat 9. This study reveals the potential of using spring imagery from Sentinel-2 and summer imagery from Landsat 9 to enhance wetland classification, underscoring how seasonality can affect classification accuracy. This can further guide scientists into selecting optimal data for monitoring wetlands within Michigan and areas with similar climate and geographical features.