Mitigating rising atmospheric CO₂ levels is critical to addressing climate change, and geological carbon storage in saline aquifers is a promising solution. This study develops and evaluates three deep learning modelsdeep neural network (DNN), gated recurrent unit (GRU), and recurrent neural network (RNN)to predict the residual trapping index (RTI) and solubility trapping index (STI) via diverse reservoir datasets. The GRU model outperformed both DNN and RNN, particularly in handling long-term dependencies and minimizing error, while DNN showed robust accuracy through effective modeling of complex nonlinear relationships. In contrast, RNN exhibited challenges with gradient instability, affecting its prediction performance. Sensitivity analysis highlighted the significance of input variables like post injection and injection rate, with DNN showing greater dependence on these features. Excluding these variables reduced predictive accuracy, especially for RTI. These findings suggest that GRU is the most effective for predicting CO₂ trapping, offering a valuable tool for optimizing carbon storage strategies and supporting global carbon reduction efforts.

The meteorology-dependent multiscale behavior of snow depth over the Tibetan Plateau was examined via moderate-resolution datasets. By analyzing spatial variance and power spectra at specific scales (5 km, 10 km, 20 km, and 50 km) over 28 intraseasonal continuous snow cover regions, key meteorological factors affecting snow depth across different seasons were identified based upon the Kullback-Leibler distance. In spring and winter, the homogeneous power spectra highlight the significant roles of precipitation, atmospheric moisture, and temperature in snow accumulation and melt processes. Summer exhibits the highest snow depth spatial variance driven by radiation and wind dynamics. Autumn with the lowest spatial heterogeneity underscores the impact of windspeed on snow distribution through sublimation and redistribution. The alignment between spatial variance maps and spectra across these scales implies the future downscaling or upscaling without substantial loss of spatial information. These findings are essential for improving snow cover modeling and forecasting, particularly in the context of climate change, as well as for effective water resource management and climate adaptation strategies in this strategically important plateau.

The impact of the COVID-19 lockdown on air quality in seven major Chinese cities was investigated by utilizing long-term datasets of air pollutants and meteorological conditions from 2016 to 2021. Generalized additive model (GAM) was developed to predict air quality during the lockdown period. The model accounting for weather conditions demonstrated high accuracy with predictions compared against measurements during the lockdown. Significant reductions in NO₂, CO, and PM₁₀ concentrations were observed primarily due to decreased vehicular traffic and industrial activities. Notable reductions were particularly evident in cities with high traffic volumes and industrial emissions prior to the lockdown. The study also employed transfer learning to enhance the accuracy of lockdown model with limited data. Despite occasional anomalies caused by specific events like fireworks and agricultural burning, the findings suggest that extended training periods and advanced modeling techniques can significantly improve air quality predictions. This research highlights the potential long-term benefits of sustained reductions in human activities and provides valuable insights for future air quality management and policy-making.

Yueqian Cao1, and Ana P. Barros21. Department of Civil and Environmental Engineering, Duke University, Durham, NC, USA2. Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, USA* Correspondence: barros@illinois.eduAbstract: A radar observing system simulator consisting of a coupled snow hydrology (MSHM) and radiative transfer model (MEMLS) was modified to include vegetation contributions to the total backscatter from the ground-snow-vegetation system. Vegetation parameters were estimated from airborne SnowSAR (X- and Ku-band) and Sentinel-1 (C-band) measurements in Grand Mesa (flat topography) and Senator Beck Basin (steep topography) by solving the inverse problem via simulated annealing. Physics-based constraints were imposed to address indeterminacy with good results, which highlights that the forward-inversion system accounting for complex multiple scattering within the ground-snow-vegetation system reliably regulated compensation effects of vegetation and snow-ground interface, including simulating observed background backscatter under snow-free conditions. The proximal goal of this study is to quantify the integrated effect of complex multiple scattering within the ground-snow-vegetation system toward isolating volume scattering from subcanopy snowpack, and subsequent retrieval of snowpack properties such as snow water equivalent (SWE). The stretch goal is to develop a vegetation correction to expand the operational utility of radar remote sensing of snow in the boreal forests at northern latitudes. The proposed approach has high operational utility for retrieving large-scale SWE from satellite-based SAR measurements.Keywords: forward-inversion system; MEMLS; simulated annealing; SAR; vegetation heterogeneity

An uncalibrated distributed multiphysics snow model driven by downscaled weather forecasts (30-m, 15-min) was implemented as a Radar Observing System Simulator (ROSS) in Senator Beck Basin (SBB), Colorado to elucidate topographic controls on C-, X-and Ku-bands active microwave sensing of mountain snowpacks. Phase-space maps of time-evolving grid-scale ROSS volume backscatter show the accumulation branch of the backscatter-snow water equivalent (σ-SWE) hysteresis seasonal loop that is the physical basis for radar retrieval (direct inference) of SWE and snowpack physical properties. ROSS results with snow-ground scattering correction inferred from snow-free conditions capture well the seasonal march of Sentinel-1 C-band backscatter, including spatial patterns tied to elevation, slope, and aspect. Root Mean Square Deviations (RMSDs) do not exceed ±3.2 dB for ripening snowpacks in early spring and ±2.4 dB for dry snowpacks in the accumulation season when the mean absolute bias is < 1 dB for all land-cover types with topographic slopes 30°. Grid-point RMSDs are attributed to the underestimation of snowfall on upwind slopes compounded with forecast errors for the weather near the ground. Like Sentinel-1, ROSS backscatter fields exhibit frequency-independent single-scaling behavior within the 60-150 m scale range for dry snowpacks in the accumulation season, while frequency-dependent scaling behavior emerges in the ablation season. This study demonstrates skillful physical modeling capabilities to emulate Sentinel-1 observations in complex terrain. Conversely, it suggests high readiness to retrieve snow mass and snowpack properties in mountainous regions from radar measurements at high-spatial resolutions enabled by SAR technology.

Ensemble predictions of the seasonal snowpack over Grand Mesa, CO were conducted for the hydrologic year 2016-2017 using a multilayer snow hydrology model. Ensembles were generated from gridded atmospheric reanalysis, model predictions were evaluated against SnowEx’17 measurements, and the signatures of the weather-dependent variability of snow physics in the behavior of multi-frequency microwave brightness temperatures and backscattering were examined through forward modeling. At sub-daily timescales , the ensemble standard deviation due to atmospheric forcing (i.e., mesoscale spatial variability of weather within the Grand Mesa) is < 3 dB for dry snow, and increases to 8-10 dB at midday when there is surficial melt that also explains the wide ensemble range (~20 dB). Further, the ensemble mean backscatter exhibits robust (R 2 > 0.95) time-varying, weather-dependent linear heuristic relationships with SWE (e.g., 5-6 cm/dB/month in January; 2-2.5 cm/dB/month in late February) as melt-refreeze cycles modify the microphysical structure in the top 50 cm of the snowpack. The nonlinear evolution of ensemble snow physics translates into seasonal hysteresis in the microwave behavior. The backscatter hysteretic offsets between accumulation and melt regimes are robust in the Land C-bands and collapse for wet shallow snow at Ku-band. The ensemble mean emissions behave as a limit-cycles with weak sensitivity in the accumulation regime, and hysteretic behavior during melt that is different for deep (winter-spring transition) and shallow snow (spring-summer) and offsets that increase with frequency. These findings suggest potential for multi-frequency active-passive remote-sensing of SWE conditional on snowpack regime, particularly suited for data-assimilation using coupled snow hydrology-microwave models.

The Snow Ensemble Uncertainty Project (SEUP) is an effort to establish a baseline characterization of snow water equivalent (SWE) uncertainty across North America with the goal of informing global snow observational needs. An ensemble-based modeling approach, encompassing a suite of current operational models, is used to assess the uncertainty in SWE and total snow storage (SWS) estimation during the 2009-2017 period. The highest modeled SWE uncertainty is observed in mountainous regions, likely due to the relatively deep snow, forcing uncertainties, and variability between the different models in resolving the snow processes over complex terrain. This highlights a need for high-resolution observations in mountains to capture the high spatial SWE variability. The greatest SWS is found in Tundra regions where, even though the spatiotemporal variability in modeled SWE is low, there is considerable uncertainty in the SWS estimates due to the large areal extent over which those estimates are spread. This highlights the need for high accuracy in snow estimations across the Tundra. In mid-latitude boreal forests, large uncertainties in both SWE and SWS indicate that vegetation-snow impacts are a critical area where focused improvements to modeled snow estimation efforts need to be made. Finally, the SEUP results indicate that SWE uncertainty is driving runoff uncertainty and measurements may be beneficial in reducing uncertainty in SWE and runoff, during the melt season at high latitudes (e.g., Tundra and Taiga regions) and in the Western mountain regions, whereas observations at (or near) peak SWE accumulation are more helpful over the mid-latitudes.