High-resolution Dataset Research Articles

Annual time-series 1 km maps of crop area and types in the conterminous US (CropAT-US): cropping diversity changes during 1850–2021

Abstract. Agricultural activities have been recognized as an important driver of land cover and land use change (LCLUC) and have significantly impacted the ecosystem feedback to climate by altering land surface properties. A reliable historical cropland distribution dataset is crucial for understanding and quantifying the legacy effects of agriculture-related LCLUC. While several LCLUC datasets have the potential to depict cropland patterns in the conterminous US, there remains a dearth of a relatively high-resolution datasets with crop type details over a long period. To address this gap, we reconstructed historical cropland density and crop type maps from 1850 to 2021 at a resolution of 1 km × 1 km by integrating county-level crop-specific inventory datasets, census data, and gridded LCLUC products. Different from other databases, we tracked the planting area dynamics of all crops in the US, excluding idle and fallow farm land and cropland pasture. The results showed that the crop acreages for nine major crops derived from our map products are highly consistent with the county-level inventory data, with a residual less than 0.2×103 ha (0.2 kha) in most counties (>75 %) during the entire study period. Temporally, the US total crop acreage has increased by 118×106 ha (118 Mha) from 1850 to 2021, primarily driven by corn (30 Mha) and soybean (35 Mha). Spatially, the hot spots of cropland distribution shifted from the Eastern US to the Midwest and the Great Plains, and the dominant crop types (corn and soybean) expanded northwestward. Moreover, we found that the US cropping diversity experienced a significant increase from the 1850s to the 1960s, followed by a dramatic decline in the recent 6 decades under intensified agriculture. Generally, this newly developed dataset could facilitate spatial data development, with respect to delineating crop-specific management practices, and enable the quantification of cropland change impacts on the environment. Annual cropland density and crop type maps are available at https://doi.org/10.6084/m9.figshare.22822838.v2 (Ye et al., 2023).

Image-derived indicators of phytoplankton community responses to Pseudo-nitzschia blooms

Phytoplankton populations in the natural environment interact with each other. Despite rising global concern with Pseudo-nitzschia blooms, which can produce the potent neurotoxin domoic acid, we still do not fully understand how other phytoplankton genera respond to the presence of Pseudo-nitzschia. Here, we used a 4-year high-resolution imaging dataset for 9 commonly found phytoplankton genera in Narragansett Bay, alongside environmental data, to identify potential interactions between phytoplankton genera and their response to elevated Pseudo-nitzschia abundance. Our results indicate that Pseudo-nitzschia tends to bloom either concurrently with or right after other phytoplankton genera. Such bloom periods coincide with higher water temperatures and lower salinity. Pseudo-nitzschia image abundance tends to increase the most from March-May and peaks during May-Jun, whereas the image-derived biovolume and width of Pseudo-nitzschia chains increase the most during Jan-Feb. For most phytoplankton genera, their relationship with Pseudo-nitzschia abundance is noticeably different from their relationship with Pseudo-nitzschia image features. Despite the complexity in the phytoplankton community, our analysis suggests several ecological indicators that may be used to determine the risk of harmful algal blooms.

Extreme precipitation events in the Brahmaputra River basin during the Indian summer monsoon and its association with atmospheric moisture transport

ABSTRACT Due to climate change, rising temperatures cause the atmosphere to hold more moisture, likely increasing the frequency and intensity of extreme precipitation events. These events, lasting several days, are the leading causes of floods. Recently, floods from multi-day precipitation extremes (MPEs) have risen significantly. The Brahmaputra River basin (BRB) in India is particularly vulnerable to flooding during the Indian summer monsoon due to its transboundary nature. It is crucial to characterize and rank MPEs to understand their risk, impact, and underlying drivers. This study ranked MPEs of different durations based on intensity and spatial extent during the Indian summer monsoon over the BRB using a high-resolution daily precipitation dataset from 1951 to 2022. In addition, it evaluates the association between atmospheric moisture transport and MPEs by quantifying integrated water vapor transport (IVT) during top-ranked MPEs. The analysis indicates intense IVT in the mid-tropospheric layer (850–500 hPa) over the MPE regions during top-ranked events and identifies significant low-pressure anomalies and shifts from ridge to trough patterns. Quantifying the connection between MPEs and IVT can aid in early prediction and risk reduction.

Shoreliner: A Sub-Pixel Coastal Waterline Extraction Pipeline for Multi-Spectral Satellite Optical Imagery

Beach morphology can be observed over large spatio-temporal scales, and future shoreline positions can be predicted and coastal risk indicators can be derived by measuring satellite-derived instantaneous waterlines. Long-term satellite missions, such as Landsat and Sentinel-2, provide decades of freely available, high-resolution optical measurement datasets, enabling large-scale data collection and relatively high-frequency monitoring of sandy beaches. Satellite-Derived Shoreline (SDS) extraction methods are emerging and are increasingly being applied over large spatio-temporal scales. SDS generally consists of two steps: a mathematical relationship is applied to obtain a ratio index or pixel classification by machine-learning algorithms, and the land/sea boundary is then determined by edge detection. Indexes from lake waterline detection, such as AWEI or NDWI, are often transferred towards the shore without taking into account that these indexes are inherently affected by wave breaking. This can be overcome by using pixel classification to filter the indices, but this comes at a computational cost. In this paper, we carry out a thorough evaluation of the relationship between scene-dependent variables and waterline extraction accuracy, as well as a robust and efficient thresholding method for coastal land–water classification that optimises the index to satellite radiometry. The method developed for sandy beaches combines a new purpose-built multispectral index (SCoWI) with a refinement method of Otsu’s threshold to derive sub-pixel waterline positions. Secondly, we present a waterline extraction pipeline, called Shoreliner, which combines the SCoWI index and the extraction steps to produce standardised outputs. Implemented on the CNES High Performance Cluster (HPC), Shoreliner has been quantitatively validated at Duck, NC, USA, using simultaneous Sentinel-2 acquisitions and in situ beach surveys over a 3-year period. Out of six dates that have a satellite acquisition and an in situ survey, five dates have a sub-pixel RMS error of less than 10 m. This sub-pixel performance of the extraction processing demonstrates the ability of the proposed SDS extraction method to extract reliable, instantaneous and stable waterlines. In addition, preliminary work demonstrates the transferability of the method, initially developed for Sentinel-2 Level1C imagery, to Landsat imagery. When evaluated at Duck on the same day, Sentinel-2 and Landsat imagery several minutes apart provide similar results for the detected waterline, within the method’s precision. Future work includes global validation using Landsat’s 40 years of data in combination with the higher resolution Sentinel-2 data at different locations around the world.

Evaluation of GPM IMERG Early, Late, and Final Run in Representing Extreme Rainfall Indices in Southwestern Iran

The growing concerns about floods have highlighted the need for accurate and detailed precipitation data as extreme precipitation occurrences can lead to catastrophic floods, resulting in significant economic losses and casualties. Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (GPM IMERG) is a commonly used high-resolution gridded precipitation dataset and is recognized as trustworthy alternative sources of precipitation data. The aim of this study is to comprehensively evaluate the performance of GPM IMERG Early (IMERG-E), Late (IMERG-L), and Final Run (IMERG-F) in precipitation estimation and their capability in detecting extreme rainfall indices over southwestern Iran during 2001–2020. The Asfezari gridded precipitation data, which are developed using a dense of ground-based observation, were utilized as the reference dataset. The findings indicate that IMERG-F performs reasonably well in capturing many extreme precipitation events (defined by various indices). All three products showed a better performance in capturing fixed and non-threshold precipitation indices across the study region. The findings also revealed that both IMERG-E and IMERG-L have problems in rainfall estimation over elevated areas showing values of overestimations. Examining the effect of land cover type on the accuracy of the precipitation products suggests that both IMERG-E and IMERG-L show large and highly unrealistic overestimations over inland water bodies and permanent wetlands. The results of the current study highlight the potential of IMERG-F as a valuable source of data for precipitation monitoring in the region.

Minnesota’s energy paradox: Household energy insecurity in the face of racial and economic disparities

The “Minnesota Paradox” points to the nation-leading socioeconomic indicators enjoyed in Minnesota that belie some of the largest racial disparities between the state’s majority White and minority Black populations. The Minnesota Paradox has been identified in indicators with complex, structural social determinants: income, employment, educational outcomes, incarceration rates, home ownership, and even drowning. In this paper, we ask if similar disparities exist in access to shared infrastructure systems, focusing on the electric system, an essential service delivered by heavily regulated public utilities. We examine disparities in access to electricity service across three dimensions: utility disconnection, service reliability, and availability of the grid to host distributed energy resources. We quantify disparities across Census block groups by leveraging unique, high-resolution datasets that have only recently been made publicly available. We find significant and pervasive evidence of the Minnesota Paradox across utility disconnection and service reliability. Across a battery of regression models, we find that living in neighborhoods with a greater concentration of people of color is associated with a statistically and practically significant difference in the likelihood of disconnection from service due to non-payment and the experience of extended power outages. We also find a positive association between communities with larger populations of people of color and hosting capacity, suggesting a potential opportunity to affirmatively address disparities in energy insecurity in the energy transition. These findings shed light on the pervasive nature of the Minnesota Paradox in the electric system, which underscores the pressing need for policy initiatives to rectify deep-seated inequalities and ensure all communities have equitable access to universal basic utility service and reliable, clean energy.

Microclimatic Zonation and Climatic Variability of Sikkim Himalaya

The Köppen classification of climate integrates precipitation and temperature information with natural vegetation patterns to create a precise representation of any particular region's climate. This integration depends on the empirical relationship of climate and vegetation, which indicates that distinct locations in the same class have similar vegetation attributes. Köppen climatic classification factors are explained and Sikkim's climate characteristics are regionalized based on it. The method for making representations of air temperatures and precipitation has been described, and an illustration of Sikkim's climatic zones with variability is generated as a result of these changes. The geographic pattern of climatic types and subtypes in Sikkim has been briefly addressed using an available high-resolution gridded dataset (ERA5-Land). This is described that the constraints of microclimatic zonation emanate from the empirical prerequisite of climate classifications, as well as the nature of data selection and the methodologies employed for climate variability analysis and classification. Based on the Köppen classification for the long term (1980–2021), the Sikkim Himalaya contains three primary climatic classes, particularly ETc (cold, tundra, and cool summer), Cfc (moderately warm, humid, and cool summer), and Cfb (moderately warm, humid, and warm summer). Climate variability on the basis of temperature and precipitation change with respect to 1980–2021 over the entire Sikkim Himalaya concludes that the climatic pattern of the Sikkim has been changed from cold-dry to warm-wet. The alteration in the corresponding climatic pattern is further supported by changes in LULC and NDVI. The results suggest that the precipitation change in the past two decades (1980–2000) is negative, while a significant positive change has been noticed in the recent two decades (2001–2021). Subsequently, the number of extremely wet days decreases in the entire ETc and Cfc climate zones. Regardless, the southern part of the Cfb climatic zone has experienced an increase in extremely wet days. This study's findings will significantly contribute to the development of future policies and initiatives by providing insights crucial for achieving sustainable development, environmental protection, and climate change adaptation.

Retrieval of dominant methane (CH4) emission sources, the first high-resolution (1–2 m) dataset of storage tanks of China in 2000–2021

Abstract. Methane (CH4) is a significant greenhouse gas in exacerbating climate change. Approximately 25 % of CH4 is emitted from storage tanks. It is crucial to spatially explore the CH4 emission patterns from storage tanks for efficient strategy proposals to mitigate climate change. However, due to the lack of publicly accessible storage tank locations and distributions, it is difficult to ascertain the CH4 emission spatial pattern over a large-scale area. To address this problem, we generated a storage tank dataset (STD) by implementing a deep learning model with manual refinement based on 4403 high-spatial-resolution images (1–2 m) from the Gaofen-1, Gaofen-2, Gaofen-6, and Ziyuan-3 satellites over city regions in China with officially reported numerous storage tanks in 2021. STD is the first storage tank dataset for over 92 typical city regions in China. The dataset can be accessed at https://doi.org/10.5281/zenodo.10514151 (Chen et al., 2024). It provides a detailed georeferenced inventory of 14 461 storage tanks wherein each storage tank is validated and assigned the construction year (2000–2021) by visual interpretation of the collected high-spatial-resolution images, historical high-spatial-resolution images of Google Earth, and field survey. The inventory comprises storage tanks with various distribution patterns in different city regions. Spatial consistency analysis with the CH4 emission product shows good agreement with storage tank distributions. The intensive construction of storage tanks significantly induces CH4 emissions from 2005 to 2020, underscoring the need for more robust measures to curb CH4 release and aid in climate change mitigation efforts. Our proposed dataset, STD, will foster the accurate estimation of CH4 released from storage tanks for CH4 control and reduction and ensure more efficient treatment strategies are proposed to better understand the impact of storage tanks on the environment, ecology, and human settlements.

Seasonal and Interannual Variability in Sea Surface Temperature Fronts in the Levantine Basin, Mediterranean Sea

Sea surface temperature (SST) fronts were analyzed in the Levantine Basin of the Mediterranean Sea over a 20-year period (2003–2022) using a high-resolution (~1 km) satellite dataset. Frontal gradients were strongest in regions of freshwater influence and around the Ierapetra eddies and Rhodes Gyre. Seasonally, maximum frontal activity was observed in fall and summer. Empirical orthogonal function (EOF) analysis revealed both monthly-to-seasonal variability and interannual variability in frontal gradients. Seasonal frontal variability is partially explained by atmospheric forcing; that is, wind stress curl (WSC) and net air–sea heat flux. The maximum frontal activity was observed in 2006, coinciding with the strongest WSC magnitude. The minimum frontal activity was observed in 2017, which saw the largest winter heat loss to the atmosphere. The highest frontal activity was typically observed in years with mild winters followed by strong Etesian winds. Over the study period (2003–2022), frontal gradients declined in the Levantine Basin. Our results suggest that years with a strong frontal boundary current (Asia Minor Current; AMC) coincide with reduced cross-shelf transport. Subsequent studies are recommended to concentrate on the variability in the frontal intensity of the AMC and associated cross-shelf transports, which are important for the oligotrophic Levantine Basin.

Amplification of temperature extremes in Arabian Peninsula under warmer worlds

The Paris Agreement and the Special Report on Global Warming of 1.5 °C from the Intergovernmental Panel on Climate Change (IPCC) highlighted the potential risks of climate change across different global warming levels (GWLs). The increasing occurrence of extreme high-temperature events is linked to a warmer climate that is particularly prevalent in the Arabian Peninsula (AP). This study investigates future changes in temperatures and related extremes over AP, under four GWLs, such as 1.5 °C, 2.0 °C, 3.0 °C, and 4.0 °C, with three different Shared Socioeconomic Pathways (SSPs: SSP1-2.6, SSP2-4.5, and SSP5-8.5). The study uses high-resolution datasets of 27 models from the NASA Earth Exchange Global Daily Downscaled Projections of the Coupled Model Intercomparison Project Phase 6 (NEX-GDDP-CMIP6). The results showed that the NEX-GDDP-CMIP6 individual models and their multi-model means reasonably captured the extreme temperature events. The summer maximum and winter minimum temperatures are projected to increase by 0.11–0.67 °C and 0.09–0.70 °C per decade under the selected SSPs. Likewise, the projected temperature extremes exhibit significant warming with varying degrees across the GWLs under the selected SSPs. The warm temperature extremes are projected to increase, while the cold extremes are projected to decrease under all GWLs and the selected SSPs. Overall, the findings provide a comprehensive assessment of temperature changes over AP in response to global warming, which can be helpful in the development of climate adaptation and mitigation strategies.

Global-scale evaluation of precipitation datasets for hydrological modelling

Abstract. Precipitation is the most important driver of the hydrological cycle, but it is challenging to estimate it over large scales from satellites and models. Here, we assessed the performance of six global and quasi-global high-resolution precipitation datasets (European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis version 5 (ERA5), Climate Hazards group Infrared Precipitation with Stations version 2.0 (CHIRPS), Multi-Source Weighted-Ensemble Precipitation version 2.80 (MSWEP), TerraClimate (TERRA), Climate Prediction Centre Unified version 1.0 (CPCU), and Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System-Climate Data Record (PERSIANN-CCS-CDR, hereafter PERCCDR) for hydrological modelling globally and quasi-globally. We forced the WBMsed global hydrological model with the precipitation datasets to simulate river discharge from 1983 to 2019 and evaluated the predicted discharge against 1825 hydrological stations worldwide, using a range of statistical methods. The results show large differences in the accuracy of discharge predictions when using different precipitation input datasets. Based on evaluation at annual, monthly, and daily timescales, MSWEP followed by ERA5 demonstrated a higher correlation (CC) and Kling–Gupta efficiency (KGE) than other datasets for more than 50 % of the stations, whilst ERA5 was the second-highest-performing dataset, and it showed the highest error and bias for about 20 % of the stations. PERCCDR is the least-well-performing dataset, with a bias of up to 99 % and a normalised root mean square error of up to 247 %. PERCCDR only show a higher KGE and CC than the other products for less than 10 % of the stations. Even though MSWEP provided the highest performance overall, our analysis reveals high spatial variability, meaning that it is important to consider other datasets in areas where MSWEP showed a lower performance. The results of this study provide guidance on the selection of precipitation datasets for modelling river discharge for a basin, region, or climatic zone as there is no single best precipitation dataset globally. Finally, the large discrepancy in the performance of the datasets in different parts of the world highlights the need to improve global precipitation data products.

Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography.

Automated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT). We begin with an extensive review of current machine learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using Hierarchical Phase-Contrast Tomography (HiP-CT) as part of the Human Organ Atlas Project. HiP-CT, pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering resolution around 20μm/voxel, and enabling highly detailed localized zooms up to 1μm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging. Our three experiments yielded Dice scores of 0.9523 and 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores-such as centerline Dice values ranging from 0.82 to 0.88, certain errors persisted. Specifically, large vessels that collapsed due to the lack of hydro-static pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Through our review and outputs, we aim to set a benchmark for subsequent model evaluations using various modalities, especially with the HiP-CT imaging database.

Assessing uncertainty in bioclimatic modelling: a comparison of two high-resolution climate datasets in northern Patagonia

Climate change is reshaping forest ecosystems, presenting urgent and complex challenges that demand attention. In this context, research that quantifies interactions between climate and forests is substantial. However, modelling at a spatial resolution relevant for ecological processes presents a significant challenge, especially given the diverse geographical contexts in which it is applied. In our study, we aimed to assess the effects of applying CHELSA v.2.1 and WorldClim v2.1 data on bioclimatic analysis within the Río Puelo catchment area in northern Patagonia. To achieve this, we inter-compared and evaluated present and future bioclimates, drawing on data from both climate datasets. Our findings underscore substantial consistency between both datasets for temperature variables, confirming the reliability of both for temperature analysis. However, a strong contrast emerges in precipitation predictions, with significant discrepancies highlighted by minimal overlap in bioclimatic classes, particularly in steep and elevated terrains. Thus, while CHELSA and WorldClim provide valuable temperature data for northern Patagonia, their use for precipitation analysis requires careful consideration of their limitations and potential inaccuracies. Nevertheless, our bioclimatic analyses of both datasets under different scenarios reveal a uniform decline in mountain climates currently occupied by N. pumilio, with projections suggesting a sharp decrease in their coverage under future climate scenarios.

Global application of a regional frequency analysis to extreme sea levels

Abstract. Coastal regions face increasing threats from rising sea levels and extreme weather events, highlighting the urgent need for accurate assessments of coastal flood risk. This study presents a novel approach to estimating global extreme sea level (ESL) exceedance probabilities using a regional frequency analysis (RFA) approach. The research combines observed and modelled hindcast data to produce a high-resolution (∼1 km) dataset of ESL exceedance probabilities, including wave setup, along the entire global coastline (excluding Antarctica). The methodology presented in this paper is an extension of the regional framework of Sweet et al. (2022), with innovations introduced to incorporate wave setup and apply the method globally. Water level records from tide gauges and a global reanalysis of tide and surge levels are integrated with a global ocean wave reanalysis. Subsequently, these data are regionalised, normalised, and aggregated and then fit with a generalised Pareto distribution. The regional distributions are downscaled to the local scale using the tidal range at every location along the global coastline obtained from a global tide model. The results show 8 cm of positive bias at the 1-in-10-year return level when compared to individual tide gauges. The RFA approach offers several advantages over traditional methods, particularly in regions with limited observational data. It overcomes the challenge of short and incomplete observational records by substituting long historical records with a collection of shorter but spatially distributed records. These spatially distributed data not only retain the volume of information but also address the issue of sparse tide gauge coverage in less populated areas and developing nations. The RFA process is illustrated using Cyclone Yasi (2011) as a case study, demonstrating how the approach can improve the characterisation of ESLs in regions prone to tropical cyclone activity. In conclusion, this study provides a valuable resource for quantifying the global coastal flood risk, offering an innovative global methodology that can contribute to preparing for – and mitigating against – coastal flooding.

Tracing Opioids Across the US: A High-Resolution Pharmaceutical Distribution Dataset

In the field of pharmaceutical supply chains, there is a lack of comprehensive historical data, representing a significant barrier to advancing research. To address this gap, we introduce a high-resolution dataset comprising drug packages distributed to approximately 300,000 pharmacies, hospitals, and practitioners across the US. We reconstruct 375 million distribution paths from ARCOS, a DEA-maintained database comprising half a billion shipping records between 2006 and 2014. While ARCOS tracks dyadic shipments, it does not provide information on the complete journey of single packages from manufacturers to final destinations. Our algorithm is able to reconstruct complete distribution paths from these dyadic records. The reconstructed dataset, with its high temporal and spatial resolution, offers an unprecedented view of US pharmaceutical distribution and is a valuable resource for investigating supply and distribution networks.

Climate change can disproportionately reduce habitats of stream fishes with restricted ranges in southern South America

Freshwater fishes are among the most threatened taxa worldwide owing to changes in land use, species introductions, and climate change. Although more than half of the freshwater fishes in the Chilean Mediterranean ecoregion are considered vulnerable or endangered, still little is known about their biogeography. Fishes of the family Perciliidae are endemic of this region and ideal cases to study potential implications of global warming given their endangered conservation status, small size, restricted range, and limited dispersal capacity in fragmented habitats. Here, we model the spatial distribution of habitats for Percilia irwini and P. gillissi under current (1970–2000) and future (2050–2080) climatic scenarios (SSP245, SSP585). We implement maximum entropy (MaxEnt) models adapted for stream networks using high-resolution datasets of selected geophysical and climatic variables. At present, both species inhabit relatively low-quality habitats. In the future (SSP585), suitable habitats for P. irwini are predicted to be reduced drastically (99%) with potential local extirpations in its northern range. Similarly, up to 62% of suitable habitats for P. gillissi would also be reduced in the future. Our study provides insights about assessing future threats and vulnerability of endemic, endangered, range-restricted, and small-bodied freshwater species in this region and elsewhere.

Advancing Hydrology through Machine Learning: Insights, Challenges, and Future Directions Using the CAMELS, Caravan, GRDC, CHIRPS, PERSIANN, NLDAS, GLDAS, and GRACE Datasets

Machine learning (ML) applications in hydrology are revolutionizing our understanding and prediction of hydrological processes, driven by advancements in artificial intelligence and the availability of large, high-quality datasets. This review explores the current state of ML applications in hydrology, emphasizing the utilization of extensive datasets such as CAMELS, Caravan, GRDC, CHIRPS, NLDAS, GLDAS, PERSIANN, and GRACE. These datasets provide critical data for modeling various hydrological parameters, including streamflow, precipitation, groundwater levels, and flood frequency, particularly in data-scarce regions. We discuss the type of ML methods used in hydrology and significant successes achieved through those ML models, highlighting their enhanced predictive accuracy and the integration of diverse data sources. The review also addresses the challenges inherent in hydrological ML applications, such as data heterogeneity, spatial and temporal inconsistencies, issues regarding downscaling the LSH, and the need for incorporating human activities. In addition to discussing the limitations, this article highlights the benefits of utilizing high-resolution datasets compared to traditional ones. Additionally, we examine the emerging trends and future directions, including the integration of real-time data and the quantification of uncertainties to improve model reliability. We also place a strong emphasis on incorporating citizen science and the IoT for data collection in hydrology. By synthesizing the latest research, this paper aims to guide future efforts in leveraging large datasets and ML techniques to advance hydrological science and enhance water resource management practices.

Disparities of compound exposure of particulate matter (PM2.5) and heat index using citywide monitoring networks

As epicenters of environmental change, cities face multiple threats from environmental hazards, including poor air quality and extreme heat. The complex features of urban environments, such as varying landscapes and emission sources, significantly affect these phenomena. This study investigated the combined exposure to PM2.5 and extreme heat within Chicago, considering socioeconomic aspects to gage environmental justice issues. This study utilized high-resolution environmental datasets and Random Forest Spatial Interpolation (RFSI) to generate hourly Heat Index and PM2.5 maps at a spatial resolution of 250 m. The RFSI showed robust performance, with cross-validation R2 ranging from 0.78 to 0.98 (air temperature (Ta): 0.98, relative humidity (RH): 0.93, and PM2.5: 0.84) and RMSE from 0.84 to 5.4 (Ta: 1.09, RH: 5.40, and PM2.5: 0.84). With this result, this research visualized the spatial variations in extreme heat, as measured by the Heat Index and PM2.5 levels, and identified areas with critical exposure that are potentially harmful to the health of vulnerable populations. Furthermore, this study found the spatial disparities in exposure linked to socioeconomic factors by conducting a Welch ANOVA test. These findings can inform the development of targeted interventions considering the temporal-spatial disparities of heat and air pollution levels.

ADGEO: A new shore‐based approach to improving spatial accuracy when mapping water bodies using low‐cost drones

AbstractOver the last three decades, satellite imagery has been instrumental in mapping and monitoring water quality. However, satellites often have limitations due to image availability and cloud cover. Today, the spatial resolution of satellite images does not provide finer detail measurements essential for small‐scale water pollution management. Drones offer a complimentary platform capable of operating below cloud cover and acquiring very high spatial resolution datasets in near real‐time. Studies have shown that drone mapping over water can be done via the Direct Georeferencing approach. However, this method is only suitable for high‐end drones with accurate GNSS/IMU. Importantly, this limitation is exacerbated because of the difficulty in placing targets over water, which can be used to improve the accuracy after the survey. This study explored a new method called Assisted Direct Georeferencing which combines the benefits of traditional Bundle Adjustment with Direct Georeferencing. The performance of the approach was evaluated over a variety of different scenarios, demonstrating significant improvement in the planimetric accuracy. From the results, the method reduced the error in XY of drone imagery from MAE of 18.9 to 3.4 m. The result shows the potential of low‐cost drones with Assisted Direct Georeferencing in closing the gap to high‐end drones.

Understanding bottom and surface marine heatwaves along the continental shelf of China

Marine heatwaves (MHWs) have become longer and more frequent over the past century under anthropogenic climate change, with devastating impacts on marine ecosystems. Surface MHWs (SMHWs) and their drivers have been extensively studied using satellite sea surface temperature data, yet the mechanism and characteristics of subsurface MHWs, especially bottom MHWs (BMHWs) along continental shelves, remain unclear. Based on a high-resolution ocean reanalysis dataset, we compare SMHWs and BMHWs along the continental shelf of China and find that BMHWs are typically longer (0–16 d) and more intense (0 °C–50 °C days) than SMHWs. The categorizing of both the BMHW and SMHW shows that moderate and strong events commonly occur in most areas with relatively large spatial coverage, whereas severe and extreme events occur with relatively small spatial coverage. There is a clear negative relationship between the BMHW intensity and ocean depth along the continental shelf, while the BMHW annual days and ocean depth are positively correlated in the Bohai and East China Seas. Generally, BMHWs and SMHWs occur more frequently in shallow coastal regions where the mixed layer depth is more likely to extend to the seafloor, resulting in high BMHW and SMHW synchrony. In addition to spatial coherence, there is a good temporal correspondence between BMHWs and SMHWs across the continental shelf of China from 1993 to 2020.