Artificial intelligence in environmental remote sensing: Progress,way forward and key considerations

Abstract

Environmental remote sensing (ERS) has been a cornerstone technology for decades, allowing for comprehensive monitoring and management of ecosystems. Recent advancements in artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), have transformed the field of ERS by offering robust, data-driven solutions for analyzing complex, large-scale remote sensing datasets and answering important scientific questions. This intervention explores the diverse applications of AI in ERS, focusing on key domains such as water, agriculture, urban, and wetland monitoring. This work also includes discussions on crucial aspects such as the impact of foundation models in ERS and outlines key points on the ethical and responsible use of AI in ERS. The intervention is concluded by identifying promising next steps for future research.

Keywords

Environment monitoring remote sensing machine learning deep learning foundation models responsible AI

Introduction

The deterioration of the Earth’s environment, driven largely by human activities, has reached unprecedented levels and poses significant challenges to sustainable development (Adger et al. 2024; Díaz et al. 2019; Fletcher et al. 2024). Issues such as resource depletion, deforestation, climate change, and pollution are no longer confined to specific regions but have become global concerns (Hirsch 2020; Nations 2021). For instance, cyanobacterial harmful algal blooms (CyanoHABs), once considered localized water quality issues, have now emerged as widespread ecological threats affecting freshwater and coastal systems across continents (Mishra et al. 2020). Eutrophication, exacerbated by nutrient runoff from agriculture and urbanization, has led to large-scale toxic blooms that endanger drinking water supplies, fisheries, and aquatic biodiversity from the Great Lakes in North America to Lake Victoria in Africa and even in remote Arctic lakes (Feng et al. 2024; Trainer et al. 2020). Similarly, deforestation in the Amazon rainforest not only accelerates regional biodiversity loss but also disrupts global carbon and hydrological cycles, contributing to climate instability worldwide (Lovejoy and Nobre 2018). The intensification of wildfires, such as the recent catastrophic wildfires in Los Angeles, CA, highlights how climate change-driven disasters are no longer isolated events (Marolla 2025). The rapid melting of polar ice due to climate change has similar far-reaching consequences, including rising sea levels that threaten coastal cities from Jakarta to Miami (Masson-Delmotte et al. 2021). Even air pollution, once considered a localized urban challenge, has now been shown to have transboundary effects, with pollutants such as fine particulate matter (PM2.5) traveling thousands of kilometers and affecting air quality and public health far from their sources (Wang et al. 2024). As the intensity and frequency of these environmental crises increase, monitoring and managing these changes has become a critical task.

Over the past 50 years, remote sensing technology has emerged as an indispensable tool for investigating Earth’s resources and monitoring environmental changes at local, regional, and global scales (Chuvieco 2020; Wulder et al. 2022). Its macro-level perspective, comprehensive coverage, and ability to provide dynamic and accurate observations make it invaluable for understanding and mitigating environmental challenges (Avtar et al. 2020; Manfreda et al. 2018; Pettorelli et al. 2014). Traditional remote sensing data processing relies heavily on physical models to derive environmental parameters from sensor observations (Nair, Mangla and Indu 2022). These models are effective in establishing relationships between observed data and environmental conditions. However, they come with significant limitations. The accuracy of physical models depends on prior knowledge of various model parameters, which often suffer from uncertainties due to the high complexity of multi-scalar environmental processes (Brynjarsdottir and OHagan 2014; Warmink et al. 2010). These uncertainties, combined with variations and heterogeneity across time and space, reduce the reliability of remote sensing outputs for environmental monitoring (Zakeri and Mariethoz 2021). As a result, there has been a growing interest in data-driven approaches, specifically artificial intelligence (AI)-based machine learning (ML) and deep learning (DL) techniques, to overcome these limitations and unlock new possibilities in environmental monitoring (Yuan et al. 2020).

The advent of “Big Earth Data” has revolutionized how we perceive, analyze, and interpret environmental changes (Li, Ma and Zhang 2021). Remote sensing data, collected through satellites, drones, and ground-based sensors, has increased exponentially in volume, variety, and velocity (Hong et al. 2024a). This abundance of data has coincided with rapid advancements in ML techniques, leading to the development of highly accurate, scalable, and efficient models for environmental remote sensing (ERS) (Yuan et al. 2020). DL has garnered particular attention due to its ability to approximate complex nonlinear relationships and extract hierarchical features from massive datasets (Ma et al. 2019). Unlike traditional approaches, DL models learn directly from data, capturing intricate associations between environmental parameters and remote sensing observations (Bengio 2013; LeCun, Bengio and Hinton 2015). This capability of AI that encompasses ML and DL has transformed the field of ERS, enabling significant improvements in image processing, classification, retrieval, and predictive modeling. Traditional image processing techniques have been complemented by AI-driven approaches, which have significantly improved tasks like multi-scale and multi-level feature extraction, representation learning, robust classification, and anomaly detection (Mai et al. 2023a; Yuan et al. 2020; Yuan, Shi and Gu 2021; Zhang and Zhang 2022). DL models, in particular, have demonstrated superior performance in handling large, heterogeneous datasets and deriving actionable insights. For instance, convolutional neural networks (CNNs) and vision transformers (Alexey 2020) are widely used for land cover mapping, while recurrent neural networks (RNNs) and video vision transformers (Arnab et al. 2021) are applied for time-series analysis in climate and hydrological studies. These models have not only improved accuracy but also introduced new capabilities, such as real-time monitoring and forecasting.

The success of AI in remote sensing spans diverse environmental applications, including water quality monitoring (O’Shea et al. 2021; Pahlevan et al. 2022), urban settlement monitoring (Li et al. 2023; Neupane, Horanont and Aryal 2021), wetlands monitoring (Campbell et al. 2022; Pei et al. 2022; Wei et al. 2017), and assessing forest health (Hamedianfar et al. 2022). This is particularly crucial in the context of global climate change, wherein timely and accurate data is essential for mitigation and adaptation efforts (Chen et al. 2023; Huntingford et al. 2019). AI in ERS continues to evolve, with significant opportunities for further innovation (Zhang and Zhang 2022). For instance, knowledge-guided ML,which combines physical models with ML and DL frameworks (Liu et al. 2024b; Mai et al. 2025; Sharma et al. 2024) has the potential to enhance interpretability and robustness, especially in the low-data regimes (Mai et al. 2023b; Wang et al. 2020), bridging the gap between traditional methods and modern AI techniques (Liu et al. 2022, 2024b).

This intervention aims to provide a comprehensive analysis of AI applications in ERS, through pilot research results incorporating different kinds of AI such as image classification, detection, and segmentation, time-series forecasting, and bio-optical modeling. We mainly focus on their transformative potential, current limitations, and future directions. These methods are applied in different environmental domains including water quality monitoring, urban environment monitoring, precision agriculture, and carbon flux monitoring. This paper also talks about the role of foundation models in ERS, both vision and language-based. Finally, some important considerations around the ethical and responsible use of AI in ERS are discussed.

Applications of AI in ERS

Sea surface salinity mapping

Sea surface salinity (SSS) is a critical parameter influencing ocean circulation, climate regulation, and the health of marine ecosystems (Boutin et al. 2021). Variations in SSS affect seawater density, thereby driving thermohaline circulation, which plays a pivotal role in distributing heat across the globe and regulating climate patterns (Rahmstorf 2006). Changes in salinity can also impact the Atlantic Meridional Overturning Circulation (AMOC), with potential consequences for global climate systems (Zhu et al. 2023). In estuarine and coastal regions, salinity dynamics are essential for maintaining biodiversity, as they influence the distribution and health of various aquatic species and habitats (Lorrain-Soligon et al. 2023; Smyth and Elliott 2016). Understanding these dynamics is particularly important in the context of climate change, sea-level rise, and extreme weather events, which can alter salinity patterns and, consequently, the structure and function of marine ecosystems (Slater et al. 2023; Wang et al. 2022).

Remote sensing technologies have advanced the monitoring of SSS, offering comprehensive spatial and temporal coverage that surpasses traditional in situ measurements (Chen and Hu 2017; Fu et al. 2021; Kim et al. 2023). Satellite missions such as the European Space Agency’s Soil Moisture and Ocean Salinity (SMOS) and NASA’s Soil Moisture Active Passive (SMAP) have been instrumental in providing global salinity data (Fournier et al. 2019). However, accurately retrieving SSS from satellite observations remains challenging due to atmospheric interference and the complex nature of coastal environments (Chen and Hu 2017). To address these challenges, researchers are increasingly integrating ML algorithms with remote sensing data to enhance SSS estimation. For instance, studies have employed random forest algorithms and other ML techniques to improve the accuracy of salinity predictions in various oceanic regions (Borovskaya et al. 2022; Kim et al. 2023).

Figure 1 presents pilot results as a part of this intervention for SSS estimation using satellite reflectance values, for Georgia’s coastal estuaries. Continuous ocean water encroachment in the estuaries creates a transition zone supporting unique biodiversity including marshes, reefs, and aquaculture. A dataset with 314 samples, comprising in situ SSS measurements (from GCE-LTER) and corresponding satellite reflectance values from Sentinel-2’s multispectral instrument (MSI), was curated. Random forest, a decision tree based ML regressor (Breiman 2001), trained on this dataset achieved an Root Mean Squared Error (RMSE) of 4.68 practical salinity units (psu) for SSS estimation from MSI spectral image input. Figure 1 shows the resulting SSS map, highlighting the model’s ability to capture salinity gradients, particularly in the estuarine regions. This highlights the potential of ERS and AI in monitoring salinity dynamics over larger spatial and temporal scales. Recent studies are focusing on using similar advanced approaches to broaden the understanding of salinity patterns linked with environmental factors such as sea level rise or climate change (Kesavakumar, Shanmugam and Venkatesan 2022; Kim et al. 2023; Nguyen et al. 2021; Nieves, Radin and Camps-Valls 2021). Ultimately, such advancements can support maritime conservation and management, ensuring resilience against future environmental stressors.

Figure 1.

(A) Sentinel-2 MSI true color image of the Georgian coastal estuaries. (B) Sea surface salinity map generated by the random forest using the same image.

CyanoHAB trend forecasting

CyanoHABs are a growing global problem due to their increasing frequency, severity, and adverse effects on ecosystems, public health, and economies (Feng et al. 2024; Huisman et al. 2018; Paerl and Otten 2013; Steffen et al. 2014). Many cyanobacteria produce potent toxins, such as microcystins, anatoxins, saxitoxins, and cylindrospermopsins, which can contaminate drinking water sources and recreational water bodies, leading to severe health risks for humans and animals (Carmichael 2001; Zhang et al. 2022b). Remote sensing has emerged as a powerful tool for tracking and assessing CyanoHABs due to its ability to provide high-frequency, synoptic observations of large water bodies using multispectral and hyperspectral satellite sensors (Dev et al. 2022; Wynne et al. 2021), as well as airborne hyperspectral imaging and drone-based surveys (Wu et al. 2019).

The integration of AI with remote sensing has significantly improved CyanoHAB monitoring infrastructure (Gaur, Pant and Jalal 2023; Marrone et al. 2023). ML and DL algorithms, trained on multi-source satellite imagery and in situ data, perform better than empirical and semi-empirical models, even in optically complex water systems (Feng et al. 2024; O’Shea et al. 2021). Despite the availability of robust monitoring tools, real-time forecasting remains a crucial next step. Recent studies show that the ability to predict bloom formation in advance is essential for mitigating risks and implementing timely interventions (Almuhtaram et al. 2021; Schaeffer et al. 2024; Zahir et al. 2024).

Figure 2 shows an example of forecasting Cyanobacterial Cell Density (CCD) trend, as part of this intervention study, for Lake Okeechobee in Florida—a lake persistently vulnerable to CyanoHABs (Jiang et al. 2025; Mazzei, Sullivan and Loftin 2025; Rosen et al. 2017). A monthly aggregated CCD trend for Lake Okeechobee from October 2016 to April 2021 was extracted using the Google Earth Engine (GEE) based CyanoKhoj dashboard (Maniyar, Kumar and Mishra 2022). Long short term memory (LSTM) networks are typically used to learn seasonal trends (Hochreiter and Schmidhuber 1997). An LSTM network was trained on 24 months of trend (October 2016- October 2018) with a 2-month window size to learn the CCD variations primarily driven by nutrient levels. The model then forecasted the next 14 months of CCD trend, with a RMSE of 19,362.25 cells/ $μ$ L, and a normalized RMSE of 16.8%. Notably, the LSTM model successfully flagged an extreme bloom event in April 2019 (Figure 2). These pilot results show the potential of AI-driven forecasting models in developing early warning systems for CyanoHABs. Integrating additional environmental drivers—such as hydrological, meteorological, and biogeochemical factors will further improve such temporal forecasts (Fournier et al. 2024; Schaeffer et al. 2024). Beyond CyanoHABs, similar AI-driven forecasting frameworks can be extended to other environmental and ecological monitoring challenges, and hold the capability to evolve into comprehensive decision-support tools (Cabay, Piedra and Ayala 2025; Madhavi et al. 2024; Marsetič and Kanjir 2024).

Figure 2.

Evaluation of LSTM-based cyanobacterial cell density trend forecasting: The blue-only trend shows training data (Oct 2016–Dec 2018). The red trend shows the forecast, validated against the blue trend in the background (Jan 2019–Apr 2020).

Turf weed management

Weeds pose a significant challenge to turfgrass management by competing with desirable vegetation for essential resources such as nutrients, water, sunlight, and space (Harker and O’Donovan 2013; McCarty and Murphy 1994). Their presence diminishes the esthetic appeal and functional integrity of turf, impacting recreational spaces, sports fields, and urban landscapes. Conventional weed control strategies rely on broad-spectrum herbicide applications, which, while effective, contribute to environmental contamination, non-target species damage, and escalating economic costs (Monteiro and Santos 2022). The non-uniform distribution of weeds further complicates management, necessitating selective herbicide placement to ensure efficacy while minimizing waste. Manual spot spraying, a common alternative to broadcast herbicide application, is time-consuming and labor-intensive, prompting a need for automated solutions (Partel, Kakarla and Ampatzidis 2019). Integrating AI with precision herbicide application presents an opportunity to reduce pesticide use, enhance weed detection accuracy, and improve the sustainability of turf management (Jin et al. 2022; Partel, Kakarla and Ampatzidis 2019; Vasileiou et al. 2024).

Advancements in remote sensing and AI-driven object detection have enabled automated and precise weed identification, fostering more sustainable agricultural and turfgrass management. Previous research has explored vegetation classification using spectral reflectance properties (Feyaerts and Van Gool 2001) and early AI-based systems such as WEEDER, which applied rule-based learning for species differentiation (Fermanian and Michalski 1989). With the emergence of DL, researchers have leveraged convolutional neural networks (CNNs) for plant species classification, weed detection, and yield prediction (Grinblat et al. 2016; Jin et al. 2022; Mohanty, Hughes and Salathé 2016). In turfgrass-specific applications, CNN-based image classification and identification architectures such as AlexNet (Krizhevsky, Sutskever and Hinton 2012), GoogLeNet (Szegedy et al. 2015), and VGGNet (Simonyan and Zisserman 2014) have been employed to distinguish broadleaf weeds from turfgrass (Prabhu, Niranjan and Rai 2024; Yu et al. 2020, 2019). Despite promising classification accuracy, automating real-time spot spraying remains challenging due to variations in lighting, weather, and weed morphology (Rakhmatulin, Kamilaris and Andreasen 2021).

As a part of this intervention study, Figure 3 shows pilot results, from an experiment aimed at addressing current gaps in automated weed detection and spot spraying. A two-step architecture consisting of image classification using ResNet-50 to identify the weed species, followed by object detection using YOLOv8 to locate the weed was developed for automated spot-spraying. A proximal dataset of over 6,000 RGB images was curated, capturing three broadleaf weed species under diverse environmental conditions. Evaluation showed strong performance for both ResNet-50 with an F1 score of 0.96 and YOLOv8 with a mean average precision of 0.92. Figure 3(B–G) shows select weed image classification and detection results. While some uncertainty persists among different weed species (Figure 3(A)), a total inference time of 6.75 ms makes this architecture suitable for deployment in ground-based robotic sprayer systems to automate selective herbicide application, reducing reliance on manual labor while mitigating pesticide overuse. By integrating AI-driven weed detection with precision spraying technologies, this research advances sustainable turf management, aligning with broader environmental conservation efforts (Ramesh et al. 2017). The adoption of real-time “see-and-spray” technologies holds significant implications for precision agriculture, enabling farmers and turf managers to optimize weed control while minimizing ecological impact (Allmendinger et al. 2022; Gerhards et al. 2022; Liu and Bruch 2020; Westwood et al. 2018).

Figure 3.

(A) Confusion matrix for resNet50-based image classification on turf pictures into different classes of weed species. (B) through (G) show YOLO8-based object detection results for different weed species against a turf background; with (B) as grassy weed, (C) as white clover, (D) as virginia buttonweed, (E) and (F) as common lespedeza, and (G) showing simultaneous detection of multiple weed species.

Urban environment mapping

Accurate mapping of urban environments using remote sensing techniques is crucial for various applications, including urban planning, climate change analysis, and disaster response (Al Shafian and Hu 2024; Li et al. 2024a; Yin et al. 2021). This section discusses two important aspects of urban monitoring: building segmentation, which involves extracting building footprints from remotely sensed imagery; and building height estimation, which determines the vertical dimension of structures. Both contribute significantly to environmental monitoring by providing insights into urban density, land-use changes, and disaster vulnerability (Gupta and Shah 2021; Li et al. 2020b; Liu et al. 2020). Recent advancements in ML and computer vision have led to the development of high-resolution remote sensing (HRS) techniques that enable precise urban mapping (Yuan, Shi and Gu 2021). Traditional methods, which relied on manual interpretation or rule-based approaches, are increasingly being replaced by DL architectures, which improve efficiency and accuracy (Maniyar and Kumar 2021). These advancements help address challenges such as occlusion, spectral similarity, and shadow-induced errors, thereby refining the accuracy of building footprints and height estimates (Cai et al. 2024; Cao and Weng 2024; Wu et al. 2023). The integration of optical, light detection and ranging (LiDAR), and synthetic aperture radar (SAR) data further enhances urban mapping by leveraging multi-source data fusion techniques to overcome individual sensor limitations (Brunner et al. 2010; Raju, Chaudhary and Jha 2014; Shao, Taff and Walsh 2011; Xu et al. 2015; Yan 2024).

Building segmentation has seen remarkable improvements with the advent of DL models such as U-Net and ResNet, which accurately extract structural details from remotely sensed images by learning to discriminate between foreground and background (Diakogiannis et al. 2020; Ronneberger, Fischer and Brox 2015). Figure 4 presents pilot results for building segmentation from satellite imagery, using a multiscale hybrid Res-U-Net based architecture, trained on aerial and satellite images with spatial resolutions varying from 0.2 m to 4 m (Maniyar and Kumar 2021; Maniyar, Kumar and Mai 2025). The Res-U-Net model was applied to a high spatial resolution (1.24 m) Worldview-3 imagery of Chandigarh, India—a city characterized by dense and complex urban patterns. The model achieved an overall accuracy of 86.5%, an intersection over union (IoU) score of 0.78, and an F1-score of 0.80, showing robust performance in complex urban settings with spectral ambiguities. However, challenges persist, particularly in delineating buildings with irregular shapes and extracting structures under shadows (Li et al. 2024a). Despite these limitations, AI-driven building segmentation plays a critical role in broader environmental monitoring applications, such as disaster management and post-disaster damage assessment (Alisjahbana et al. 2024; Gupta and Shah 2021). Rapid and precise segmentation of urban structures allows for immediate evaluation of affected areas after natural disasters like earthquakes and floods, aiding in emergency response efforts (Al Shafian and Hu 2024). Looking ahead, advancements in self-supervised learning (Cong et al. 2022; Noman et al. 2024), multimodal data fusion (Fuller, Millard and Green 2024; Li et al. 2025), and spatial representation learning (Hu et al. 2023b; Liu et al. 2023b; Mai et al. 2023a, 2024b; Wu et al. 2024) hold promise to further improve building segmentation, enabling more accurate and real-time urban analysis (Gui et al. 2024; Li et al. 2024a).

Figure 4.

Multiscale hybrid res-U-net building segmentation results. (A) is the true color Cartosat 2F image. (B) is the binary label mask (ground truth) for building footprint. (C) are the building footprints predicted by the model. (D) is model evaluations with true positives in white, true negatives in black, false positives in red, and false negatives in yellow. Figure reproduced and modified with permission from Maniyar, Kumar and Mai (2025).

Similarly, building height estimation has progressed significantly with the integration of SAR, optical imagery, and ML algorithms. Techniques such as Random Forest and Support Vector Machine regression models are commonly used for large-scale estimation, while DL models like convolutional neural networks (CNNs) are better suited for smaller areas (Frantz et al. 2021; Li et al. 2020a; Mahendra and Seto 2019; Wu et al. 2023). The fusion of SAR backscatter, LiDAR, and optical data has proven effective in refining height estimation, addressing issues like double-bounce scattering and terrain-induced distortions (Franceschetti, Iodice and Riccio 2002; Frolking et al. 2013; Li et al. 2020b; Mahendra and Seto 2019). Figure 5 shows pilot results of building height estimation with 10 m spatial resolution based on ensembled multimodel and colocated multi-temporal data from Sentinel-2 (optical) and Sentinel-1 (radar) by establishing a set of spatial–spectral–temporal feature databases, with an $R^{2}$ of 0.78 (Yan 2024). Despite advancements, challenges remain, particularly in overfitting issues caused by varying urban distribution patterns, accurate estimation of skyscrapers with different building materials and roof designs, the effects of double or multiple bounces in SAR data, and resolving mismatches between building footprints and raster datasets (Huang et al. 2022). Accurate height estimation is essential for understanding urban heat islands, wind flow modeling, and population density assessments, further emphasizing its importance in ERS. The future of building height estimation lies in improving model interpretability, reducing computational costs through feature selection techniques like Shapley additive explanations (SHAP) and permutation importance, and leveraging high-resolution multi-temporal datasets for continuous monitoring (Genuer, Poggi and Tuleau-Malot 2010; Miao and Niu 2016). The combination of AI and Earth observation data is poised to revolutionize urban analytics, making remote sensing a cornerstone of sustainable urban development and climate resilience.

Figure 5.

Heatmap of building heights (in feet) as predicted by the random forest model, overlayed onto the open street map of Des Moines, Iowa, USA. Figure reproduced and modified with permission from Yan (2024).

Carbon flux monitoring

The global carbon cycle is primarily driven by vegetation gross primary productivity (GPP), which represents the total carbon fixed through photosynthesis (Anav et al. 2015; Grace 2004). GPP is commonly estimated using the light use efficiency (LUE) model, where $GPP = APAR \times LUE$ , as formulated by Monteith (1972). Here, APAR (absorbed photosynthetically active radiation) accounts for the total light absorbed by vegetation, while LUE quantifies how efficiently this absorbed light is converted into biomass (Hawman et al. 2021). Wetlands, particularly salt marshes, play a significant role in carbon sequestration, acting as both sinks and sources of carbon due to their highly dynamic environmental conditions (Moomaw et al. 2018). Monitoring carbon fluxes (CF) in these ecosystems is crucial for understanding their response to climate change and natural disturbances. Several models have been proposed to estimate GPP using LUE-based approaches that integrate ground measurements and remote sensing products (Bloomfield et al. 2023). Given the spatiotemporal complexity of salt marshes, remote sensing enables large-scale estimation of LUE by incorporating spectral vegetation indices and environmental variables (Hawman et al. 2021).

Traditional empirical models, while widely applied, rely on static parameterizations and simplified assumptions, often failing to account for the nonlinear and dynamic nature of LUE responses to environmental fluctuations (Bloomfield et al. 2023). In contrast, AI-based models, such as random forest, offer improved predictive capabilities by learning complex interactions from large datasets (Jafarzadeh et al. 2022). The Random Forest model used in this intervention was trained using flux tower data from the Georgia Coastal Ecosystems LTER Flux Tower (US-GCE), available via AmeriFlux (Robinson et al. 2010). Figure 6 presents the validation fit for LUE predictions at the US-GCE site, where the random forest model, trained on daily data of 7 years, achieved an $R^{2}$ of 0.79 and an RMSE of 0.07 gC/MJ. In Figure 6, points clustering tightly around the red dashed reference line indicate higher prediction accuracy, while residuals (green dots) quantify the prediction error. These results highlight the potential of Random Forest based models for generating GPP estimates in salt marshes that lack direct flux tower measurements (Wei et al. 2017). Recent studies have increasingly adopted AI-driven approaches to improve the monitoring of LUE and CF in wetland ecosystems (Cao, Lu and Liu 2024; Jafarzadeh et al. 2022; Sharma and Naik 2024; Sharma et al. 2022).

Figure 6.

Validation fit for random forest based LUE estimation at US-GCE flux tower site. The red dashed line is the 1:1 line, blue dots are validation scatterplot, green dots are corresponding residual values.

Foundation models in ERS

Foundation models refer to large-scale AI models pre-trained on internet-scale datasets in a task-agnostic manner which can be later adapted to various downstream tasks (Bommasani et al. 2021). Examples include language foundation models viz ChatGPT, LLAMA (Touvron et al. 2023), OPT (Zhang et al. 2022a), and Deepseek R1 (Guo et al. 2025), vision foundation models viz SAM (Kirillov et al. 2023) and DINO (Caron et al. 2021), and multimodal foundation models viz CLIP, GPT-4V (Achiam et al. 2023), LLAVA (Liu et al. 2023a), and O1. The paradigm of foundation models has transformed the research practices of many domains including geography and geoscience (Mai 2024; Mai et al. 2024a; Xie et al. 2023).

Foundation models that integrate vision and language modalities are transforming ERS by enabling more comprehensive and nuanced data analysis (Huo et al. 2025). Traditionally, remote sensing has relied on task-specific ML models tailored for specialized objectives like land cover classification or vegetation health assessment (Ma et al. 2019). In contrast, vision-language foundation models, pre-trained on extensive datasets in a task-agnostic manner, offer a generalized understanding of visual and textual relationships, enhancing adaptability for various remote sensing applications (Mai et al. 2024a; Mall et al. 2023; Manvi et al. 2024). For instance, vision-language models can process both image and text data to provide contextual awareness, allowing AI systems not only to classify images but also to generate natural language explanations of their observations (Lu et al. 2024; Mall et al. 2023). This capability is crucial in environmental monitoring, where interpretability is as important as accuracy. Recent advancements have demonstrated that such models can be adapted for remote sensing tasks using zero-shot learning (Li et al. 2024b; Wu and Osco 2023; Zhang et al. 2024), few-shot learning (Hu et al. 2023a; Mai et al. 2024a; Osco et al. 2023), or fine-tuning (Ferreira et al. 2025; Liu et al. 2024a), which bypass the need for extensive labeled datasets.

Other than simply adopting the general-purpose foundation models to remote sensing applications, massive efforts have also been devoted to developing remote sensing foundation models (RSFMs) such as Segment Any Change (Zheng et al. 2024), SatCLIP (Klemmer et al. 2023), ClimaX (Nguyen et al. 2023), SatMAE (Cong et al. 2022), SatMAE++ (Noman et al. 2024), CROMA (Fuller, Millard and Green 2024), SpectralGPT (Hong et al. 2024b), SkySense (Guo et al. 2024), GeoChat+ (Kuckreja et al. 2024) etc. These RSFMs show promising model performance and generalizability across diverse remote sensing tasks, showing their potential on many RS tasks which usually suffer from low performance due to low labeled data availability (Chang et al. 2024; Shoaib, Emara and Zhao 2023).

AI tools and resources for ERS

The integration of AI into ERS has been significantly enhanced by the evolution of R and Python packages, which now increasingly offer robust geospatial components. This development has democratized access to sophisticated geospatial analysis, enabling researchers from various disciplines to apply AI techniques without requiring extensive programming expertise. Moreover, the emergence of domain-specific AI tools has further facilitated the application of ML and DL in environmental studies, making these technologies more accessible to non-AI specialists. Table 1 lists some of the most popular tools and resources commonly used for AI-based analysis in ERS, identifying the ones used in this paper.

Table 1.

A non-exhaustive list of popular and commonly used AI tools and resources used in ERS. Each tool/resource is hyperlinked to its online source.

Tool/Resource Name	Description	Used in this paper for	Source Availability
Google Earth Engine	Cloud-based AI	CyanoHAB trend extraction	Partially Open-Source
Pix4D	Data annotation	Annotating weed images	Closed-Source
ArcGIS	ML/DL tools	Shape index for Building height estimation	Closed-Source
EOfactory	Cloud-based AI	–	Closed-Source
Google Colab (Python)	Cloud-based AI	CyanoHAB trend forecasting (LSTM)	Partially Open-Source
Scikit-learn (Python)	ML Algorithms	SSS, Building height estimation, LUE (Random Forest)	Fully Open-Source
TensorFlow (Python)	ML/DL Algorithms	Weed image classification	Fully Open-Source
PyTorch (Python)	ML/DL Algorithms	Building Segmentation (Hybrid Res-U-Net)	Fully Open-Source
YOLO (Python)	Object detection	Weed detection	Fully Open-Source
GeoTorch (Python)	Geo-AI Algorithms	–	Fully Open-Source
caret (R)	ML Algorithms	Building height estimation	Fully Open-Source
tidymodels (R)	ML Algorithms	–	Fully Open-Source
mlr3 (R)	ML Algorithms	–	Fully Open-Source
e1071 (R)	ML Algorithms	Building height estimation	Fully Open-Source

Key considerations

AI in ERS offers unprecedented capabilities for understanding, monitoring and managing Earth’s ecosystems. However, this advancement brings forth significant ethical concerns and technological limitations that necessitate careful consideration from the ERS community.

Ethical concerns in AI-driven ERS

One pressing issue is the potential generation of synthetic or manipulated satellite imagery, often referred to as deepfake geography (Zhao et al. 2021). While ancient map-making also showed a version of deepfake geography in the form of fake locations or inaccuracies, such falsified data can mislead policymakers, scientists, and the public, leading to erroneous conclusions and potentially harmful decisions (Abedi and Rajabifard 2024; Romano 2025). In the context of ERS, similar manipulations could distort environmental assessments and undermine trust in geospatial data (McKenzie, Zhang and Gambs 2023). Another concern is the inadvertent invasion of privacy through AI-enhanced remote sensing. High-resolution satellite imagery, combined with AI analysis, can inadvertently capture sensitive information, leading to potential misuse or unauthorized surveillance (Paolanti et al. 2024). For example, researchers using AI and camera imagery in wildlife surveys have unintentionally captured images of poachers and other illegal activities, raising ethical questions about privacy and consent (Jamrin 2024).

Additionally, maintaining transparency in AI-driven ERS applications is crucial to maintaining trust and scientific credibility. Explainable AI (XAI) plays a vital role in this context by enabling researchers and decision-makers to understand, interpret, and validate AI-generated insights. XAI refers to a set of techniques that make AI models more interpretable, providing insights into how decisions are made rather than treating models as opaque black boxes (Arrieta et al. 2020; Gunning et al. 2021). Recently, several efforts have been made to incorporate the geographic aspect into XAI model development such as Geoshapley (Li 2024) and SpaCE (Zhang et al. 2025). In ERS, where AI-driven analyses are used for tasks such as land cover classification, climate modeling, and pollution monitoring, the lack of interpretability can pose risks to decision-making and policy development (Gevaert et al. 2021). For example, an AI model predicting harmful algal blooms must not only provide results but also explain which spectral features or environmental parameters contributed to its decision (Karpatne, Jia and Kumar 2024; Veerman et al. 2024; Weber et al. 2020). Without explainability, AI systems may reinforce biases, misinterpret environmental patterns, or generate misleading conclusions, ultimately affecting resource management and mitigation strategies (Arashpour 2023; Gevaert et al. 2021). Techniques such as feature attribution, saliency maps, and attention mechanisms help reveal how AI models derive their predictions, allowing scientists to identify errors and refine models for greater reliability (Molnar 2020). Furthermore, incorporating XAI enhances accountability, as stakeholders can independently verify AI-generated outputs and ensure that models align with ecological and societal values. Promoting explainability in AI-based ERS applications is a fundamental step toward responsible AI adoption, fostering trust and ensuring that AI augments, rather than obscures environmental decision-making.

Challenges and limitations of AI in ERS

Despite its advantages, AI in ERS faces several challenges and limitations. One major concern is the high computational cost associated with training and deploying large AI models or foundation models. Training these learning models requires significant computing power and electricity, raising concerns about sustainability and carbon footprint (Shi et al. 2023; Strubell, Ganesh and McCallum 2020). Computational costs can be a limiting factor for many research institutions and organizations. Advanced AI models often require specialized hardware such as GPUs or TPUs, which may not be readily available in resource-constrained settings (Cottier et al. 2024). The cost of acquiring, processing, and storing large-scale environmental datasets further adds to the financial burden, potentially restricting access to cutting-edge AI solutions in ERS (Ghamisi et al. 2024). This necessitates the need for re-usable AI (Moradi and Uta 2021).

Additionally, many AI applications in ERS rely on large, high-quality labeled datasets, but environmental data availability remains a challenge. While satellite imagery is abundant, labeled datasets required for supervised learning are often scarce or inconsistent across different geographic regions (Tuia et al. 2022). This data limitation hinders model generalizability and transferability, especially in underrepresented or rapidly changing environments. Model interpretability also remains a critical issue. While XAI techniques are advancing, many complex AI models, such as deep neural networks and foundation models, still function as black boxes, making it difficult for researchers and policymakers to fully understand their decision-making processes (Arrieta et al. 2020; Molnar 2020). In ERS applications where AI-driven insights influence conservation policies or disaster response, a lack of interpretability could reduce stakeholder trust and lead to unintended consequences (Bruynseels, Asveld and van den Hoven 2025).

Responsible AI use in ERS

To address these ethical challenges, the research community must adopt a framework of responsible AI practices in ERS. Based on recent comprehensive reviews and interventions highlighting the importance of ethical AI practices in ERS (Bennett et al. 2024; Ghamisi et al. 2024; McKenzie, Zhang and Gambs 2023; Paolanti et al. 2024), we propose the following considerations:

Implementing robust verification protocols to help detect and mitigate the impact of synthetic or manipulated data.

Safeguarding privacy and adhering to guidelines that protect individuals’ and communities’ rights when collecting and analyzing geospatial data.

Maintaining transparency in AI methodologies to foster trust and allowing for independent verification of findings.

Engaging with local communities and stakeholders to ensure that AI applications in ERS align with societal values and ethical standards.

By proactively implementing these ethical considerations through approaches like federated learning (Kairouz et al. 2021; Zhang et al. 2021) interpretable AI (Du, Liu and Hu 2019; Molnar 2020), and re-usable AI (Moradi and Uta 2021), the ERS community can benefit from AI while minimizing potential harms. This ensures that technological advancements contribute positively to environmental stewardship and societal well-being.

Conclusion

This technical intervention highlights the expanding role of state-of-the-art AI techniques in multiple domains of ERS. Specifically, we discussed AI applications in image classification, object detection and segmentation, time-series forecasting, and bio-optical modeling. Additionally, we reviewed the emergence of vision- and language-based foundation models and their potential to revolutionize ERS workflows. The increasing adoption of AI has significantly improved the performance, scalability, and accessibility of critical ERS applications, like climate monitoring and rapid response to extreme events and disasters. The current state of AI-driven research in ERS stands at the intersection of multiple disciplines, creating important new opportunities for interdisciplinary collaboration between computer scientists, remote sensing experts, ecologists, and policymakers. By integrating domain expertise from diverse fields, researchers can develop AI solutions that are both scientifically robust and directly applicable to pressing environmental challenges.

Beyond these immediate applications, AI is expanding the scope of remotely sensed data by enabling the estimation of environmental or biophysical parameters and phenomena that are not optically evident, that is, do not have apparent spectral characteristics. For example, ML models have successfully estimated dynamic and unapparent parameters such as sea surface salinity (discussed in this article), belowground biomass (O’Connell et al. 2021; Runion et al. 2024) and soil organic carbon (Sharma et al. 2022), by learning complex relationships between remotely sensed spectral data, aboveground or surface optically active proxies, co-related environmental variables and underlying in situ measurements. These advancements continue to broaden the future applicability of AI in environmental and ecological research, allowing scientists to: (a) derive new insights into ecosystem dynamics and biogeochemical processes, and (b) develop stronger linkages between large-scale long-term climate change stressors to ecosystem response; for example, sea level rise and belowground biomass in wetlands (O’Connell et al. 2021; Runion et al. 2024), and temperature extremes and CyanoHABs phenology in fresh waterbodies (Ho and Michalak 2020).

The rapid expansion of AI in ERS also raises ethical concerns, including issues of data integrity, model transparency, and potential biases. This review outlines key ethical challenges and provides considerations for responsible AI implementation in ERS to ensure trustworthy and equitable applications. While AI holds transformative potential, its application in ERS remains in its early stages, with several fundamental challenges and possibilities open to further investigation. Among these, crucial research directions relate to the generalizability of AI models in ERS and their explainability.

Footnotes

Acknowledgments

The work on Sea Surfaca Salinity was funded through a GA Space Consotrium Grant (NIF project #AWD-001629-G24). The work on CyanoHAB forecasting was mainly supported through a NASA FINESST Grant (Award #80NSSC24K0068) and a NASA ESROGSS Grant (Award #80NSSC23K1258). The authors thank Georgia Coastal Ecosystems LTER for providing in situ salinity data and flux tower data. The authors thank Mississippi State University’s Geosystems Research Institute for providing turf weed data, and their High Performance Computing Collaboratory for providing computational resources to support the work on turf weed management. The authors thank Indian Institute of Remote Sensing, ISRO for providing access to high resolution Cartosat-2F imagery as well as high performance computational resources to support the building segmentation work. Finally, authors thank the reviewers for their review and comments, which helped in improving the quality of this manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Chintan B. Maniyar

Xin Yan

Gengchen Mai

Sathishkumar Samiappan

Abhishek Kumar

Deepak R. Mishra

References

Abedi

Rajabifard

. 2024. “Ethics, Privacy and the Perils of ‘Deepfake Geography’.” https://pursuit.unimelb.edu.au/articles/Ethic. Accessed 17 Febreuary 2025.

Achiam

Adler

Agarwal

Ahmad

Akkaya

Aleman

F. L.

Almeida

Altenschmidt

Altman

Anadkat

, et al. 2023. “Gpt-4 Technical Report.” arXiv preprint arXiv:2303.08774.

Adger

W. N.

Fransen

Safra de Campos

Clark

W. C.

. 2024. “Scientific Frontiers on Migration and Sustainability.” Proceedings of the National Academy of Sciences 121 (3): e2321325121. DOI: https://doi.org/10.1073/pnas.2321325121.

Alexey

2020. “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.” arXiv preprint arXiv: 2010.11929.

Alisjahbana

Zhang

, et al. 2024. “Deepdamagenet: A Two-Step Deep-Learning Model for Multi-Disaster Building Damage Segmentation and Classification Using Satellite Imagery.” arXiv preprint arXiv:2405.04800.

Allmendinger

Spaeth

Saile

Peteinatos

G. G.

Gerhards

. 2022. “Precision Chemical Weed Management Strategies: A Review and a Design of a New CNN-Based Modular Spot Sprayer.” Agronomy 12 (7): 1620. DOI: https://doi.org/10.3390/agronomy12071620.

Almuhtaram

Kibuye

F. A.

Ajjampur

Glover

C. M.

Hofmann

Gaget

Owen

Wert

E. C.

Zamyadi

. 2021. “State of Knowledge on Early Warning Tools for Cyanobacteria Detection.” Ecological Indicators 133: 108442. DOI: https://doi.org/10.1016/j.ecolind.2021.108442.

Al Shafian

. 2024. “Integrating Machine Learning and Remote Sensing in Disaster Management: A Decadal Review of Post-Disaster Building Damage Assessment.” Buildings 14 (8): 2344. DOI: https://doi.org/10.3390/buildings14082344.

Anav

Friedlingstein

Beer

Ciais

Harper

Jones

Murray-Tortarolo

Papale

Parazoo

N. C.

Peylin

, et al. 2015. “Spatiotemporal Patterns of Terrestrial Gross Primary Production: A Review.” Reviews of Geophysics 53 (3): 785–818. DOI: https://doi.org/10.1002/2015RG000483.

10.

Arashpour

2023. “AI Explainability Framework for Environmental Management Research.” Journal of Environmental Management 342: 118149. DOI: https://doi.org/10.1016/j.jenvman.2023.118149.

11.

Arnab

Dehghani

Heigold

Sun

Lučić

Schmid

. 2021. “Vivit: A Video Vision Transformer.” In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6836–6846. DOI: 10.1109/ICCV48922.2021.00676.

12.

Arrieta

A. B.

Díaz-Rodríguez

Del Ser

Bennetot

Tabik

Barbado

García

Gil-López

Molina

Benjamins

, et al. 2020. “Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges Toward Responsible Ai.” Information Fusion 58: 82–115. DOI: https://doi.org/10.1016/j.inffus.2019.12.012.

13.

Avtar

Komolafe

A. A.

Kouser

Singh

Yunus

A. P.

Dou

Kumar

Gupta

R. D.

Johnson

B. A.

Minh

H. V. T.

, et al. 2020. “Assessing Sustainable Development Prospects Through Remote Sensing: A Review.” Remote Sensing Applications: Society and Environment 20: 100402. DOI: https://doi.org/10.1016/j.rsase.2020.100402.

14.

Bengio

2013. “Deep Learning of Representations: Looking Forward.” In International Conference on Statistical Language and Speech Processing. Springer, 1–37. DOI: 10.1007/978-3-642-39593-2_1.

15.

Bennett

M. M.

Gleason

C. J.

Tellman

Leon

L. F. A.

Friedrich

H. K.

Ovienmhada

Mathews

A. J.

. 2024. “Bringing Satellites Down to Earth: Six Steps to More Ethical Remote Sensing.” Global Environmental Change Advances 2: 100003. DOI: https://doi.org/10.1016/j.gecadv.2023.100003.

16.

Bloomfield

K. J.

Stocker

B. D.

Keenan

T. F.

Prentice

I. C.

. 2023. “Environmental Controls on the Light Use Efficiency of Terrestrial Gross Primary Production.” Global Change Biology 29 (4): 1037–53. DOI: https://doi.org/10.1111/gcb.16511.

17.

Bommasani

Hudson

D. A.

Adeli

Altman

Arora

von Arx

Bernstein

M. S.

Bohg

Bosselut

Brunskill

, et al. 2021. “On the Opportunities and Risks of Foundation Models.” arXiv preprint arXiv:2108.07258.

18.

Borovskaya

Krivoguz

Chernyi

Kozhurin

Khorosheltseva

Zinchenko

. 2022. “Surface Water Salinity Evaluation and Identification for Using Remote Sensing Data and Machine Learning Approach.” Journal of Marine Science and Engineering 10 (2): 257. DOI: https://doi.org/10.3390/jmse10020257.

19.

Boutin

Reul

Köhler

Martin

Catany

Guimbard

Rouffi

Vergely

J. L.

Arias

Chakroun

, et al. 2021. “Satellite-Based Sea Surface Salinity Designed for Ocean and Climate Studies.” Journal of Geophysical Research: Oceans 126 (11): e2021JC017676. DOI: https://doi.org/10.1029/2021JC017676.

20.

Breiman

2001. “Random Forests.” Machine Learning 45: 5–32.

21.

Brunner

Lemoine

Bruzzone

Greidanus

. 2010. “Building Height Retrieval From Vhr Sar Imagery Based on An Iterative Simulation and Matching Technique.” IEEE Transactions on Geoscience and Remote Sensing 48 (3): 1487–504. DOI: https://doi.org/10.1109/TGRS.2009.2031910.

22.

Bruynseels

Asveld

van den Hoven

. 2025. “Foundation Models for Research: A Matter of Trust?” Artificial Intelligence in the Life Sciences 100126. https://doi.org/10.1016/j.ailsci.2025.100126.

23.

Brynjarsdottir

OHagan

. 2014. “Learning About Physical Parameters: The Importance of Model Discrepancy.” Inverse Problems 30 (11): 114007. DOI: https://doi.org/10.1088/0266-5611/30/11/114007.

24.

Cabay

M. J. C.

Piedra

J. A.

Ayala

R. M.

. 2025. “Modeling Temperature in the Ecuadorian Paramo Through Deep Learning.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensinghttps://doi.org/10.1109/JSTARS.2025.3536878.

25.

Cai

Guo

Xiao

Guo

Zhang

Song

. 2024. “Automated Building Height Estimation Using Ice, Cloud, and Land Elevation Satellite 2 Light Detection and Ranging Data and Building Footprints.” Remote Sensing 16 (2): https://doi.org/10.3390/rs16020263.

26.

Campbell

A. D.

Fatoyinbo

Charles

S. P.

Bourgeau-Chavez

L. L.

Goes

Gomes

Halabisky

Holmquist

Lohrenz

Mitchell

, et al. 2022. “A Review of Carbon Monitoring in Wet Carbon Systems Using Remote Sensing.” Environmental Research Letters 17 (2): 025009. DOI: https://doi.org/10.1088/1748-9326/ac4d4d.

27.

Cao

Liu

. 2024. “Research on Innovative Design Mode of Wetland Landscape Ecological Restoration in the Era of Artificial Intelligence.” Applied Mathematics and Nonlinear Sciences 9 (1): https://doi.org/10.2478/amns-2024-0188.

28.

Cao

Weng

. 2024. “A Deep Learning-Based Super-Resolution Method for Building Height Estimation at 2.5 m Spatial Resolution in the Northern Hemisphere.” Remote Sensing of Environment 310: 114241. DOI: https://doi.org/10.1016/j.rse.2024.114241.

29.

Carmichael

W. W.

2001. “Health Effects of Toxin-Producing Cyanobacteria: The Cyanohabs.” Human and Ecological Risk Assessment: An International Journal 7 (5): 1393–407.

30.

Caron

Touvron

Misra

Jégou

Mairal

Bojanowski

Joulin

. 2021. “Emerging Properties in Self-Supervised Vision Transformers.” In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9650–9660. DOI: 10.1109/ICCV48922.2021.00951.

31.

Chang

Y. C.

Stewart

A. J.

Bastani

Wolters

Kannan

Huber

G. R.

Wang

Banerjee

. 2024. “On the Generalizability of Foundation Models for Crop Type Mapping.” arXiv preprint arXiv:2409.09451.

32.

Chen

Zhang

Liu

Osman

A. I.

Farghali

Hua

Al-Fatesh

Ihara

Rooney

D. W.

, et al. 2023. “Artificial Intelligence-Based Solutions for Climate Change: A Review.” Environmental Chemistry Letters 21 (5): 2525–57. DOI: https://doi.org/10.1007/s10311-023-01617-y.

33.

Chen

. 2017. “Estimating Sea Surface Salinity in the Northern Gulf of Mexico From Satellite Ocean Color Measurements.” Remote Sensing of Environment 201: 115–32. DOI: https://doi.org/10.1016/j.rse.2017.09.004.

34.

Chuvieco

2020. Fundamentals of Satellite Remote Sensing: An Environmental Approach. CRC Press. DOI: DOI: https://doi.org/10.1201/9780429506482.

35.

Cong

Khanna

Meng

Liu

Rozi

Burke

Lobell

Ermon

. 2022. “Satmae: Pre-Training Transformers for Temporal and Multi-Spectral Satellite Imagery.” Advances in Neural Information Processing Systems 35: 197–211.

36.

Cottier

Rahman

Fattorini

Maslej

Besiroglu

Owen

. 2024. “The Rising Costs of Training Frontier AI Models.” arXiv preprint arXiv:2405.21015.

37.

Dev

P. J.

Sukenik

Mishra

D. R.

Ostrovsky

. 2022. “Cyanobacterial Pigment Concentrations in Inland Waters: Novel Semi-Analytical Algorithms for Multi- and Hyperspectral Remote Sensing Data.” Science of the Total Environment 805: 150423. DOI: https://doi.org/10.1016/j.scitotenv.2021.150423.

38.

Díaz

Settele

Brondízio

E. S.

Ngo

H. T.

Agard

Arneth

Balvanera

Brauman

K. A.

Butchart

S. H.

Chan

K. M.

, et al. 2019. “Pervasive Human-Driven Decline of Life on Earth Points to the Need for Transformative Change.” Science (New York, N.Y.) 366 (1): eaax3100. DOI: https://doi.org/10.1126/science.aax3100.

39.

Diakogiannis

F. I.

Waldner

Caccetta

. 2020. “Resunet-a: A Deep Learning Framework for Semantic Segmentation of Remotely Sensed Data.” ISPRS Journal of Photogrammetry and Remote Sensing 162: 94–114. DOI: https://doi.org/10.1016/j.isprsjprs.2020.01.013.

40.

Liu

. 2019. “Techniques for Interpretable Machine Learning.” Communications of the ACM 63 (1): 68–77. DOI: https://doi.org/10.1145/3359786.

41.

Feng

Wang

Hou

Qin

Kuster

Chen

Paerl

H. W.

Zheng

. 2024. “Harmful Algal Blooms in Inland Waters.” Nature Reviews Earth and Environment 5 (9): 631–44. DOI: https://doi.org/10.1038/s43017-024-00578-2.

42.

Fermanian

T. W.

Michalski

R. S.

. 1989. “Weeder: An Advisory System for the Identification of Grasses in Turf.” Agronomy Journal 81 (2): 312–6. DOI: https://doi.org/10.2134/agronj1989.00021962008100020034x.

43.

Ferreira

L. B.

Martins

V. S.

Aires

U. R.

Wijewardane

Zhang

Samiappan

. 2025. “Fieldseg: A Scalable Agricultural Field Extraction Framework Based on the Segment Anything Model and 10-m Sentinel-2 Imagery.” Computers and Electronics in Agriculture 232: 110086. DOI: https://doi.org/10.1016/j.compag.2025.110086.

44.

Feyaerts

Van Gool

. 2001. “Multi-Spectral Vision System for Weed Detection.” Pattern Recognition Letters 22 (6–7): 667–74. DOI: https://doi.org/10.1016/S0167-8655(01)00006-X.

45.

Fletcher

Ripple

W. J.

Newsome

Barnard

Beamer

Behl

Bowen

Cooney

Crist

Field

, et al. 2024. “Earth At Risk: An Urgent Call to End the Age of Destruction and Forge a Just and Sustainable Future.” PNAS Nexus 3 (4): 106. DOI: https://doi.org/10.1093/pnasnexus/pgae106.

46.

Fournier

Fernandez-Fernandez

Cirés

López-Orozco

J. A.

Besada-Portas

Quesada

. 2024. “Lstm Networks Provide Efficient Cyanobacterial Blooms Forecasting Even with Incomplete Spatio-Temporal Data.” Water Research 267: 122553. DOI: https://doi.org/10.1016/j.watres.2024.122553.

47.

Fournier

Lee

Tang

Steele

Olmedo

. 2019. “Evaluation and Intercomparison of Smos, Aquarius, and Smap Sea Surface Salinity Products in the Arctic Ocean.” Remote Sensing 11 (24): 3043. DOI: https://doi.org/10.3390/rs11243043.

48.

Franceschetti

Iodice

Riccio

. 2002. “A Canonical Problem in Electromagnetic Backscattering From Buildings.” IEEE Transactions on Geoscience and Remote Sensing 40 (8): 1787–801. DOI: https://doi.org/10.1109/TGRS.2002.802459.

49.

Frantz

Schug

Okujeni

Navacchi

Wagner

van der Linden

Hostert

. 2021. “National-Scale Mapping of Building Height Using Sentinel-1 and Sentinel-2 Time Series.” Remote Sensing of Environment 252: 112128. DOI: https://doi.org/10.1016/j.rse.2020.112128.

50.

Frolking

Milliman

Seto

K. C.

Friedl

M. A.

. 2013. “A Global Fingerprint of Macro-Scale Changes in Urban Structure From 1999 to 2009.” Environmental Research Letters 8 (2): 024004. DOI: https://doi.org/10.1088/1748-9326/8/2/024004.

51.

Zhang

Shi

Liu

. 2021. “Sea Surface Salinity Estimation and Spatial-Temporal Heterogeneity Analysis in the Gulf of Mexico.” Remote Sensing 13 (5). DOI: https://doi.org/10.3390/rs13050881. https://www.mdpi.com/2072-4292/13/5/881 .

52.

Fuller

Millard

Green

2024. “Croma: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoders.” Advances in Neural Information Processing Systems 36: 5506–5538. DOI: https://doi.org/10.48550/arXiv.2311.00566.

53.

Gaur

Pant

Jalal

A. S.

. 2023. “Comparative Assessment of Artificial Intelligence (AI)-Based Algorithms for Detection of Harmful Bloom-Forming Algae: An Eco-Environmental Approach Toward Sustainability.” Applied Water Science 13 (5): 115. DOI: https://doi.org/10.1007/s13201-023-01919-0.

54.

Genuer

Poggi

J. M.

Tuleau-Malot

. 2010. “Variable Selection Using Random Forests.” Pattern Recognition Letters 31 (14): 2225–36. DOI: https://doi.org/10.1016/j.patrec.2010.03.014.

55.

Gerhards

Andujar Sanchez

Hamouz

Peteinatos

G. G.

Christensen

Fernandez-Quintanilla

. 2022. “Advances in Site-Specific Weed Management in Agriculture: A Review.” Weed Research 62 (2): 123–33.

56.

Gevaert

C. M.

Carman

Rosman

Georgiadou

Soden

. 2021. “Fairness and Accountability of Ai in Disaster Risk Management: Opportunities and Challenges.” Patterns 2 (11). DOI: https://doi.org/10.1016/j.patter.2021.100363.

57.

Ghamisi

Marinoni

Gevaert

C. M.

Persello

Selvakumaran

Girotto

Horton

B. P.

Rufin

Hostert

, et al. 2024. “Responsible AI for Earth Observation.” arXiv preprint arXiv:2405.20868.

58.

Grace

2004. “Understanding and Managing the Global Carbon Cycle.” Journal of Ecology 92 (2): 189–202. DOI: https://doi.org/10.1111/j.0022-0477.2004.00874.x.

59.

Grinblat

G. L.

Uzal

L. C.

Larese

M. G.

Granitto

P. M.

. 2016. “Deep Learning for Plant Identification Using Vein Morphological Patterns.” Computers and Electronics in Agriculture 127: 418–24. DOI: https://doi.org/10.1016/j.compag.2016.07.003.

60.

Gui

Song

Qin

Tang

. 2024. “Remote Sensing Object Detection in the Deep Learning Era: A Review.” Remote Sensing 16 (2): 327.

61.

Gunning

Vorm

Wang

Turek

. 2021. “Darpas Explainable AI (XAI) Program: A Retrospective.” Authorea Preprints.

62.

Guo

Yang

Zhang

Song

Zhang

Zhu

Wang

, et al. 2025. “DeepSeek-R1: Incentivizing Reasoning Capability in LLMS Via Reinforcement Learning.” arXiv preprint arXiv:2501.12948.

63.

Guo

Lao

Dang

Zhang

Zhong

Huang

, et al. 2024. “Skysense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 27672–27683. DOI: 10.1109/CVPR52733.2024.02613.

64.

Gupta

Shah

. 2021. “Rescuenet: Joint Building Segmentation and Damage Assessment from Satellite Imagery.” In 2020 25th International Conference on Pattern Recognition (ICPR). IEEE, 4405–4411. DOI: 10.1109/ICPR48806.2021.9412295.

65.

Hamedianfar

Mohamedou

Kangas

Vauhkonen

. 2022. “Deep Learning for Forest Inventory and Planning: A Critical Review on the Remote Sensing Approaches so Far and Prospects for Further Applications.” Forestry 95 (4): 451–65. DOI: https://doi.org/10.1093/forestry/cpac002.

66.

Harker

K. N.

O’Donovan

J. T.

. 2013. “Recent Weed Control, Weed Management, and Integrated Weed Management.” Weed Technology 27 (1): 1–11. DOI: https://doi.org/10.1614/WT-D-12-00109.1.

67.

Hawman

P. A.

Mishra

D. R.

O’Connell

J. L.

Cotten

D. L.

Narron

C. R.

Mao

. 2021. “Salt Marsh Light Use Efficiency is Driven by Environmental Gradients and Species-Specific Physiology and Morphology.” Journal of Geophysical Research: Biogeosciences 126 (5): e2020JG006213. DOI: https://doi.org/10.1029/2020JG006213.

68.

Hirsch

S. L.

2020. “Making Global Problems Local: Understanding Large-Scale Environmental Issues Using Ethnographic Methods.” Science, Technology, & Human Values 45 (5): 963–73. DOI: https://doi.org/10.1177/0162243920903767.

69.

J. C.

Michalak

A. M.

. 2020. “Exploring Temperature and Precipitation Impacts on Harmful Algal Blooms Across Continental Us Lakes.” Limnology and Oceanography 65 (5): 992–1009. DOI: https://doi.org/10.1002/lno.11365.

70.

Hochreiter

Schmidhuber

. 1997. “Long Short-Term Memory.” Neural Computation 9 (8): 1735–80. DOI: https://doi.org/10.1162/neco.1997.9.8.1735.

71.

Hong

Zhang

Yokoya

Benediktsson

A. J.

Chanussot

. 2024a. “Multimodal Artificial Intelligence Foundation Models: Unleashing the Power of Remote Sensing Big Data in Earth Observation.” Innovation 2 (1): 100055. DOI: https://doi.org/10.59717/j.xinn-geo.2024.100055.

72.

Hong

Zhang

Yao

Yokoya

Ghamisi

Jia

, et al. 2024b. “Spectralgpt: Spectral Remote Sensing Foundation Model.” IEEE Transactions on Pattern Analysis and Machine Intelligence. DOI: https://doi.org/10.1109/TPAMI.2024.3362475.

73.

Mai

Cundy

Choi

Lao

Liu

Lakhanpal

Zhou

R. Z.

Joseph

. 2023a. “Geo-Knowledge-Guided Gpt Models Improve the Extraction of Location Descriptions From Disaster-Related Social Media Messages.” International Journal of Geographical Information Science 37 (11): 2289–318. DOI: https://doi.org/10.1080/13658816.2023.2266495.

74.

Wang

Huang

Liu

. 2023b. “Polybuilding: Polygon Transformer for Building Extraction.” ISPRS Journal of Photogrammetry and Remote Sensing 199: 15–27. DOI: https://doi.org/10.1016/j.isprsjprs.2023.03.021.

75.

Huang

Chen

Liu

Wang

Liu

Clinton

Gong

. 2022. “Estimating Building Height in China From Alos Aw3d30.” ISPRS Journal of Photogrammetry and Remote Sensing 185: 146–57. DOI: https://doi.org/10.1016/j.isprsjprs.2022.01.022.

76.

Huisman

Codd

G. A.

Paerl

H. W.

Ibelings

B. W.

Verspagen

J. M.

Visser

P. M.

. 2018. “Cyanobacterial Blooms.” Nature Reviews Microbiology 16 (8): 471–83. DOI: https://doi.org/10.1038/s41579-018-0040-1.

77.

Huntingford

Jeffers

E. S.

Bonsall

M. B.

Christensen

H. M.

Lees

Yang

. 2019. “Machine Learning and Artificial Intelligence to Aid Climate Change Research and Preparedness.” Environmental Research Letters 14 (12): 124007. DOI: https://doi.org/10.1088/1748-9326/ab4e55.

78.

Huo

Chen

Zhang

Wang

Yan

Shen

Hong

Fang

Wang

. 2025. “When Remote Sensing Meets Foundation Model: A Survey and Beyond.” Remote Sensing 17 (2). DOI: https://doi.org/10.3390/rs17020179.

79.

Jafarzadeh

Mahdianpari

Gill

E. W.

Brisco

Mohammadimanesh

. 2022. “Remote Sensing and Machine Learning Tools to Support Wetland Monitoring: A Meta-Analysis of Three Decades of Research.” Remote Sensing 14 (23): 6104. DOI: https://doi.org/10.3390/rs14236104.

80.

Jamrin

2024. “Ethical Evaluation of Remote Sensing in Conservation.” https://ethicalgeo.org/ethical-evaluation-of-remote-sensing-in-conservation. Accessed 17 February 2025.

81.

Jiang

Brereton

Beckler

Moore

Brewton

R. A.

Lapointe

B. E.

McFarland

M. N.

. 2025. “Modeling Water Quality and Cyanobacteria Blooms in Lake Okeechobee: I. Model Descriptions, Seasonal Cycles, and Spatial Patterns.” Ecological Modelling 502: 111018. DOI: https://doi.org/10.1016/j.ecolmodel.2025.111018.

82.

Jin

Liu

Chen

. 2022. “Deep Learning-Based Weed Detection in Turf: A Review.” Agronomy 12 (12): 3051. DOI: https://doi.org/10.3390/agronomy12123051.

83.

Kairouz

McMahan

H. B.

Avent

Bellet

Bennis

Bhagoji

A. N.

Bonawitz

Charles

Cormode

Cummings

, et al. 2021. “Advances and Open Problems in Federated Learning.” Foundations and Trends® in Machine Learning 14 (1–2): 1–210. DOI: https://doi.org/10.1561/2200000083.

84.

Karpatne

Jia

Kumar

. 2024. “Knowledge-Guided Machine Learning: Current Trends and Future Prospects.” arXiv preprint arXiv:2403.15989.

85.

Kesavakumar

Shanmugam

Venkatesan

. 2022. “Enhanced Sea Surface Salinity Estimates Using Machine-Learning Algorithm With Smap and High-Resolution Buoy Data.” IEEE Access 10: 74304–17. DOI: https://doi.org/10.1109/ACCESS.2022.3189784.

86.

Kim

Y. J.

Han

Jang

Sung

. 2023. “Remote Sensing of Sea Surface Salinity: Challenges and Research Directions.” GIScience & Remote Sensing 60 (1): 2166377. DOI: https://doi.org/10.1080/15481603.2023.2166377.

87.

Kirillov

Mintun

Ravi

Mao

Rolland

Gustafson

Xiao

Whitehead

Berg

A. C.

W. Y.

, et al. 2023. “Segment Anything.” In Proceedings of the IEEE/CVF International Conference on Computer Vision, 4015–4026. DOI: 10.1109/ICCV51070.2023.00371.

88.

Klemmer

Rolf

Robinson

Mackey

Rußwurm

. 2023. “Satclip: Global, General-Purpose Location Embeddings with Satellite Imagery.” arXiv preprint arXiv:2311.17179.

89.

Krizhevsky

Sutskever

Hinton

G. E.

. 2012. “Imagenet Classification with Deep Convolutional Neural Networks.” Advances in Neural Information Processing Systems 25. DOI: https://doi.org/10.1145/3065386.

90.

Kuckreja

Danish

M. S.

Naseer

Das

Khan

F. S.

. 2024. “Geochat: Grounded Large Vision-Language Model for Remote Sensing.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 27831–27840.

91.

LeCun

Bengio

Hinton

. 2015. “Deep Learning.” Nature 521 (7553): 436–44. DOI: https://doi.org/10.1038/nature14539.

92.

Yigitcanlar

Nepal

Nguyen

Dur

. 2023. “Machine Learning and Remote Sensing Integration for Leveraging Urban Sustainability: A Review and Framework.” Sustainable Cities and Society 96: 104653. DOI: https://doi.org/10.1016/j.scs.2023.104653.

93.

Deuser

Yin

Luo

Walther

Mai

Huang

Werner

. 2025. “Cross-View Geolocalization and Disaster Mapping With Street-View and Vhr Satellite Imagery: A Case Study of Hurricane Ian.” ISPRS Journal of Photogrammetry and Remote Sensing 220: 841–54. DOI: https://doi.org/10.1016/j.isprsjprs.2025.01.003.

94.

Koks

Taubenböck

van Vliet

. 2020a. “Continental-Scale Mapping and Analysis of 3D Building Structure.” Remote Sensing of Environment 245: 111859. DOI: https://doi.org/10.1016/j.rse.2020.111859.

95.

Mou

Sun

Hua

Shi

Zhu

X. X.

. 2024a. “A Review of Building Extraction From Remote Sensing Imagery: Geometrical Structures and Semantic Attributes.” IEEE Transactions on Geoscience and Remote Sensing. DOI: https://doi.org/10.1109/TGRS.2024.3369723.

96.

Hsu

C. Y.

Wang

Yang

Lee

Liljedahl

Witharana

Yang

Rogers

B. M.

Arundel

S. T.

, et al. 2024b. “Segment Anything Model Can Not Segment Anything: Assessing Ai Foundation Models Generalizability in Permafrost Mapping.” Remote Sensing 16 (5): 797. DOI: https://doi.org/10.3390/rs16050797.

97.

Zhou

Gong

Seto

K. C.

Clinton

. 2020b. “Developing a Method to Estimate Building Height From Sentinel-1 Data.” Remote Sensing of Environment 240: 111705. DOI: https://doi.org/10.1016/j.rse.2020.111705.

98.

Zhang

. 2021. “Image Retrieval From Remote Sensing Big Data: A Survey.” Information Fusion 67: 94–115. DOI: https://doi.org/10.1016/j.inffus.2020.10.008.

99.

2024. “Geoshapley: A Game Theory Approach to Measuring Spatial Effects in Machine Learning Models.” Annals of the American Association of Geographers 114 (7): 1365–85. DOI: https://doi.org/10.1080/24694452.2024.2350982.

100.

Liu

Bruch

. 2020. “Weed Detection for Selective Spraying: A Review.” Current Robotics Reports. DOI: https://doi.org/10.1007/s43154-020-00001-w.

101.

Liu

C. J.

Krylov

V. A.

Kane

Kavanagh

Dahyot

. 2020. “Im2elevation: Building Height Estimation From Single-View Aerial Imagery.” Remote Sensing 12 (17). DOI: https://doi.org/10.3390/rs12172719.

102.

Liu

Chen

Guan

Zhou

Zhu

Zhou

. 2024a. “Remoteclip: A Vision Language Foundation Model for Remote Sensing.” IEEE Transactions on Geoscience and Remote Sensing. DOI: https://doi.org/10.1109/TGRS.2024.3390838.

103.

Liu

Lee

Y. J.

. 2023a. “Visual Instruction Tuning.” arXiv preprint arXiv:2304.08485.

104.

Liu

Ding

Cai

Zhang

Satzoda

R. K.

Mahadevan

Manmatha

. 2023b. “Polyformer: Referring Image Segmentation as Sequential Polygon Generation.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18653–18663. DOI: 10.1109/CVPR52729.2023.01789.

105.

Liu

Tang

Guan

Griffis

T. J.

Erickson

M. D.

Frie

A. L.

Jia

Kim

Miller

L. T.

, et al. 2022. “Kgml-ag: A Modeling Framework of Knowledge-Guided Machine Learning to Simulate Agroecosystems: A Case Study of Estimating N 2 O Emission Using Data From Mesocosm Experiments.” Geoscientific Model Development 15 (7): 2839–58. DOI: https://doi.org/10.5194/gmd-15-2839-2022.

106.

Liu

Zhou

Guan

Peng

Tang

Zhu

Till

Jia

Jiang

, et al. 2024b. “Knowledge-Guided Machine Learning Can Improve Carbon Cycle Quantification in Agroecosystems.” Nature Communications 15 (1): 357. DOI: https://doi.org/10.1038/s41467-023-43860-5.

107.

Lorrain-Soligon

Robin

Barbraud

Brischoux

. 2023. “Some Like it Salty: Spatio-Temporal Dynamics of Salinity Differentially Affect Anurans and Caudates in Coastal Wetlands.” Freshwater Biology 68 (7): 1279–92. DOI: https://doi.org/10.1111/fwb.14103.

108.

Lovejoy

T. E.

Nobre

. 2018. “Amazon Tipping Point.” DOI: 10.1126/sciadv.aat2340.

109.

Guo

Zimmer-Dauphinee

J. R.

Nieusma

J. M.

Wang

VanValkenburgh

Wernke

S. A.

Huo

. 2024. “AI Foundation Models in Remote Sensing: A Survey.” arXiv preprint arXiv:2408.03464.

110.

Liu

Zhang

Yin

Johnson

B. A.

. 2019. “Deep Learning in Remote Sensing Applications: A Meta-Analysis and Review.” ISPRS Journal of Photogrammetry and Remote Sensing 152: 166–77. DOI: https://doi.org/10.1016/j.isprsjprs.2019.04.015.

111.

Madhavi

Kolikipogu

Prabakar

Banerjee

Maguluri

L. P.

Raj

G. B.

Balaram

. 2024. “Experimental Evaluation of Remote Sensing-Based Climate Change Prediction Using Enhanced Deep Learning Strategy.” Remote Sensing in Earth Systems Sciences 1–15. DOI: https://doi.org/10.1007/s41976-024-00152-w.

112.

Mahendra

Seto

K. C.

. 2019. “Upward and Outward Growth: Managing Urban Expansion for More Equitable Cities in the Global South.” World Resources Institute.

113.

Mai

2024. “Geo-Foundation Model.” International Encyclopedia of Geography: People, the Earth, Environment and Technology. DOI: https://doi.org/10.1002/9781118786352.wbieg2206.

114.

Mai

Huang

Sun

Song

Mishra

Liu

Gao

Liu

Cong

, et al. 2024a. “On the Opportunities and Challenges of Foundation Models for Geoai (vision Paper).” ACM Transactions on Spatial Algorithms and Systems. DOI: https://doi.org/10.1145/3653070.

115.

Mai

Jiang

Sun

Zhu

Xuan

Cai

Janowicz

Ermon

Lao

. 2023a. “Towards General-Purpose Representation Learning of Polygonal Geometries.” GeoInformatica 27 (2): 289–340. DOI: https://doi.org/10.1007/s10707-022-00481-2.

116.

Mai

Lao

Song

Ermon

. 2023b. “CSP: Self-Supervised Contrastive Spatial Pre-Training for Geospatial-Visual Representations.” In International Conference on Machine Learning. PMLR, 23498–23515.

117.

Mai

Xie

Jia

Lao

Rao

Zhu

Liu

Chiang

Y. Y.

Jiao

. 2025. “Towards the Next Generation of Geospatial Artificial Intelligence.” International Journal of Applied Earth Observation and Geoinformation 136: 104368. DOI: https://doi.org/10.1016/j.jag.2025.104368.

118.

Mai

Yao

Xie

Rao

Zhu

Lao

. 2024b. “SRL: Towards a General-Purpose Framework for Spatial Representation Learning.” In Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems, 465–468. DOI: 10.1145/3678717.3691246.

119.

Mall

Phoo

Liu

Vondrick

Hariharan

Bala

. 2023. “Remote Sensing Vision-Language Foundation Models Without Annotations Via Ground Remote Alignment.” arxiv 2023. arXiv preprint arXiv:2312.06960.

120.

Manfreda

McCabe

M. F.

Miller

P. E.

Lucas

Pajuelo Madrigal

Mallinis

Ben Dor

Helman

Estes

Ciraolo

, et al. 2018. “On the Use of Unmanned Aerial Systems for Environmental Monitoring.” Remote Sensing 10 (4): 641. DOI: https://doi.org/10.3390/rs10040641.

121.

Maniyar

C. B.

Kumar

Mishra

D. R.

. 2022. “Continuous and Synoptic Assessment of Indian Inland Waters for Harmful Algae Blooms.” Harmful Algae 111: 102160. DOI: https://doi.org/10.1016/j.hal.2021.102160.

122.

Maniyar

C. B.

Kumar

. 2021. “Deep Learning-Based Improved Automatic Building Extraction from Open-Source High Resolution Unmanned Aerial Vehicle (UAV) imagery.” In International Conference on Unmanned Aerial System in Geomatics. Springer, 51–66. DOI: 10.1007/978-3-031-19309-5_5.

123.

Maniyar

C. B.

Kumar

Mai

. 2025. ““Feature-Augmented Deep Networks for Multiscale Building Segmentation in High-Resolution UAV and Satellite Imagery.” arXiv preprint arXiv:2505.05321. DOI: 10.48550/arXiv.2505.05321.

124.

Manvi

Khanna

Mai

Burke

Lobell

Ermon

. 2024. “GeoLLM: Extracting Geospatial Knowledge from Large Language Models.” In The Twelfth International Conference on Learning Representations (ICLR 2024), 1–17.

125.

Marolla

2025. California Wildfires: A Holistic Systemic Management Approach to Urban Resilience. Cambridge University Press.

126.

Marrone

B. L.

Banerjee

Talapatra

Gonzalez-Esquer

C. R.

Pilania

. 2023. “Toward a Predictive Understanding of Cyanobacterial Harmful Algal Blooms Through Ai Integration of Physical, Chemical, and Biological Data.” ACS ES&T Water 4 (3): 844–58. DOI: https://doi.org/10.1021/acsestwater.3c00369.

127.

Marsetič

Kanjir

. 2024. “Forecasting Vegetation Behavior Based on Planetscope Time Series Data Using RNN-Based Models.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 17: 5015–25.

128.

Masson-Delmotte

Zhai

Pirani

Connors

Péan

Berger

Caud

Chen

Goldfarb

Scheel Monteiro

P. M.

. 2021. “IPCC, 2021: Summary for Policymakers.” In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.

129.

Mazzei

Sullivan

K. L.

Loftin

. 2025. “Phytoplankton Assemblage Structure, Drivers, and Thresholds with a Focus on Harmful Algal Bloom Ecology in the Lake Okeechobee System, Florida, Usa.” Harmful Algae 142: 102744. DOI: https://doi.org/10.1016/j.hal.2024.102744.

130.

McCarty

L. B.

Murphy

T. R.

. 1994. “Control of Turf Grass Weeds.” Turf Weeds and Their Control 209–48. DOI: https://doi.org/10.2134/turfweedsandtheircontrol1994.acsesspublicatiturfweedsandthe209x.

131.

McKenzie

Zhang

Gambs

. 2023. “Privacy and Ethics in GeoAI.” In Handbook of Geospatial Artificial Intelligence. CRC Press, 388–405. DOI: 10.1201/9781003308423-19.

132.

Miao

Niu

. 2016. “A Survey on Feature Selection.” Procedia Computer Science 91: 919–26. DOI: https://doi.org/10.1016/j.procs.2016.07.111.

133.

Mishra

D. R.

Kumar

Ramaswamy

Boddula

V. K.

Das

M. C.

Page

B. P.

Weber

S. J.

. 2020. “Cyanotracker: A Cloud-Based Integrated Multi-Platform Architecture for Global Observation of Cyanobacterial Harmful Algal Blooms.” Harmful Algae 96: 101828. DOI: https://doi.org/10.1016/j.hal.2020.101828.

134.

Mohanty

S. P.

Hughes

D. P.

Salathé

. 2016. “Using Deep Learning for Image-Based Plant Disease Detection.” Frontiers in Plant Science 7: 1419. DOI: https://doi.org/10.3389/fpls.2016.01419.

135.

Molnar

. 2020. “Interpretable Machine Learning.” Lulu.com.

136.

Monteiro

Santos

. 2022. “Sustainable Approach to Weed Management: The Role of Precision Weed Management.” Agronomy 12 (1): 118. DOI: https://doi.org/10.3390/agronomy12010118.

137.

Monteith

J. L.

1972. “Solar Radiation and Productivity in Tropical Ecosystems.” Journal of Applied Ecology 9 (3): 747–66. DOI: https://doi.org/10.2307/2401901.

138.

Moomaw

W. R.

Chmura

Davies

G. T.

Finlayson

Middleton

B. A.

Natali

S. M.

Perry

Roulet

Sutton-Grier

A. E.

. 2018. “Wetlands in a Changing Climate: Science, Policy and Management.” Wetlands 38 (2): 183–205. DOI: https://doi.org/10.1007/s13157-018-1023-8.

139.

Moradi

Uta

. 2021. “Reproducible Model Sharing for AI Practitioners.” In Proceedings of the Fifth Workshop on Distributed Infrastructures for Deep Learning (DIDL) 2021, 1–6. DOI: 10.1145/3493652.3505630.

140.

Nair

A. S.

Mangla R

P. T.

Indu

. 2022. “Remote Sensing Data Assimilation.” Hydrological Sciences Journal 67 (16): 2457–89. DOI: https://doi.org/10.1080/02626667.2020.1761021.

141.

Nations

2021. “Climate Crises—A Race We Can Win.” https://www.un.org/sites/un2.un.org/files/2020/01/un75_climate_crisis.pdf. Accessed 07 February 2025.

142.

Neupane

Horanont

Aryal

. 2021. “Deep Learning-Based Semantic Segmentation of Urban Features in Satellite Images: A Review and Meta-Analysis.” Remote Sensing 13 (4): 808. DOI: https://doi.org/10.3390/rs13040808.

143.

Nguyen

Brandstetter

Kapoor

Gupta

J. K.

Grover

. 2023. “Climax: A Foundation Model for Weather and Climate.” In International Conference on Machine Learning. PMLR, 25904–25938.

144.

Nguyen

T. G.

Tran

N. A.

P. L.

Nguyen

Q. H.

Nguyen

H. D.

Bui

Q. T.

. 2021. “Salinity Intrusion Prediction Using Remote Sensing and Machine Learning in Data-Limited Regions: A Case Study in Vietnam’s Mekong Delta.” Geoderma Regional 27: e00424. DOI: https://doi.org/10.1016/j.geodrs.2021.e00424.

145.

Nieves

Radin

Camps-Valls

. 2021. “Predicting Regional Coastal Sea Level Changes with Machine Learning.” Scientific Reports 11 (1): 7650. DOI: https://doi.org/10.1038/s41598-021-87460-z.

146.

Noman

Naseer

Cholakkal

Anwer

R. M.

Khan

F. S.

. 2024. “Rethinking Transformers Pre-Training for Multi-Spectral Satellite Imagery.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 27811–27819. DOI: 10.1109/CVPR52733.2024.02627.

147.

O’Connell

J. L.

Mishra

D. R.

Alber

Byrd

K. B.

. 2021. “Berm: A Belowground Ecosystem Resiliency Model for Estimating Spartina Alterniflora Belowground Biomass.” New Phytologist 232 (1): 425–39.

148.

Osco

L. P.

de Lemos

E. L.

Gonçalves

W. N.

Ramos

A. P. M.

Junior

J. M.

. 2023. “The Segment Anything Model (sam) for Remote Sensing Applications: From Zero to One Shot.” International Journal of Applied Earth Observation and Geoinformation 124: 103540. DOI: https://doi.org/10.1016/j.jag.2023.103540.

149.

O’Shea

R. E.

Pahlevan

Smith

Bresciani

Egerton

Giardino

Moore

Ruiz-Verdu

Ruberg

, et al. 2021. “Advancing Cyanobacteria Biomass Estimation From Hyperspectral Observations: Demonstrations with Hico and Prisma Imagery.” Remote Sensing of Environment 266: 112693. DOI: https://doi.org/10.1016/j.rse.2021.112693.

150.

Paerl

H. W.

Otten

T. G.

. 2013. “Harmful Cyanobacterial Blooms: Causes, Consequences, and Controls.” Microbial Ecology 65: 995–1010. DOI: https://doi.org/10.1007/s00248-012-0159-y.

151.

Pahlevan

Smith

Alikas

Anstee

Barbosa

Binding

Bresciani

Cremella

Giardino

Gurlin

, et al. 2022. “Simultaneous Retrieval of Selected Optical Water Quality Indicators From Landsat-8, Sentinel-2, and Sentinel-3.” Remote Sensing of Environment 270: 112860. DOI: https://doi.org/10.1016/j.rse.2021.112860.

152.

Paolanti

Tiribelli

Giovanola

Mancini

Frontoni

Pierdicca

. 2024. “Ethical Framework to Assess and Quantify the Trustworthiness of Artificial Intelligence Techniques: Application Case in Remote Sensing.” Remote Sensing 16 (23): 4529. DOI: https://doi.org/10.3390/rs16234529.

153.

Partel

Kakarla

S. C.

Ampatzidis

. 2019. “Development and Evaluation of a Low-Cost and Smart Technology for Precision Weed Management Utilizing Artificial Intelligence.” Computers and Electronics in Agriculture 157: 339–50. DOI: https://doi.org/10.1016/j.compag.2018.12.048.

154.

Pei

Dong

Zhang

Yuan

Doughty

Yang

Zhou

Zhang

Xiao

. 2022. “Evolution of Light Use Efficiency Models: Improvement, Uncertainties, and Implications.” Agricultural and Forest Meteorology 317: 108905. DOI: https://doi.org/10.1016/j.agrformet.2022.108905.

155.

Pettorelli

Laurance

W. F.

O’Brien

T. G.

Wegmann

Nagendra

Turner

. 2014. “Satellite Remote Sensing for Applied Ecologists: Opportunities and Challenges.” Journal of Applied Ecology 51 (4): 839–48. DOI: https://doi.org/10.1111/1365-2664.12261.

156.

Prabhu

Niranjan

Rai

S. S.

. 2024. “Deepweed: A Dynamically Adaptive Lightweight Deep Learning Model for Efficient Weed Classification.” In 2024 4th Asian Conference on Innovation in Technology (ASIANCON). IEEE, 1–6. DOI: 10.1109/ASIANCON62057.2024.10837861.

157.

Rahmstorf

2006. “Thermohaline Ocean Circulation.” Encyclopedia of Quaternary Sciences 5 (18): 191.

158.

Raju

Chaudhary

Jha

. 2014. “Shadow Analysis Technique for Extraction of Building Height Using High Resolution Satellite Single Image and Accuracy Assessment.” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 40: 1185–92. DOI: https://doi.org/10.5194/isprsarchives-XL-8-1185-2014.

159.

Rakhmatulin

Kamilaris

Andreasen

. 2021. “Deep Neural Networks to Detect Weeds From Crops in Agricultural Environments in Real-Time: A Review.” Remote Sensing 13 (21): 4486. DOI: https://doi.org/10.3390/rs13214486.

160.

Ramesh

Matloob

Aslam

Florentine

S. K.

Chauhan

B. S.

. 2017. “Weeds in a Changing Climate: Vulnerabilities, Consequences, and Implications for Future Weed Management.” Frontiers in Plant Science 8: 95. DOI: https://doi.org/10.3389/fpls.2017.00095.

161.

Robinson

J. D.

Diaz-Ferguson

Poelchau

M. F.

Pennings

Bishop

T. D.

Wares

. 2010. “Multiscale Diversity in the Marshes of the Georgia Coastal Ecosystems Lter.” Estuaries and Coasts 33: 865–77. DOI: https://doi.org/10.1007/s12237-009-9188-2.

162.

Romano

. 2025. “Synthetic Geospatial Data and Fake Geography: A Case Study on the Implications of AI-Derived Data in a Data-Intensive Society.” Digital Geography and Society 8: 100108. DOI: https://doi.org/10.1016/j.diggeo.2024.100108.

163.

Ronneberger

Fischer

Brox

. 2015. “U-Net: Convolutional Networks for Biomedical Image Segmentation.” In Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015: 18th International Conference, Munich, Germany, October 5–9, 2015, Proceedings, Part III 18. Springer, 234–241. DOI: 10.1007/978-3-319-24574-4_28.

164.

Rosen

B. H.

Davis

T. W.

Gobler

C. J.

Kramer

B. J.

Loftin

K. A.

. 2017. “Cyanobacteria of the 2016 Lake Okeechobee and Okeechobee Waterway Harmful Algal Bloom.” Technical Report, US Geological Survey.

165.

Runion

K. D.

Mishra

D. R.

Alber

Lever

M. A.

O’Connell

J. L.

. 2024. “Capturing Spatiotemporal Variation in Salt Marsh Belowground Biomass, a Key Resilience Metric, Through Geoinformatics.” Ecosphere (Washington, DC) 15 (12): e70110. DOI: https://doi.org/10.1002/ecs2.70110.

166.

Schaeffer

B. A.

Reynolds

Ferriby

Salls

Smith

Johnston

J. M.

Myer

. 2024. “Forecasting Freshwater Cyanobacterial Harmful Algal Blooms for Sentinel-3 Satellite Resolved Us Lakes and Reservoirs.” Journal of Environmental Management 349: 119518.

167.

Shao

Taff

G. N.

Walsh

S. J.

. 2011. “Shadow Detection and Building-Height Estimation Using Ikonos Data.” International Journal of Remote Sensing 32 (22): 6929–44. DOI: https://doi.org/10.1080/01431161.2010.517226.

168.

Sharma

L. K.

Naik

. 2024. “Future Technologies for Wetland Studies.” In Conservation of Saline Wetland Ecosystems: An Initiative Towards UN Decade of Ecological Restoration. Springer, 175–215. DOI: 10.1007/978-981-97-5069-6_6.

169.

Sharma

Levi

M. R.

Ricker

M. C.

Thompson

King

E. G.

Robertson

. 2024. “Scaling of Soil Organic Carbon in Space and Time in the Southern Coastal Plain, Usa.” Science of the Total Environment 933: 173060. DOI: https://doi.org/10.1016/j.scitotenv.2024.173060.

170.

Sharma

Mishra

D. R.

Levi

M. R.

Sutter

L. A.

. 2022. “Remote Sensing of Surface and Subsurface Soil Organic Carbon in Tidal Wetlands: A Review and Ideas for Future Research.” Remote Sensing 14 (12): 2940. DOI: https://doi.org/10.3390/rs14122940.

171.

Shi

Currier

Liu

Janowicz

Wiedemann

Verstegen

McKenzie

Graser

Zhu

Mai

. 2023. “Thinking Geographically About Ai Sustainability.” AGILE: GIScience Series 4: 42. DOI: https://doi.org/10.5194/agile-giss-4-42-2023.

172.

Shoaib

M. R.

Emara

H. M.

Zhao

. 2023. “Revolutionizing Global Food Security: Empowering Resilience Through Integrated AI Foundation Models and Data-Driven Solutions.” arXiv preprint arXiv:2310.20301.

173.

Simonyan

Zisserman

. 2014. “Very Deep Convolutional Networks for Large-Scale Image Recognition.” arXiv preprint arXiv:1409.1556.

174.

Slater

L. J.

Arnal

Boucher

M. A.

Chang

A. Y. Y.

Moulds

Murphy

Nearing

Shalev

Shen

Speight

, et al. 2023. “Hybrid Forecasting: Blending Climate Predictions with Ai Models.” Hydrology and Earth System Sciences 27 (9): 1865–89. DOI: https://doi.org/10.5194/hess-27-1865-2023.

175.

Smyth

Elliott

. 2016. “Effects of Changing Salinity on the Ecology of the Marine Environment.” Stressors in the Marine Environment: Physiological and Ecological Responses; Societal Implications 161–74. DOI: https://doi.org/10.1093/acprof:oso/9780198718826.003.0009.

176.

Steffen

M. M.

Belisle

B. S.

Watson

S. B.

Boyer

G. L.

Wilhelm

S. W.

. 2014. “Status, Causes and Controls of Cyanobacterial Blooms in Lake Erie.” Journal of Great Lakes Research 40 (2): 215–25. DOI: https://doi.org/10.1016/j.jglr.2013.12.012.

177.

Strubell

Ganesh

McCallum

. 2020. “Energy and Policy Considerations for Modern Deep Learning Research.” In Proceedings of the AAAI Conference on Artificial Intelligence, Volume 34, 13693–13696. DOI: 10.1609/aaai.v34i09.7123.

178.

Szegedy

Liu

Jia

Sermanet

Reed

Anguelov

Erhan

Vanhoucke

Rabinovich

. 2015. “Going Deeper with Convolutions.” In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1–9. DOI: 10.1109/CVPR.2015.7298594.

179.

Touvron

Lavril

Izacard

Martinet

Lachaux

M. A.

Lacroix

Rozière

Goyal

Hambro

Azhar

, et al. 2023. “LLaMA: Open and Efficient Foundation Language Models.” arXiv preprint arXiv:2302.13971.

180.

Trainer

V. L.

Moore

S. K.

Hallegraeff

Kudela

R. M.

Clement

Mardones

J. I.

Cochlan

W. P.

. 2020. “Pelagic Harmful Algal Blooms and Climate Change: Lessons From Nature’s Experiments with Extremes.” Harmful Algae 91: 101591. DOI: https://doi.org/10.1016/j.hal.2019.03.009.

181.

Tuia

Kellenberger

Beery

Costelloe

B. R.

Zuffi

Risse

Mathis

M. W.

Van Langevelde

Burghardt

, et al. 2022. “Perspectives in Machine Learning for Wildlife Conservation.” Nature Communications 13 (1): 792. DOI: https://doi.org/10.1038/s41467-022-27980-y.

182.

Vasileiou

Kyrgiakos

L. S.

Kleisiari

Kleftodimos

Vlontzos

Belhouchette

Pardalos

P. M.

. 2024. “Transforming Weed Management in Sustainable Agriculture with Artificial Intelligence: A Systematic Literature Review Towards Weed Identification and Deep Learning.” Crop Protection 176: 106522. DOI: https://doi.org/10.1016/j.cropro.2023.106522.

183.

Veerman

Mishra

D. R.

Kumar

Karidozo

. 2024. “Environmental Drivers Behind the Exceptional Increase in Cyanobacterial Blooms in Okavango Delta, Botswana.” Harmful Algae 102677. DOI: https://doi.org/10.1016/j.hal.2024.102677.

184.

Wang

Gao

R. X.

Zhang

. 2022. “Hybrid Physics-Based and Data-Driven Models for Smart Manufacturing: Modelling, Simulation, and Explainability.” Journal of Manufacturing Systems 63: 381–91. DOI: https://doi.org/10.1016/j.jmsy.2022.04.004.

185.

Wang

Liu

Chao

Zhang

Zheng

Zhang

Xue

Wang

Lei

. 2024. “Assessing the Effectiveness of Pm2. 5 Pollution Control From the Perspective of Interprovincial Transport and Pm2. 5 Mitigation Costs Across China.” Environmental Science and Ecotechnology 22: 100448.

186.

Wang

Yao

Kwok

J. T.

L. M.

. 2020. “Generalizing From a Few Examples: A Survey on Few-Shot Learning.” ACM Computing Surveys (CSUR) 53 (3): 1–34. DOI: https://doi.org/10.1145/3386252.

187.

Warmink

J. J.

Janssen

Booij

M. J.

Krol

M. S.

. 2010. “Identification and Classification of Uncertainties in the Application of Environmental Models.” Environmental Modelling & Software 25 (12): 1518–27. DOI: https://doi.org/10.1016/j.envsoft.2010.04.011.

188.

Weber

S. J.

Mishra

D. R.

Wilde

S. B.

Kramer

. 2020. “Risks for Cyanobacterial Harmful Algal Blooms Due to Land Management and Climate Interactions.” Science of the Total Environment 703: 134608. DOI: https://doi.org/10.1016/j.scitotenv.2019.134608.

189.

Wei

Fang

Hendrey

. 2017. “A Global Study of Gpp Focusing on Light-Use Efficiency in a Random Forest Regression Model.” Ecosphere (Washington, DC) 8 (5): e01724. DOI: https://doi.org/10.1002/ecs2.1724.

190.

Westwood

J. H.

Charudattan

Duke

S. O.

Fennimore

S. A.

Marrone

Slaughter

D. C.

Swanton

Zollinger

. 2018. “Weed Management in 2050: Perspectives on the Future of Weed Science.” Weed Science 66 (3): 275–85. DOI: https://doi.org/10.1017/wsc.2017.78.

191.

Zhang

Liu

. 2019. “A Review on Drone-Based Harmful Algae Blooms Monitoring.” Environmental Monitoring and Assessment 191: 1–11. DOI: https://doi.org/10.1007/s10661-019-7365-8.

192.

Cao

Wang

Liu

Zhang

Yao

X. A.

, et al. 2024. “Torchspatial: A Location Encoding Framework and Benchmark for Spatial Representation Learning.” In The 38th Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 1–24.

193.

Osco

L. P.

. 2023. “samgeo: A Python Package for Segmenting Geospatial Data With the Segment Anything Model (SAM).” Journal of Open Source Software 8 (89): 5663. DOI: https://doi.org/10.21105/joss.05663.

194.

W. B.

Banzhaf

Meadows

M. E.

Z. W.

Guo

F. X.

Sengupta

Cai

X. X.

Zhao

. 2023. “A First Chinese Building Height Estimate at 10 m Resolution (cnbh-10 M) Using Multi-Source Earth Observations and Machine Learning.” Remote Sensing of Environment 291: 113578. DOI: https://doi.org/10.1016/j.rse.2023.113578.

195.

Wulder

M. A.

Roy

D. P.

Radeloff

V. C.

Loveland

T. R.

Anderson

M. C.

Johnson

D. M.

Healey

Zhu

Scambos

T. A.

Pahlevan

, et al. 2022. “Fifty Years of Landsat Science and Impacts.” Remote Sensing of Environment 280: 113195. DOI: https://doi.org/10.1016/j.rse.2022.113195.

196.

Wynne

T. T.

Mishra

Meredith

Litaker

R. W.

Stumpf

R. P.

. 2021. “Intercalibration of Meris, Modis, and Olci Satellite Imagers for Construction of Past, Present, and Future Cyanobacterial Biomass Time Series.” Remote Sensing 13 (12): 2305. DOI: https://doi.org/10.3390/rs13122305.

197.

Xie

Wang

Mai

Jia

Gao

Wang

. 2023. “Geo-Foundation Models: Reality, Gaps and Opportunities.” In Proceedings of the 31st ACM International Conference on Advances in Geographic Information Systems, 1–4. DOI: 10.1145/3589132.3625616.

198.

Lin

. 2015. “Fusion of Worldview-2 Stereo and Multitemporal Terrasar-x Images for Building Height Extraction in Urban Areas.” IEEE Geoscience and Remote Sensing Letters 12 (8): 1795–9.

199.

Yan

2024. “Estimate the Building Height at a 10-Meter Resolution Based on Sentinel Data.” arXiv preprint arXiv:2405.00989.

200.

Yin

Dong

Hamm

N. A.

Wang

Xing

. 2021. “Integrating Remote Sensing and Geospatial Big Data for Urban Land Use Mapping: A Review.” International Journal of Applied Earth Observation and Geoinformation 103: 102514. DOI: https://doi.org/10.1016/j.jag.2021.102514.

201.

Schumann

A. W.

Sharpe

S. M.

Boyd

N. S.

. 2020. “Detection of Grassy Weeds in Bermudagrass With Deep Convolutional Neural Networks.” Weed Science 68 (5): 545–52. DOI: https://doi.org/10.1017/wsc.2020.46.

202.

Sharpe

S. M.

Schumann

A. W.

Boyd

N. S.

. 2019. “Detection of Broadleaf Weeds Growing in Turfgrass With Convolutional Neural Networks.” Pest Management Science 75 (8): 2211–8. DOI: https://doi.org/10.1002/ps.5349.

203.

Yuan

Shen

Jiang

Tan

Yang

Wang

, et al. 2020. “Deep Learning in Environmental Remote Sensing: Achievements and Challenges.” Remote Sensing of Environment 241: 111716. DOI: https://doi.org/10.1016/j.rse.2020.111716.

204.

Yuan

Shi

. 2021. “A Review of Deep Learning Methods for Semantic Segmentation of Remote Sensing Imagery.” Expert Systems With Applications 169: 114417. DOI: https://doi.org/10.1016/j.eswa.2020.114417.

205.

Zahir

Shahzad

M. I.

Ayub

Rehman

S. U.

Ijaz

. 2024. “A Review on Monitoring, Forecasting, and Early Warning of Harmful Algal Bloom.” Aquaculture (Amsterdam, Netherlands) 741351. DOI: https://doi.org/10.1016/j.aquaculture.2024.741351.

206.

Zakeri

Mariethoz

. 2021. “A Review of Geostatistical Simulation Models Applied to Satellite Remote Sensing: Methods and Applications.” Remote Sensing of Environment 259: 112381. DOI: https://doi.org/10.1016/j.rse.2021.112381.

207.

Zhang

Xie

Bai

Gao

. 2021. “A Survey on Federated Learning.” Knowledge-Based Systems 216: 106775. DOI: https://doi.org/10.1109/ICCA51439.2020.9264412.

208.

Zhang

Chen

Rajbhandari-Thapa

Pagán

J. A.

Mai

Zhou

. 2025. “Space: A Spatial Counterfactual Explainable Deep Learning Model for Predicting Out-of-Hospital Cardiac Arrest Survival Outcome.” International Journal of Geographical Information Science 1–32. DOI: https://doi.org/10.1080/13658816.2024.2443757.

209.

Zhang

Zhou

Mai

Guan

. 2024. “Text2Seg: Zero-Shot Remote Sensing Image Semantic Segmentation Via Text-Guided Visual Foundation Models.” In Proceedings of the 7th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery, 63–66. DOI: 10.1145/3687123.3698287.

210.

Zhang

. 2022. “Artificial Intelligence for Remote Sensing Data Analysis: A Review of Challenges and Opportunities.” IEEE Geoscience and Remote Sensing Magazine 10 (2): 270–94. DOI: https://doi.org/10.1109/MGRS.2022.3145854.

211.

Zhang

Roller

Goyal

Artetxe

Chen

Dewan

Diab

Lin

X. V.

, et al. 2022a. “OPT: Open Pre-Trained Transformer Language Models.” arXiv preprint arXiv:2205.01068.

212.

Zhang

Duy

S. V.

Munoz

Sauvé

. 2022b. “Phytotoxic Effects of Microcystins, Anatoxin-a and Cylindrospermopsin to Aquatic Plants: A Meta-Analysis.” Science of the Total Environment 810: 152104. DOI: https://doi.org/10.1016/j.scitotenv.2021.152104.

213.

Zhao

Zhang

Sun

Deng

. 2021. “Deep Fake Geography? when Geospatial Data Encounter Artificial Intelligence.” Cartography and Geographic Information Science 48 (4): 338–52. DOI: https://doi.org/10.1080/15230406.2021.1910075.

214.

Zheng

Zhong

Zhang

Ermon

. 2024. “Segment Any Change.” In NeurIPS 2024, 1–15.

215.

Zhu

Liu

Zhang

. 2023. “Likely Accelerated Weakening of Atlantic Overturning Circulation Emerges in Optimal Salinity Fingerprint.” Nature Communications. DOI: https://doi.org/10.1038/s41467-023-36288-4.