pith. sign in

arxiv: 2606.07780 · v1 · pith:M2U7PMYUnew · submitted 2026-06-05 · 💻 cs.AI · cs.CV· cs.LG

Land cover and flood type govern the detection limits of satellite-based flood mapping across diverse global flood events

Venkatesh Kolluru , Rajat Shinde , Abdelhak Marouane , Caden Helbling , Deepak Shah , Othneil Drew , Iksha Gurung , Manil Maskey
show 1 more author
Rahul Ramachandran
This is my paper

Pith reviewed 2026-06-27 21:38 UTC · model grok-4.3

classification 💻 cs.AI cs.CVcs.LG
keywords flood mappingsatellite imageryland coverflood typedetection limitsgeospatial modelsvalidationdisaster response
0
0 comments X

The pith

Land cover and flood type jointly determine satellite flood mapping accuracy across global events.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests a geospatial foundation model on 19 unseen flood events spanning six continents and multiple mechanisms to measure real-world reliability. It establishes that detection success is controlled by the specific combination of land cover and flood type rather than model design in isolation. Cropland produces the strongest agreement while riverine floods are detected most reliably, but tree cover and built-up zones yield near-zero performance. Some measured errors arise from inconsistencies between the two reference datasets instead of outright model failure. Pipeline engineering choices affect outcomes more than the base model capacity.

Core claim

Across 19 out-of-distribution flood events, the model reaches highest agreement in cropland (IoU 52 percent) and riverine floods (F1 0.69), yet shows near-zero detection in tree cover and built-up areas (IoU 4 percent) regardless of flood mechanism. Dual-reference validation shows that part of the apparent error reflects definitional inconsistency between the reference products. Iterative testing isolates 23 failure modes where pipeline engineering dominates initial error over model capacity, establishing environment-dependent detection boundaries for operational satellite flood mapping.

What carries the argument

Joint dependence of detection metrics (IoU and F1) on land cover class and flood mechanism

If this is right

  • Detection performance reaches its highest levels in cropland and for riverine flood events.
  • Tree cover and built-up areas produce near-zero detection irrespective of the flood mechanism.
  • Inconsistencies between reference products account for a portion of measured model error.
  • Pipeline engineering choices exert greater influence on results than model capacity alone.
  • Operational satellite flood mapping is bounded by these environment-specific limits.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Monitoring systems could prioritize cropland and riverine zones where satellite detection is most reliable.
  • Urban and forested flood response may need supplementary ground or radar data sources.
  • Model training could incorporate land-cover stratification to reduce systematic blind spots.

Load-bearing premise

The two independent reference products supply sufficiently consistent and accurate benchmarks for true flood extent despite definitional differences between them.

What would settle it

A documented flood event in tree-covered or built-up terrain where the model reaches IoU above 30 percent against both independent references would undermine the near-zero detection claim.

read the original abstract

Floods are among the most destructive natural hazards, and their increasing frequency under climate change makes satellite-based inundation mapping essential for disaster response. Geospatial foundation models pretrained on satellite archives offer geographic transferability, but their operational reliability across diverse, unseen events remains uncharacterized. Here we deploy Prithvi-EO-2.0 across 19 out-of-distribution flood events (2017-2025) spanning six continents, eight climate zones, and six flood mechanisms, validating against two independent reference products. Detection accuracy depended jointly on land cover and flood type, with cropland yielding the highest agreement (IoU=52%) and riverine events the strongest detection (F1=0.69), while tree cover and built-up areas showed near-zero detection (IoU=4%) regardless of flood mechanism. Dual-reference validation revealed that apparent model error partly reflects definitional inconsistency between reference products rather than detection failure. Iterative pipeline testing identified 23 failure modes, with pipeline engineering dominating initial error over model capacity. These findings establish environment-dependent detection boundaries for operational satellite flood mapping.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript evaluates the Prithvi-EO-2.0 geospatial foundation model on 19 out-of-distribution flood events (2017-2025) spanning six continents, eight climate zones, and six flood mechanisms. It validates model outputs against two independent reference products and reports that detection accuracy depends jointly on land cover and flood type, with cropland yielding the highest agreement (IoU=52%) and riverine events the strongest detection (F1=0.69), while tree cover and built-up areas show near-zero detection (IoU=4%) regardless of flood mechanism. Dual-reference validation indicates that some apparent model errors reflect definitional inconsistencies between references rather than detection failure, and iterative testing identifies 23 failure modes where pipeline engineering dominates over model capacity.

Significance. If the central empirical findings hold after addressing reference-consistency quantification, the work provides valuable environment-dependent detection boundaries for operational satellite flood mapping. The multi-continent, multi-mechanism out-of-distribution evaluation and explicit acknowledgment of reference inconsistencies are strengths that move beyond single-benchmark assessments common in the field.

major comments (2)
  1. [Abstract] Abstract and validation results: The claim that dual-reference validation shows 'apparent model error partly reflects definitional inconsistency' lacks a quantitative breakdown (e.g., per-land-cover agreement rates between the two references or the fraction of reported IoU/F1 'error' attributable to reference mismatch). This is load-bearing for the central claim because the near-zero IoU=4% in tree cover and built-up areas could be an artifact of higher benchmark disagreement in those classes rather than a true model detection limit.
  2. [Methods] Methods (event selection and validation protocol): No details are provided on data exclusion criteria, statistical tests for the reported IoU/F1 values, or how the 19 events were chosen from the larger set of possible floods. This affects interpretability of the joint land-cover/flood-type dependence, as post-hoc selection or inconsistent reference handling could inflate the reported environment-specific patterns.
minor comments (1)
  1. [Abstract] The abstract reports specific metrics (IoU=52%, F1=0.69, IoU=4%) without accompanying confidence intervals or sample sizes per category; adding these would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which help clarify the presentation of our results on environment-dependent flood detection limits with Prithvi-EO-2.0. We address each major comment below and will revise the manuscript accordingly to improve transparency and quantification.

read point-by-point responses

  1. Referee: [Abstract] Abstract and validation results: The claim that dual-reference validation shows 'apparent model error partly reflects definitional inconsistency' lacks a quantitative breakdown (e.g., per-land-cover agreement rates between the two references or the fraction of reported IoU/F1 'error' attributable to reference mismatch). This is load-bearing for the central claim because the near-zero IoU=4% in tree cover and built-up areas could be an artifact of higher benchmark disagreement in those classes rather than a true model detection limit.

    Authors: We agree that the current presentation would benefit from explicit quantification to support the claim. In the revised manuscript, we will add a new table or section reporting per-land-cover agreement rates (e.g., IoU and F1) between the two reference products, along with the estimated fraction of model-reference discrepancies attributable to definitional mismatch versus actual detection failure. This will directly address whether the low performance in tree cover and built-up areas is inflated by reference inconsistency. revision: yes

  2. Referee: [Methods] Methods (event selection and validation protocol): No details are provided on data exclusion criteria, statistical tests for the reported IoU/F1 values, or how the 19 events were chosen from the larger set of possible floods. This affects interpretability of the joint land-cover/flood-type dependence, as post-hoc selection or inconsistent reference handling could inflate the reported environment-specific patterns.

    Authors: We acknowledge that additional methodological transparency is warranted. We will expand the Methods section to include: (1) explicit data exclusion criteria applied to the initial pool of flood events, (2) the rationale and process for selecting the final 19 events (including any stratification by continent, climate zone, or mechanism to avoid post-hoc bias), and (3) statistical tests or uncertainty measures (e.g., bootstrap confidence intervals) for the reported IoU and F1 scores. These additions will strengthen the interpretability of the land-cover and flood-type dependence findings. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical evaluation against external references

full rationale

The paper conducts an empirical deployment and validation of a pretrained geospatial model on 19 out-of-distribution flood events, computing standard agreement metrics (IoU, F1) stratified by land cover and flood type against two independent external reference products. No mathematical derivations, parameter fitting presented as prediction, self-citation load-bearing premises, or ansatz smuggling appear in the described chain. All reported performance numbers are direct comparisons to external benchmarks, with the dual-reference analysis serving as an explicit check on reference inconsistency rather than a reduction to the paper's own inputs. The central claims therefore remain self-contained against outside data.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an empirical validation study applying an existing pretrained model to new satellite data; no new free parameters, axioms, or invented entities are introduced in the abstract.

pith-pipeline@v0.9.1-grok · 5764 in / 1115 out tokens · 28058 ms · 2026-06-27T21:38:42.720292+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 42 canonical work pages

  1. [1]

    and Liu, J

    Liu, Q., Du, M., Wang, Y., Deng, J., Yan, W., Qin, C., Liu, M. and Liu, J. Global, regional and national trends and impacts of natural floods, 1990–2022. Bulletin of the World Health Organization 102 , 410-420 (2024). https://doi.org/10.2471/BLT.23.290243

    work page doi:10.2471/blt.23.290243 1990

  2. [2]

    & Zhang, J

    Xiao, Y., Dai, Q., Liu, R., Ji, D., Lai, X. & Zhang, J. Modeling dynamic flood population exposure in coupled human ‐ water systems: The role of reservoir regulation and population movement. Water Resources Research 61 , e2025WR041234 (2025). https://doi.org/10.1029/2025WR041234

    work page doi:10.1029/2025wr041234 2025

  3. [3]

    N., Allen, G

    Shen, X., Anagnostou, E. N., Allen, G. H., Brakenridge, G. R. & Kettner, A. J. Near-real-time non-obstructed flood inundation mapping using synthetic aperture radar. Remote Sensing of Environment 221 , 302-315 (2019). https://doi.org/10.1016/j.rse.2018.11.008

    work page doi:10.1016/j.rse.2018.11.008 2019

  4. [4]

    and Huang, Z

    Wu, H., Kimball, J.S., Zhou, N., Alfieri, L., Luo, L., Du, J. and Huang, Z. Evaluation of real-time global flood modeling with satellite surface inundation observations from SMAP. Remote Sensing of Environment 233 , 111360 (2019). https://doi.org/10.1016/j.rse.2019.111360

    work page doi:10.1016/j.rse.2019.111360 2019

  5. [5]

    The Harmonized Landsat and Sentinel-2 surface reflectance data set,

    Claverie, M., Ju, J., Masek, J.G., Dungan, J.L., Vermote, E.F., Roger, J.C., Skakun, S.V. and Justice, C. The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sensing of Environment 219 , 145-161 (2018). https://doi.org/10.1016/j.rse.2018.09.002

    work page doi:10.1016/j.rse.2018.09.002 2018

  6. [6]

    & Gamba, P

    Ban, Y., Jacob, A. & Gamba, P. Spaceborne SAR data for global urban mapping at 30 m resolution using a robust urban extractor. ISPRS Journal of Photogrammetry and Remote Sensing 103 , 28-37 (2017). https://doi.org/10.1016/j.isprsjprs.2014.08.004

    work page doi:10.1016/j.isprsjprs.2014.08.004 2017

  7. [7]

    G., Broich, M., Stehman, S

    Tulbure, M. G., Broich, M., Stehman, S. V. & Kommareddy, A. Surface water extent dynamics from three decades of seasonally continuous Landsat time series at subcontinental scale in a semi-arid region. Remote Sensing of Environment 178 , 142-157 (2022). https://doi.org/10.1016/j.rse.2016.02.034

    work page doi:10.1016/j.rse.2016.02.034 2022

  8. [8]

    Munasinghe, D., Frasson, R. P. D. M., David, C. H., Bonnema, M., Schumann, G. & Brakenridge, G. R. A multi-sensor approach for increased measurements of floods and their societal impacts from space. Communications Earth & Environment 4 , 462 (2023). https://doi.org/10.1038/s43247-023-01129-1 16

    work page doi:10.1038/s43247-023-01129-1 2023

  9. [9]

    I., Tsagkatakis, G., Fotiadou, K

    Drakonakis, G. I., Tsagkatakis, G., Fotiadou, K. & Tsakalides, P. Ombrianet—supervised flood mapping via convolutional neural networks using multitemporal sentinel-1 and sentinel-2 data fusion. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 15 , 2341-2356 (2022). https://doi.org/10.1109/JSTARS.2022.3155559

    work page doi:10.1109/jstars.2022.3155559 2022

  10. [10]

    Bentivoglio, R., Isufi, E., Jonkman, S. N. & Taormina, R. Deep learning methods for flood mapping: a review of existing applications and future research directions. Hydrology and Earth System Sciences 26 , 4345-4378 (2022). https://doi.org/10.5194/hess-26-4345-2022

    work page doi:10.5194/hess-26-4345-2022 2022

  11. [11]

    & Bromley, L

    Nemni, E., Bullock, J., Belabbes, S. & Bromley, L. Fully convolutional neural network for rapid flood segmentation in synthetic aperture radar imagery. Remote Sensing 12 , 2532 (2020). https://doi.org/10.3390/rs12162532

    work page doi:10.3390/rs12162532 2020

  12. [12]

    & Gómez-Chova, L

    Portalés-Julià, E., Mateo-García, G. & Gómez-Chova, L. Understanding flood detection models across Sentinel-1 and Sentinel-2 modalities and benchmark datasets. Remote Sensing of Environment 328 , 114882 (2025). https://doi.org/10.1016/j.rse.2025.114882

    work page doi:10.1016/j.rse.2025.114882 2025

  13. [13]

    and Trital, P

    Shinde, R., Ankur, K., Phillips, C.E., Gupta, A., Pfreundschuh, S., Roy, S., Kirkland, S., Gaur, V., Kolluru, V., Lin, A. and Trital, P. WxC-Bench: A novel dataset for weather and climate downstream tasks. Scientific Data (2026). https://doi.org/10.1038/s41597-026-06839-7

    work page doi:10.1038/s41597-026-06839-7 2026

  14. [14]

    and Brynjolfsson, E

    Bommasani, R., Hudson, D.A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M.S., Bohg, J., Bosselut, A., Brunskill, E. and Brynjolfsson, E. On the opportunities and risks of foundation models. (2021). Preprint at https://arxiv.org/abs/2108.07258

    Pith/arXiv arXiv 2021

  15. [15]

    and Huo, Y

    Lu, S., Guo, J., Zimmer-Dauphinee, J.R., Nieusma, J.M., Wang, X., VanValkenburgh, P., Wernke, S.A. and Huo, Y. Vision foundation models in remote sensing: A survey. IEEE Geoscience and Remote Sensing Magazine (2025). https://doi.org/10.1109/MGRS.2025.3541952

    work page doi:10.1109/mgrs.2025.3541952 2025

  16. [16]

    and Kimura, D

    Jakubik, J., Roy, S., Phillips, C.E., Fraccaro, P., Godwin, D., Zadrozny, B., Szwarcman, D., Gomes, C., Nyirjesy, G., Edwards, B. and Kimura, D. Foundation models for generalist geospatial artificial intelligence. arXiv preprint. https://doi.org/10.48550/arXiv.2310.18660

    work page doi:10.48550/arxiv.2310.18660

  17. [17]

    and Chakraborty, S

    Szwarcman, D., Roy, S., Fraccaro, P., Gíslason, O.E., Blumenstiel, B., Ghosal, R., De Oliveira, P.H., de Sousa Almeida, J.L., Sedona, R., Kang, Y. and Chakraborty, S. Prithvi-EO-2.0: A Versatile Multitemporal Foundation Model for Earth Observation Applications. IEEE Transactions on Geoscience and Remote Sensing 64 , 1-20 (2026). https://doi.org/10.1109/TG...

    work page doi:10.1109/tgrs.2025.3642610 2026

  18. [18]

    Adaptive Weighted Attention Network with Camera Spectral Sensitivity Prior for Spectral Reconstruction from RGB Images,

    Bonafilia, D., Tellman, B., Anderson, T. & Issenberg, E. Sen1Floods11: a georeferenced dataset to train and test deep learning flood algorithms for Sentinel-1. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (pp. 835–845). IEEE (2020). https://doi.org/10.1109/CVPRW50498.2020.00113

    work page doi:10.1109/cvprw50498.2020.00113 2020

  19. [19]

    Li, W., Lee, H., Wang, S., Hsu, C. Y. & Arundel, S. T. Assessment of a new GeoAI foundation model for flood inundation mapping. In Proceedings of the 6th ACM SIGSPATIAL International workshop on AI for geographic knowledge discovery (pp. 102-109) (2023). https://doi.org/10.1145/3615886.3627747 17

    work page doi:10.1145/3615886.3627747 2023

  20. [20]

    & Šatanová, A

    Kostejn, J., Janečka, K. & Šatanová, A. U-Prithvi: integrating a foundation model and U-Net for enhanced flood inundation mapping. In GIScience 2025, LIPIcs, Vol. 343, Article 18 (2025). https://doi.org/10.4230/LIPIcs.GIScience.2025.18

    work page doi:10.4230/lipics.giscience.2025.18 2025

  21. [21]

    & Zhang, Z

    Kaushik, S., Maurya, L., Tellman, B. & Zhang, Z. Assessing geo-foundational models for flood inundation mapping: benchmarking models for Sentinel-1, Sentinel-2, and PlanetScope. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 19 , 5649-5665 (2026). https://doi.org/10.1109/JSTARS.2026.3656855

    work page doi:10.1109/jstars.2026.3656855 2026

  22. [22]

    and Atkinson, P.M

    Ghamisi, P., Rasti, B., Yokoya, N., Wang, Q., Hofle, B., Bruzzone, L., Bovolo, F., Chi, M., Anders, K., Gloaguen, R. and Atkinson, P.M. Multisource and multitemporal data fusion in remote sensing: a comprehensive review of the state of the art. IEEE Geoscience and Remote Sensing Magazine 7 , 6-39 (2025). https://doi.org/10.1109/MGRS.2018.2890023

    work page doi:10.1109/mgrs.2018.2890023 2025

  23. [23]

    and Shi, Y

    Zhu, X.X., Xiong, Z., Wang, Y., Stewart, A.J., Heidler, K., Wang, Y., Yuan, Z., Dujardin, T., Xu, Q. and Shi, Y. On the foundations of Earth foundation models. Communications Earth & Environment. 7, 103 (2026). https://doi.org/10.1038/s43247-025-03127-x

    work page doi:10.1038/s43247-025-03127-x 2026

  24. [24]

    and Hostert, P

    Tulbure, M.G., Caineta, J., Broich, M., Gaines, M.D., Rufin, P., Thomas, L.F., Alemohammad, H., Hemmerling, J. and Hostert, P. Leveraging AI multimodal geospatial foundation models for improved near-real-time flood mapping at a global scale. arXiv preprint arXiv:2512.02055 . https://doi.org/10.48550/arXiv.2512.02055

    work page doi:10.48550/arxiv.2512.02055

  25. [25]

    and Lavista, J

    Misra, A., White, K., Nsutezo, S.F., Straka III, W. and Lavista, J. Mapping global floods with 10 years of satellite radar data. Nature Communications 16 , 5762 (2025). https://doi.org/10.1038/s41467-025-60973-1

    work page doi:10.1038/s41467-025-60973-1 2025

  26. [26]

    and Unduche, F

    Li, K., Razavi, S., Maier, H.R., Hrachowitz, M., Nabavi, E., Harvey, N., Akhtar, K. and Unduche, F. When are AI models ready for deployment? Reassessing Google's global AI flood forecasting system through the lens of responsible modelling. Journal of Hydrology X, 100215 (2026). https://doi.org/10.1016/j.hydroa.2026.100215

    work page doi:10.1016/j.hydroa.2026.100215 2026

  27. [28]

    H., Lipton, Z

    Feffer, M., Sinha, A., Deng, W. H., Lipton, Z. C. & Heidari, H. Red-teaming for generative AI: Silver bullet or security theater? In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society 7 421-437 (2024). https://doi.org/10.1609/aies.v7i1.31647

    work page doi:10.1609/aies.v7i1.31647 2024

  28. [29]

    and Lutz, R

    Bullwinkel, B., Minnich, A., Chawla, S., Lopez, G., Pouliot, M., Maxwell, W., de Gruyter, J., Pratt, K., Qi, S., Chikanov, N. and Lutz, R. Lessons from red teaming 100 generative AI products. arXiv preprint arXiv:2501.07238 (2025). https://doi.org/10.48550/arXiv.2501.07238 18

    work page doi:10.48550/arxiv.2501.07238 2025

  29. [30]

    & Chen, Y

    Wang, P., Wu, X. & Chen, Y. AI-driven approaches to flood risk management: overcoming data bias and enhancing decision-making. Climate Risk Management 50 , 100752 (2025). https://doi.org/10.1016/j.crm.2025.100752

    work page doi:10.1016/j.crm.2025.100752 2025

  30. [31]

    I., Yin, K., Li, X

    Liu, Z., Coleman, N., Patrascu, F. I., Yin, K., Li, X. & Mostafavi, A. Artificial intelligence for flood risk management: a comprehensive state-of-the-art review and future directions. International Journal of Disaster Risk Reduction 117 , 105110 (2025). https://doi.org/10.1016/j.ijdrr.2024.105110

    work page doi:10.1016/j.ijdrr.2024.105110 2025

  31. [32]

    and Weng, Q

    Li, Z., Xu, S. and Weng, Q. Beyond clouds: seamless flood mapping using Harmonized Landsat and Sentinel-2 time series imagery and water occurrence data. ISPRS Journal of Photogrammetry and Remote Sensing 216 , 185-199 (2024). https://doi.org/10.1016/j.isprsjprs.2024.07.022

    work page doi:10.1016/j.isprsjprs.2024.07.022 2024

  32. [33]

    & Hosseiny, H

    Mohamadiazar, N., Ebrahimian, A. & Hosseiny, H. Integrating deep learning, satellite image processing, and spatial-temporal analysis for urban flood prediction. Journal of Hydrology 639 , 131508 (2024). https://doi.org/10.1016/j.jhydrol.2024.131508

    work page doi:10.1016/j.jhydrol.2024.131508 2024

  33. [34]

    Marsocci, V. et al. PANGAEA: Assessing Geospatial Foundation Models Capabilities through a Global and Inclusive Benchmark. IEEE geoscience and remote sensing magazine (2025). https://doi.org/10.1109/MGRS.2025.3628194

    work page doi:10.1109/mgrs.2025.3628194 2025

  34. [35]

    and Nevo, S., 2024

    Nearing, G., Cohen, D., Dube, V., Gauch, M., Gilon, O., Harrigan, S., Hassidim, A., Klotz, D., Kratzert, F., Metzger, A. and Nevo, S., 2024. Global prediction of extreme floods in ungauged watersheds. Nature 627 , 559-563 (2024). https://doi.org/10.1038/s41586-024-07145-1

    work page doi:10.1038/s41586-024-07145-1 2024

  35. [36]

    & Woodcock, C

    Zhu, Z., Wang, S. & Woodcock, C. E. Improvement and expansion of the Fmask algorithm: cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel 2 images. Remote Sensing of Environment 159 , 269-277 (2015). https://doi.org/10.1016/j.rse.2014.12.014

    work page doi:10.1016/j.rse.2014.12.014 2015

  36. [37]

    and Chini, M

    Salamon, P., Mctlormick, N., Reimer, C., Clarke, T., Bauer-Marschallinger, B., Wagner, W., Martinis, S., Chow, C., Böhnke, C., Matgen, P. and Chini, M. The new, systematic global flood monitoring product of the Copernicus Emergency Management Service. In 2021 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 1053–1056 (2021). https://doi...

    work page doi:10.1109/igarss47720.2021.9554214 2021

  37. [38]

    OPERA Dynamic Surface Water Extent from Harmonized Landsat Sentinel-2 Version 1

    OPERA. OPERA Dynamic Surface Water Extent from Harmonized Landsat Sentinel-2 Version 1. NASA Physical Oceanography Distributed Active Archive Center (2023). https://doi.org/10.5067/opdsw-pl3v1

    work page doi:10.5067/opdsw-pl3v1 2023

  38. [39]

    Cédric Villani.Optimal Transport: Old and New

    He, K., Chen, X., Xie, S., Li, Y., Dollár, P. & Girshick, R. Masked autoencoders are scalable vision learners. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 16000–16009). IEEE (2022). https://doi.org/10.1109/CVPR52688.2022.01553

    work page doi:10.1109/cvpr52688.2022.01553 2022

  39. [40]

    and Ganzorig, U., 2026

    Kolluru, V., John, R., Chen, J., Henebry, G.M., Xiao, J., Shinde, R., Kussainova, M. and Ganzorig, U., 2026. Leveraging sUAS-Sentinel-2 synergy for cross-scale mapping of canopy cover and aboveground biomass across Mongolia and Kazakhstan. Remote Sensing of Environment, 336, p.115302. https://doi.org/10.1016/j.rse.2026.115302

    work page doi:10.1016/j.rse.2026.115302 2026

  40. [41]

    and Kussainova, M., 2024

    Kolluru, V., John, R., Chen, J., Konkathi, P., Kolluru, S., Saraf, S., Henebry, G.M., Xiao, J., Jain, K. and Kussainova, M., 2024. Dominant role of grazing and snow cover variability on vegetation shifts in 19 the drylands of Kazakhstan. Communications Earth & Environment, 5(1), p.424. https://doi.org/10.1038/s43247-024-01587-1

    work page doi:10.1038/s43247-024-01587-1 2024

  41. [42]

    and Yuan, J., 2022

    Venkatesh, K., John, R., Chen, J., Jarchow, M., Amirkhiz, R.G., Giannico, V., Saraf, S., Jain, K., Kussainova, M. and Yuan, J., 2022. Untangling the impacts of socioeconomic and climatic changes on vegetation greenness and productivity in Kazakhstan. Environmental Research Letters, 17(9), p.095007. https://doi.org/10.1088/1748-9326/ac8c59

    work page doi:10.1088/1748-9326/ac8c59 2022

  42. [43]

    & Belward, A

    Pekel, J.-F., Cottam, A., Gorelick, N. & Belward, A. S. High-resolution mapping of global surface water and its long-term changes. Nature 540 , 418-422 (2016). https://doi.org/10.1038/nature20584

    work page doi:10.1038/nature20584 2016

  43. [44]

    and Lesiv, M

    Zanaga, D., Van De Kerchove, R., Daems, D., De Keersmaecker, W., Brockmann, C., Kirches, G., Wevers, J., Cartus, O., Santoro, M., Fritz, S. and Lesiv, M. ESA WorldCover 10 m 2021 v200. Zenodo (2022). https://doi.org/10.5281/zenodo.7254221 20

    work page doi:10.5281/zenodo.7254221 2021