pith. sign in

arxiv: 2606.29447 · v1 · pith:WW7T4JV3new · submitted 2026-06-28 · 💻 cs.CV

Miti360: A Comprehensive Dataset for Improved Reforestation Monitoring

Cedric Kiplimo , Samuel Mbatia , Ciira wa Maina , Arthur Sichangi , Dennis Gitundu This is my paper

Pith reviewed 2026-06-30 07:41 UTC · model grok-4.3

classification 💻 cs.CV
keywords reforestation monitoringmachine learning datasettree detectiondrone imageryAfrican forestrybounding box annotationslongitudinal data
0
0 comments X

The pith

A dataset of drone imagery and tree annotations from a Kenyan forest site allows machine learning models to track tree crowns over three years and boosts detection performance when used for fine-tuning.

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

The paper presents Miti360, a new collection of high-resolution aerial and terrestrial images with tree bounding box annotations from a reforested area in Kenya. Existing machine learning datasets for forest monitoring come mostly from other continents, creating a gap for African contexts. By including longitudinal data over three years along with weather records, the dataset supports tasks like tree census automation and growth modeling. The authors show its value by tracking crowns across years and fine-tuning an existing model to raise box precision by 12 percent and recall by 69 percent.

Core claim

Miti360 supplies orthophotos, tiled images, stereo terrestrial photos, tree biophysical parameters, species labels, GPS locations, and historical weather data from a 770-hectare site in Kieni Forest, Kenya, collected between 2023 and 2025. This enables machine learning systems for reforestation monitoring in Sub-Saharan Africa, as demonstrated by successful multi-year tree crown tracking and improved performance of the DeepForest model after fine-tuning.

What carries the argument

The Miti360 dataset, which pairs high-resolution drone orthophotos and ground-truth tree bounding boxes with terrestrial imagery, species data, and weather records to support model training for African forest monitoring.

If this is right

  • Machine learning models can be trained to accelerate tree censuses in similar reforestation projects.
  • Species matching to geographical areas becomes feasible with the included annotations.
  • Growth modeling based on weather conditions can use the longitudinal records.
  • Digital twin frameworks for forests can be developed from the multi-modal data.
  • Fine-tuning on this data improves box precision by 12% and recall by 69% on tree detection tasks.

Where Pith is reading between the lines

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

  • Similar datasets from other African sites could test whether the improvements generalize beyond the single Kieni location.
  • The three-year span of imagery might allow studies of how specific weather patterns affect crown growth rates.
  • Combining the aerial and stereo terrestrial images could support 3D reconstruction of individual trees for volume estimation.

Load-bearing premise

The annotation process and data from this one 770-hectare Kenyan site using drone stitching will produce labels that work reliably for machine learning training across other African reforestation areas.

What would settle it

A test showing that the fine-tuned DeepForest model performs no better than the original on imagery from a different African forest site would indicate the dataset does not generalize as claimed.

read the original abstract

Over the past decade, interest in applying machine learning (ML) to automate forest monitoring has grown significantly. However, existing training datasets are predominantly drawn from North America, Europe, Asia, and Australia, leaving a critical gap in African forestry data. To address this limited geographic diversity, we present Miti360, a comprehensive dataset for reforestation monitoring that comprises high-resolution imagery, ground truth data, and longitudinal weather data. Data collection occurred within a 770-ha reforested section of the Kieni Forest in Kenya between March 2023 and February 2025. Miti360 comprises aerial photos (orthophotos and tiles) with tree bounding box annotations, terrestrial images (single and stereo), and detailed data records including tree biophysical parameters, species, and GPS coordinates, alongside historical weather data. Aerial surveys utilized a DJI Mavic 2 Pro, with imagery stitched via Agisoft Metashape and tiled using ArcGIS Pro, while terrestrial captures used smartphones and custom stereo cameras. Miti360 enables the training of ML systems for tasks such as accelerating tree censuses, matching species to geographical areas, modelling growth based on weather conditions, and developing digital twin frameworks. Models can be trained on Miti360 to address challenges specific to Sub-Saharan Africa, ultimately advancing reforestation monitoring and fostering sustainable forestry practices in underrepresented regions. We demonstrate the utility of this dataset by successfully tracking tree crowns across three years and improving the DeepForest model's box precision and box recall by 12% and 69% respectively through fine-tuning on Miti360.

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 paper presents Miti360, a dataset of high-resolution aerial orthophotos/tiles with tree bounding-box annotations, terrestrial images, biophysical parameters, species labels, GPS coordinates, and historical weather data collected from a single 770-ha reforested site in Kieni Forest, Kenya (March 2023–February 2025) using DJI Mavic imagery stitched via Agisoft Metashape. It claims the dataset fills a geographic gap in African forestry ML data and demonstrates utility via three-year tree-crown tracking plus fine-tuning of DeepForest that improves box precision by 12% and box recall by 69%.

Significance. A publicly released, well-documented dataset from an underrepresented Sub-Saharan African reforestation context would be a useful addition to the forestry-ML literature, especially given the inclusion of longitudinal weather and biophysical metadata that could support growth modeling. The multi-modal collection (aerial + terrestrial + metadata) is a positive design choice.

major comments (2)
  1. [Abstract] Abstract: the claimed 12% precision / 69% recall gains after fine-tuning DeepForest are presented with no experimental protocol, no description of train/validation/test splits, no statement of whether the test set is held-out from the same 770-ha site, no baseline comparisons beyond the original model, and no error analysis. These details are load-bearing for the utility demonstration.
  2. [Dataset Description] Dataset collection and annotation sections: all imagery originates from one Kenyan site and is produced by Agisoft Metashape stitching of DJI Mavic frames; no analysis of potential stitching artifacts, illumination/site-specific biases, or species/terrain variability is supplied, undermining the claim that the dataset supports reliable ML training for other Sub-Saharan reforestation contexts.
minor comments (1)
  1. [Abstract] Clarify the exact temporal span of the three-year crown-tracking experiment relative to the March 2023–February 2025 collection window.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major point below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claimed 12% precision / 69% recall gains after fine-tuning DeepForest are presented with no experimental protocol, no description of train/validation/test splits, no statement of whether the test set is held-out from the same 770-ha site, no baseline comparisons beyond the original model, and no error analysis. These details are load-bearing for the utility demonstration.

    Authors: We agree these experimental details are essential and were omitted from the abstract. In the revision we will expand the abstract to reference the protocol, specify the train/validation/test splits with confirmation that the test set is held-out from the 770-ha site, note additional baseline comparisons, and mention the error analysis. A new subsection in the main text will provide the full protocol and results. revision: yes

  2. Referee: [Dataset Description] Dataset collection and annotation sections: all imagery originates from one Kenyan site and is produced by Agisoft Metashape stitching of DJI Mavic frames; no analysis of potential stitching artifacts, illumination/site-specific biases, or species/terrain variability is supplied, undermining the claim that the dataset supports reliable ML training for other Sub-Saharan reforestation contexts.

    Authors: We acknowledge that all data come from a single site and that the current text provides no explicit analysis of stitching artifacts or biases. We will add a dedicated limitations paragraph discussing potential Agisoft Metashape artifacts, illumination and terrain biases, and species variability. We will also moderate claims of broad applicability to other Sub-Saharan contexts while retaining the value of the data as an initial resource from an underrepresented region. revision: yes

Circularity Check

0 steps flagged

No circularity; dataset paper with empirical demonstration only

full rationale

The paper presents Miti360 as a new dataset collected from one Kenyan site and reports measured improvements (12% precision, 69% recall) after fine-tuning DeepForest on it, along with three-year crown tracking. No equations, derivations, fitted parameters renamed as predictions, self-citations, or uniqueness theorems appear in the provided text. The central claim is an empirical result on the introduced data, which is the standard non-circular contribution of a dataset paper and remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Dataset creation paper. No free parameters, mathematical axioms, or invented entities are required to support the central claim of providing new training data.

pith-pipeline@v0.9.1-grok · 5827 in / 1081 out tokens · 45401 ms · 2026-06-30T07:41:39.459903+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

45 extracted references · 41 canonical work pages

  1. [1]

    An assessment of data sources, data quality and changes in national forest monitoring capacities in the Global Forest Resources Assessment 2005 – 2020,

    K. Nesha et al. , “An assessment of data sources, data quality and changes in national forest monitoring capacities in the Global Forest Resources Assessment 2005 – 2020,” Environmental Research Letters , vol. 16, no. 5, p. 054029, May 2021, doi: 10.1088/1748 - 9326/abd81b

    work page doi:10.1088/1748 2005

  2. [2]

    An Inventory of the Above Ground Biomass in the Mau Forest Ecosystem, Kenya,

    M. J. Kinyanjui, P. Latva - Käyrä, P. S. Bhuwneshwar, P. Kariuki, A. Gichu, and K. Wamichwe, “An Inventory of the Above Ground Biomass in the Mau Forest Ecosystem, Kenya,” Open J. Ecol. , vol. 04, no. 10, pp. 619 – 627, 2014, doi: 10.4236/oje.2014.410052

    work page doi:10.4236/oje.2014.410052 2014

  3. [3]

    doi: 10.1101/2020.09.08.287839

    work page doi:10.1101/2020.09.08.287839 2020

  4. [4]

    Ego4d: Around the world in 3, 000 hours of egocentric video

    S. Beery et al. , “The Auto Arborist Dataset: A Large - Scale Benchmark for Multiview Urban Forest Monitoring Under Domain Shift,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , IEEE, Jun. 2022, pp. 21262 – 21275. doi: 10.1109/CVPR52688.2022.02061

    work page doi:10.1109/cvpr52688.2022.02061 2022

  5. [5]

    OpenForest: a data catalog for machine learning in forest monitoring,

    A. Ouaknine, T. Kattenborn, E. Laliberté, and D. Rolnick, “OpenForest: a data catalog for machine learning in forest monitoring,” Environmental Data Science , vol. 4, p. e15, Feb. 2025, doi: 10.1017/eds.2024.53

    work page doi:10.1017/eds.2024.53 2025

  6. [6]

    Seeing the Trees from Above: A Survey on Real and Synthetic Agroforestry Datasets for Remote Sensing Applications,

    B. Chehreh, A. Moutinho, and C. Viegas, “Seeing the Trees from Above: A Survey on Real and Synthetic Agroforestry Datasets for Remote Sensing Applications,” Remote Sens. (Basel). , vol. 17, no. 19, p. 3346, Oct. 2025, doi: 10.3390/rs17193346

    work page doi:10.3390/rs17193346 2025

  7. [7]

    Deep learning enables image - based tree counting, crown segmentation, and height prediction at national scale,

    S. Li et al. , “Deep learning enables image - based tree counting, crown segmentation, and height prediction at national scale,” PNAS Nexus , vol. 2, no. 4, Apr. 2023, doi: 10.1093/pnasnexus/pgad076

    work page doi:10.1093/pnasnexus/pgad076 2023

  8. [8]

    A Systematic Review of Individual Tree Crown Detection and Delineation with Convolutional Neural Networks (CNN),

    H. Zhao, J. Morgenroth, G. Pearse, and J. Schindler, “A Systematic Review of Individual Tree Crown Detection and Delineation with Convolutional Neural Networks (CNN),” Current Forestry Reports , vol. 9, no. 3, pp. 149 – 170, Apr. 2023, doi: 10.1007/s40725 - 023 - 00184 -

    work page doi:10.1007/s40725 2023

  9. [9]

    Two - Stage Deep Learning Framework for Individual Tree Crown Detection and Delineation in Mixed - Wood Forests Using High - Resolution Light Detection and Ranging Data,

    Q. Li, B. Hu, J. Shang, and T. K. Remmel, “Two - Stage Deep Learning Framework for Individual Tree Crown Detection and Delineation in Mixed - Wood Forests Using High - Resolution Light Detection and Ranging Data,” Remote Sens. (Basel). , vol. 17, no. 9, p. 1578, Apr. 2025, doi: 10.3390/rs17091578

    work page doi:10.3390/rs17091578 2025

  10. [10]

    Above ground biomass estimation from UAV high resolution RGB images and LiDAR data in a pine forest in Southern Italy,

    M. Maesano, G. Santopuoli, F. Moresi, G. Matteucci, B. Lasserre, and G. Scarascia Mugnozza, “Above ground biomass estimation from UAV high resolution RGB images and LiDAR data in a pine forest in Southern Italy,” IForest , vol. 15, no. 6, pp. 451 – 457, Dec. 2022, doi: 10.3832/ifor3781 -

    work page doi:10.3832/ifor3781 2022

  11. [11]

    Forest Aboveground Biomass Estimation Based on Unmanned Aerial Vehicle – Light Detection and Ranging and Machine Learning,

    Y. Yan, J. Lei, and Y. Huang, “Forest Aboveground Biomass Estimation Based on Unmanned Aerial Vehicle – Light Detection and Ranging and Machine Learning,” Sensors , vol. 24, no. 21, p. 7071, Nov. 2024, doi: 10.3390/s24217071

    work page doi:10.3390/s24217071 2024

  12. [12]

    Above ground tree biomass modeling using machine learning algorithms in western Terai Sal Forest of Nepal,

    B. Singh, A. K. Verma, K. Tiwari, and R. Joshi, “Above ground tree biomass modeling using machine learning algorithms in western Terai Sal Forest of Nepal,” Heliyon , vol. 9, no. 11, p. e21485, Nov. 2023, doi: 10.1016/j.heliyon.2023.e21485

    work page doi:10.1016/j.heliyon.2023.e21485 2023

  13. [13]

    ReforesTree: A Dataset for Estimating Tropical Forest Carbon Stock with Deep Learning and Aerial Imagery,

    G. Reiersen et al. , “ReforesTree: A Dataset for Estimating Tropical Forest Carbon Stock with Deep Learning and Aerial Imagery,” Proceedings of the AAAI Conference on Artificial Intelligence , vol. 36, no. 11, pp. 12119 – 12125, Jun. 2022, doi: 10.1609/aaai.v36i11.21471

    work page doi:10.1609/aaai.v36i11.21471 2022

  14. [14]

    Deep Learning with Northern Australian Savanna Tree Species: A Novel Dataset,

    A. J. Jansen et al. , “Deep Learning with Northern Australian Savanna Tree Species: A Novel Dataset,” Data (Basel). , vol. 8, no. 2, p. 44, Feb. 2023, doi: 10.3390/data8020044

    work page doi:10.3390/data8020044 2023

  15. [15]

    M., & von Bloh, W

    D. Koldasbayeva, P. Tregubova, M. Gasanov, A. Zaytsev, A. Petrovskaia, and E. Burnaev, “Challenges in data - driven geospatial modeling for environmental research and practice,” Nat. Commun. , vol. 15, no. 1, p. 10700, Dec. 2024, doi: 10.1038/s41467 - 024 - 55240 -

    work page doi:10.1038/s41467 2024

  16. [16]

    The First Open Dataset of Mangrove Above - and Belowground Biomass and Soil Carbon Stocks in Côte d’Ivoire: Insights from Fresco and Sassandra,

    S. Kouassi, F. Zapfack, T. Temfack, G. Fonlladosa, and K. Kassoum, “The First Open Dataset of Mangrove Above - and Belowground Biomass and Soil Carbon Stocks in Côte d’Ivoire: Insights from Fresco and Sassandra,” 2025, Figshare

    2025

  17. [17]

    Phenological Dataset for Ecological Forecasting,

    B. Ofosu - Bamfo et al. , “Phenological Dataset for Ecological Forecasting,” 2025, Zenodo

    2025

  18. [18]

    Advancements in Forest Monitoring: Applications and Perspectives of Airborne Laser Scanning and Complementarity with Satellite Optical Data,

    C. Borghi, S. Francini, G. D’Amico, R. Valbuena, and G. Chirici, “Advancements in Forest Monitoring: Applications and Perspectives of Airborne Laser Scanning and Complementarity with Satellite Optical Data,” Land (Basel). , vol. 14, no. 3, p. 567, Mar. 2025, doi: 10.3390/land14030567

    work page doi:10.3390/land14030567 2025

  19. [19]

    Enhanced forest inventories in Canada: implementation, status, and research needs,

    J. C. White et al. , “Enhanced forest inventories in Canada: implementation, status, and research needs,” Canadian Journal of Forest Research , vol. 55, pp. 1 – 37, Jan. 2025, doi: 10.1139/cjfr - 2024 -

    work page doi:10.1139/cjfr 2025

  20. [20]

    Improving availability and utilization of forest inventory and land use map data using Linked Open Data,

    J. M. Giménez - García, G. Vega - Gorgojo, C. Ordóñez, N. Crespo - Lera, and F. Bravo, “Improving availability and utilization of forest inventory and land use map data using Linked Open Data,” Frontiers in Forests and Global Change , vol. 7, Sep. 2024, doi: 10.3389/ffgc.2024.1329812

    work page doi:10.3389/ffgc.2024.1329812 2024

  21. [21]

    Advancing forest inventorying and monitoring,

    M. Ferretti et al. , “Advancing forest inventorying and monitoring,” Ann. For. Sci. , vol. 81, no. 1, p. 6, Feb. 2024, doi: 10.1186/s13595 - 023 - 01220 -

    work page doi:10.1186/s13595 2024

  22. [22]

    Monitoring forest dynamics in Africa during 2000 – 2020 using a remotely sensed fractional tree cover dataset,

    X. Wei et al. , “Monitoring forest dynamics in Africa during 2000 – 2020 using a remotely sensed fractional tree cover dataset,” Int. J. Digit. Earth , vol. 16, no. 1, pp. 2212 – 2232, Oct. 2023, doi: 10.1080/17538947.2023.2220613. 23

    work page doi:10.1080/17538947.2023.2220613 2000

  23. [23]

    Exploring sustainability challenges in national forest inventories: Tunisia and Mongolia as case studies,

    P. Knostmann, T. Neeff, M. Krott, and C. Kleinn, “Exploring sustainability challenges in national forest inventories: Tunisia and Mongolia as case studies,” Forests Monitor , vol. 2, no. 1, pp. 1 – 48, Feb. 2025, doi: 10.62320/fm.v2i1.16

    work page doi:10.62320/fm.v2i1.16 2025

  24. [24]

    doi: 10.21203/rs.3.rs - 8036525/v1

    work page doi:10.21203/rs.3.rs

  25. [25]

    Mapping deforestation and recovery of tropical montane forests of East Africa,

    S. Ojoatre, C. Zhang, G. Yesuf, and M. C. Rufino, “Mapping deforestation and recovery of tropical montane forests of East Africa,” Front. Environ. Sci. , vol. 11, Jul. 2023, doi: 10.3389/fenvs.2023.1084764

    work page doi:10.3389/fenvs.2023.1084764 2023

  26. [26]

    Available: https://cabinetoffice.go.ke/sites/default/files/2025 - 04/Despatch%20from %20Cabinet%20 - %2006.12.22.pdf

    [Online]. Available: https://cabinetoffice.go.ke/sites/default/files/2025 - 04/Despatch%20from %20Cabinet%20 - %2006.12.22.pdf

    2025

  27. [27]

    Tree detection and diameter estimation based on deep learning,

    V. Grondin, J. - M. Fortin, F. Pomerleau, and P. Giguère, “Tree detection and diameter estimation based on deep learning,” Forestry: An International Journal of Forest Research , vol. 96, no. 2, pp. 264 – 276, Mar. 2023, doi: 10.1093/forestry/cpac043

    work page doi:10.1093/forestry/cpac043 2023

  28. [28]

    Individual Tree Diameter Growth Models of Larch – Spruce – Fir Mixed Forests Based on Machine Learning Algorithms,

    Q. Ou, X. Lei, and C. Shen, “Individual Tree Diameter Growth Models of Larch – Spruce – Fir Mixed Forests Based on Machine Learning Algorithms,” Forests , vol. 10, no. 2, p. 187, Feb. 2019, doi: 10.3390/f10020187

    work page doi:10.3390/f10020187 2019

  29. [29]

    Detecting and mapping tree seedlings in UAV imagery using convolutional neural networks and field - verified data,

    G. D. Pearse, A. Y. S. Tan, M. S. Watt, M. O. Franz, and J. P. Dash, “Detecting and mapping tree seedlings in UAV imagery using convolutional neural networks and field - verified data,” ISPRS Journal of Photogrammetry and Remote Sensing , vol. 168, pp. 156 – 169, Oct. 2020, doi: 10.1016/j.isprsjprs.2020.08.005

    work page doi:10.1016/j.isprsjprs.2020.08.005 2020

  30. [30]

    Advances in Digital Twin Technology in Industry: A Review of Applications, Challenges, and Standardization,

    L. Ai and P. Ziehl, “Advances in Digital Twin Technology in Industry: A Review of Applications, Challenges, and Standardization,” Journal of Intelligent Construction , vol. 3, no. 2, pp. 1 – 19, Jun. 2025, doi: 10.26599/Jic.2025.9180083

    work page doi:10.26599/jic.2025.9180083 2025

  31. [31]

    Digital twin for smart manufacturing, A review,

    M. Soori, B. Arezoo, and R. Dastres, “Digital twin for smart manufacturing, A review,” Sustainable Manufacturing and Service Economics , vol. 2, p. 100017, Apr. 2023, doi: 10.1016/j.smse.2023.100017

    work page doi:10.1016/j.smse.2023.100017 2023

  32. [32]

    Digital twin in manufacturing: conceptual framework and case studies,

    I. Onaji, D. Tiwari, P. Soulatiantork, B. Song, and A. Tiwari, “Digital twin in manufacturing: conceptual framework and case studies,” Int. J. Comput. Integr. Manuf. , vol. 35, no. 8, pp. 831 – 858, Aug. 2022, doi: 10.1080/0951192X.2022.2027014

    work page doi:10.1080/0951192x.2022.2027014 2022

  33. [33]

    A Proposal for a Forest Digital Twin Framework and Its Perspectives,

    L. Buonocore, J. Yates, and R. Valentini, “A Proposal for a Forest Digital Twin Framework and Its Perspectives,” Forests , vol. 13, no. 4, p. 498, Mar. 2022, doi: 10.3390/f13040498. 24

    work page doi:10.3390/f13040498 2022

  34. [35]

    Forestry Digital Twin With Machine Learning in Landsat 7 Data,

    X. Jiang, M. Jiang, Y. Gou, Q. Li, and Q. Zhou, “Forestry Digital Twin With Machine Learning in Landsat 7 Data,” Front. Plant Sci. , vol. 13, Jun. 2022, doi: 10.3389/fpls.2022.916900

    work page doi:10.3389/fpls.2022.916900 2022

  35. [36]

    Editorial: Digital twins of plant and forest,

    Z. Lv, H. H. Song, J. Shen, and N. Vaughan, “Editorial: Digital twins of plant and forest,” Front. Plant Sci. , vol. 13, Dec. 2022, doi: 10.3389/fpls.2022.1049240

    work page doi:10.3389/fpls.2022.1049240 2022

  36. [37]

    A Digital Twin Architecture for Forest Restoration: Integrating AI, IoT, and Blockchain for Smart Ecosystem Management,

    N. Sasaki and I. Abe, “A Digital Twin Architecture for Forest Restoration: Integrating AI, IoT, and Blockchain for Smart Ecosystem Management,” Future Internet , vol. 17, no. 9, p. 421, Sep. 2025, doi: 10.3390/fi17090421

    work page doi:10.3390/fi17090421 2025

  37. [38]

    Digital Twins in Agriculture and Forestry: A Review,

    A. C. Tagarakis, L. Benos, G. Kyriakarakos , S. Pearson, C. G. Sørensen, and D. Bochtis, “Digital Twins in Agriculture and Forestry: A Review,” Sensors , vol. 24, no. 10, p. 3117, May 2024, doi: 10.3390/s24103117

    work page doi:10.3390/s24103117 2024

  38. [39]

    Tropical Forest Monitoring: Challenges and Recent Progress in Research,

    J. Murrins Misiukas, S. Carter, and M. Herold, “Tropical Forest Monitoring: Challenges and Recent Progress in Research,” Remote Sens. (Basel). , vol. 13, no. 12, p. 2252, Jun. 2021, doi: 10.3390/rs13122252

    work page doi:10.3390/rs13122252 2021

  39. [40]

    Available: https://agu.confex.com/agu/agu25/meetingapp.cgi/Paper/1987602

    [Online]. Available: https://agu.confex.com/agu/agu25/meetingapp.cgi/Paper/1987602

    work page arXiv

  40. [41]

    Ecosystem service preferences across multilevel stakeholders in co - managed forests: Case of Aberdare protected forest ecosystem in Kenya,

    F. Kilonzi and T. Ota, “Ecosystem service preferences across multilevel stakeholders in co - managed forests: Case of Aberdare protected forest ecosystem in Kenya,” One Ecosystem , vol. 4, Aug. 2019, doi: 10.3897/oneeco.4.e36768

    work page doi:10.3897/oneeco.4.e36768 2019

  41. [42]

    DSAIL - TreeVision: A software tool for extracting tree biophysical parameters from stereoscopic images,

    C. Kiplimo, C. E. Epege, C. wa Maina, and B. Okal, “DSAIL - TreeVision: A software tool for extracting tree biophysical parameters from stereoscopic images,” SoftwareX , vol. 26, p. 101661, May 2024, doi: 10.1016/j.softx.2024.101661

    work page doi:10.1016/j.softx.2024.101661 2024

  42. [43]

    The Trans‐African Hydro - Meteorological Observatory (TAHMO),

    N. van de Giesen, R. Hut, and J. Selker, “The Trans‐African Hydro - Meteorological Observatory (TAHMO),” WIREs Water , vol. 1, no. 4, pp. 341 – 348, Jul. 2014, doi: 10.1002/wat2.1034

    work page doi:10.1002/wat2.1034 2014

  43. [44]

    A two - dimensional interpolation function for irregularly - spaced data,

    D. Shepard, “A two - dimensional interpolation function for irregularly - spaced data,” in Proceedings of the 1968 23rd ACM national conference on - , New York, New York, USA: ACM Press, 1968, pp. 517 –

    1968

  44. [45]

    doi: 10.1145/800186.810616

    work page doi:10.1145/800186.810616

  45. [46]

    Low - Cost Non - Contact Forest Inventory: A Case Study of Kieni Forest in Kenya,

    C. Kiplimo, C. wa Maina, and B. Okal, “Low - Cost Non - Contact Forest Inventory: A Case Study of Kieni Forest in Kenya,” Challenges , vol. 15, no. 1, p. 16, Mar. 2024, doi: 10.3390/challe15010016

    work page doi:10.3390/challe15010016 2024