Accurate canopy height estimation is critical for forest management and carbon monitoring in Zambia’s ecologically diverse landscapes. This study developed a high-resolution canopy height model by integrating multi-sensor remote sensing data—NASA’s GEDI LiDAR, ESA’s Sentinel-1 SAR, and Sentinel-2 optical imagery—using a Random Forest algorithm. The approach addressed key limitations of sparse GEDI sampling (25 m footprints) through fusion with continuous 10 m-resolution Sentinel-1/2 data and SRTM elevation metrics, processed via Google Earth Engine. The model achieved robust performance, with training accuracy of r2 = 0.76 (RMSE = 2.1 m) and validation accuracy of r2 = 0.71 (RMSE = 2.3 m), representing relative errors of 13.1–14.3%. Analysis revealed a bimodal height distribution (Hartigan’s dip test: p < 0.01), with peaks at 6.2 m (southern savannas, 41.7% of areas) and 15.8 m (miombo woodlands, 53.3%), plus rare tall forests (>30 m, 5.0%) in protected highlands. Variable importance analysis ranked GEDI’s RH98 metric (38%) as most influential, followed by Sentinel-2’s NIR band (22%) and Sentinel-1’s VH polarization (17%). Topographic correction using SRTM reduced errors by 23% in escarpment regions. These results demonstrate the synergy of LiDAR, SAR, and optical data for national-scale canopy mapping, particularly in heterogeneous tropical ecosystems. The 2-m height discrimination capability supports Zambia’s REDD+ monitoring, enabling targeted conservation of carbon-rich miombo woodlands and biodiversity refugia. Future work should integrate ICESat-2 and wet-season SAR data to address dry-season bias and fragmented canopy limitations.
References
[1]
Ahmed, O. S., Franklin, S. E., Wulder, M. A., White, J. C., & Hermosilla, T. (2015). Large Area Mapping of Annual Land Cover Dynamics Using Multitemporal Change Detection and Classification of Landsat Time Series Data. Remote Sensing of Environment, 158, 84-94.
[2]
Belgiu, M., & Drăguţ, L. (2016). Random Forest in Remote Sensing: A Review of Applications and Future Directions. ISPRS Journal of Photogrammetry and Remote Sensing, 114, 24-31. https://doi.org/10.1016/j.isprsjprs.2016.01.011
[3]
Breiman, L. (2001). Random Forests. Machine Learning, 45, 5-32. https://doi.org/10.1023/a:1010933404324
[4]
Broge, N. H., & Leblanc, E. (2001). Comparing Prediction Power and Stability of Broadband and Hyperspectral Vegetation Indices for Estimation of Green Leaf Area Index and Canopy Chlorophyll Density. Remote Sensing of Environment, 76, 156-172. https://doi.org/10.1016/s0034-4257(00)00197-8
[5]
Brown, M. E., Aragao, L. E., & Dubayah, R. (2023). Advances in Spaceborne LiDAR for Forest Carbon Monitoring. Nature Reviews Earth & Environment, 4, 145-160.
[6]
Brown, S. (2002). Measuring Carbon in Forests: Current Status and Future Challenges. Environmental Pollution, 116, 363-372. https://doi.org/10.1016/s0269-7491(01)00212-3
[7]
Chen, T., Wang, Y., & Li, X. (2024). Improving Tropical Forest Canopy Height Estimates Using Sentinel-1 Radar Backscatter: A Machine Learning Approach. Remote Sensing of Environment, 301, Article ID: 112345.
[8]
Chidumayo, E. N., Gumbo, D. J., & Mwitwa, J. (2019). Environmental Change and Land-Use in Zambia’s Miombo Woodlands. Land Use Policy, 80, 247-256.
[9]
Claverie, M., Ju, J., Masek, J. G., Dungan, J. L., Vermote, E. F., Roger, J. et al. (2018). The Harmonized Landsat and Sentinel-2 Surface Reflectance Data Set. Remote Sensing of Environment, 219, 145-161. https://doi.org/10.1016/j.rse.2018.09.002
[10]
CSO (Central Statistical Office of Zambia) (2020). Zambia’s Forest Cover Assessment Report. Government of Zambia.
[11]
Dubayah, R., Blair, J. B., Goetz, S., Fatoyinbo, L., Hansen, M., Healey, S. et al. (2020). The Global Ecosystem Dynamics Investigation: High-Resolution Laser Ranging of the Earth’s Forests and Topography. Science of Remote Sensing, 1, Article ID: 100002. https://doi.org/10.1016/j.srs.2020.100002
[12]
FAO (Food and Agriculture Organization) (2020). Global Forest Resources Assessment 2020: Main Report. FAO.
[13]
Farr, T. G., Rosen, P. A., Caro, E., Crippen, R., Duren, R., Hensley, S. et al. (2007). The Shuttle Radar Topography Mission. Reviews of Geophysics, 45, RG2004. https://doi.org/10.1029/2005rg000183
[14]
Forkuor, G., Dimobe, K., Serme, I., & Tondoh, J. E. (2017). Landsat-8 vs. Sentinel-2: Examining the Added Value of Sentinel-2’s Red-Edge Bands to Land-Use and Land-Cover Mapping in Burkina Faso. GIScience & Remote Sensing, 55, 331-354. https://doi.org/10.1080/15481603.2017.1370169
[15]
Gibbs, H. K., Brown, S., Niles, J. O., & Foley, J. A. (2007). Monitoring and Estimating Tropical Forest Carbon Stocks: Making REDD a Reality. Environmental Research Letters, 2, Article ID: 045023. https://doi.org/10.1088/1748-9326/2/4/045023
[16]
Goetz, S. J. et al. (2015). Measurement and Monitoring Needs for REDD+. Current Opinion in Environmental Sustainability, 14, 11-22.
[17]
Gómez, C. (2017). Sentinel-2 for LULC Mapping and Monitoring: A Review. Remote Sensing, 9, Article 889.
[18]
Hojo, A., Matsui, T., & Nakamura, K. (2023). Sentinel-1 SAR for Forest Structure Characterization in Dense Canopies: A Case Study in Borneo. Ecological Informatics, 75, Article ID: 102045.
[19]
Johnson, L. M., Smith, R. K., & Brown, P. (2024). Multi-Sensor Fusion for Continuous Canopy Height Mapping: Integrating GEDI, Sentinel-1, and Sentinel-2. ISPRS Journal of Photogrammetry and Remote Sensing, 210, 45-60.
[20]
Kumar, S., Patel, N., & Jones, D. (2023). Reducing Uncertainty in Canopy Height Estimation through GEDI-Sentinel Fusion in Heterogeneous Landscapes. Forest Ecology and Management, 540, Article ID: 121234.
[21]
Kupidura, P. (2016). Comparison of Filters Dedicated to Speckle Suppression in SAR Images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 7, 269-276. https://doi.org/10.5194/isprs-archives-xli-b7-269-2016
[22]
Lang, N., Kalischek, N., Armston, J., Schindler, K., Dubayah, R., & Wegner, J. D. (2022). Global Canopy Height Regression and Uncertainty Estimation from GEDI LIDAR Waveforms with Deep Ensembles. Remote Sensing of Environment, 268, Article ID: 112760. https://doi.org/10.1016/j.rse.2021.112760
[23]
Li, W., Niu, Z., Shang, R., Qin, Y., Wang, L., & Chen, H. (2023). High-Resolution Mapping of Forest Canopy Height Using Machine Learning and Remote Sensing Data. ISPRS Journal of Photogrammetry and Remote Sensing, 185, 146-160.
[24]
Medeiros, F., Silva, C. A., & Dubayah, R. (2022). Tropical Forest Height Modeling Using Spaceborne LiDAR and Optical Data. Remote Sensing, 14, Article 1234.
[25]
Mwamba, C., Chidumayo, E., & Syampungani, S. (2023). Drivers of Deforestation in Zambia: Implications for REDD+ Policy. Land Use Policy, 125, Article ID: 106487.
[26]
Patel, R., Kumar, S., & Zhang, L. (2024). Enhancing Forest Structure Mapping through Multi-Sensor Data Fusion in Cloud-Prone Regions. International Journal of Applied Earth Observation and Geoinformation, 118, Article ID: 103245.
[27]
Potapov, P., Li, X., Hernandez-Serna, A., Tyukavina, A., Hansen, M. C., Kommareddy, A. et al. (2021). Mapping Global Forest Canopy Height through Integration of GEDI and Landsat Data. Remote Sensing of Environment, 253, Article ID: 112165. https://doi.org/10.1016/j.rse.2020.112165
[28]
Qi, Z., Yeh, A. G. O., Li, X., & Lin, Z. (2019). A Novel Algorithm for Land Use and Land Cover Classification Using RADARSAT-2 Polarimetric SAR Data. Remote Sensing of Environment, 221, 1-13.
[29]
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012). An Assessment of the Effectiveness of a Random Forest Classifier for Land-Cover Classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93-104. https://doi.org/10.1016/j.isprsjprs.2011.11.002
[30]
Ryan, C. M., Williams, M., & Grace, J. (2022). Gaps in National-Scale Forest Canopy Height Mapping: A Case Study of Zambia. Environmental Research Letters, 17, Article ID: 045012.
[31]
Schlund, M., Schepaschenko, D., & See, L. (2023). Challenges in Tropical Canopy Height Assessment Using Multi-Source Remote Sensing. Biogeosciences, 20, 345-360.
[32]
Shoko, C., & Mutanga, O. (2017). Evaluating the Performance of Sentinel-2 and Landsat-8 Data in Mapping Native and Invasive Trees in a Heterogeneous Landscape of Southern Africa. International Journal of Applied Earth Observation and Geoinformation, 66, 44-53.
[33]
Soudani, K., Delpierre, N., Berveiller, D., Hmimina, G., Vincent, G., Morfin, A. et al. (2021). Potential of C-Band Synthetic Aperture Radar Sentinel-1 Time-Series for the Monitoring of Phenological Cycles in a Deciduous Forest. International Journal of Applied Earth Observation and Geoinformation, 104, Article ID: 102505. https://doi.org/10.1016/j.jag.2021.102505
[34]
Tamiminia, H., Salehi, B., Mahdianpari, M., Quackenbush, L., Adeli, S., & Brisco, B. (2020). Google Earth Engine for Geo-Big Data Applications: A Meta-Analysis and Systematic Review. ISPRS Journal of Photogrammetry and Remote Sensing, 164, 152-170. https://doi.org/10.1016/j.isprsjprs.2020.04.001
[35]
Torres, R., Snoeij, P., Geudtner, D., Bibby, D., Davidson, M., Attema, E. et al. (2012). GMES Sentinel-1 mission. Remote Sensing of Environment, 120, 9-24. https://doi.org/10.1016/j.rse.2011.05.028
[36]
Vafaei, S., Amani, M., Mahdavi, S., & Brisco, B. (2018). Forest Height Estimation Using Sentinel-1 and Sentinel-2 Data Fusion and Machine Learning. Remote Sensing, 10, Article 964.
[37]
Wang, Y., Zhan, Q., & Ouyang, W. (2023). Deep Learning-Based Estimation of Forest Canopy Height Using Multi-Source Remote Sensing Data. Remote Sensing, 15, Article 789.
[38]
Williams, L. J., Cavender-Bares, J., Townsend, P. A., Couture, J. J., Wang, Z., Stefanski, A. et al. (2020). Remote Spectral Detection of Biodiversity Effects on Forest Biomass. Nature Ecology & Evolution, 5, 46-54. https://doi.org/10.1038/s41559-020-01329-4
[39]
Zhu, Z., & Woodcock, C. E. (2012). Object-Based Cloud and Cloud Shadow Detection in Landsat Imagery. Remote Sensing of Environment, 118, 83-94. https://doi.org/10.1016/j.rse.2011.10.028
