Abstract
NASA's Global Ecosystem Dynamics Investigation (GEDI) is collecting spaceborne full waveform lidar data with a primary science goal of producing accurate estimates of forest aboveground biomass density (AGBD). This paper presents the development of the models used to create GEDI's footprint-level (~25 m) AGBD (GEDI04_A) product, including a description of the datasets used and the procedure for final model selection. The data used to fit our models are from a compilation of globally distributed spatially and temporally coincident field and airborne lidar datasets, whereby we simulated GEDI-like waveforms from airborne lidar to build a calibration database. We used this database to expand the geographic extent of past waveform lidar studies, and divided the globe into four broad strata by Plant Functional Type (PFT) and six geographic regions. GEDI's waveform-to-biomass models take the form of parametric Ordinary Least Squares (OLS) models with simulated Relative Height (RH) metrics as predictor variables. From an exhaustive set of candidate models, we selected the best input predictor variables, and data transformations for each geographic stratum in the GEDI domain to produce a set of comprehensive predictive footprint-level models. We found that model selection frequently favored combinations of RH metrics at the 98th, 90th, 50th, and 10th height above ground-level percentiles (RH98, RH90, RH50, and RH10, respectively), but that inclusion of lower RH metrics (e.g. RH10) did not markedly improve model performance. Second, forced inclusion of RH98 in all models was important and did not degrade model performance, and the best performing models were parsimonious, typically having only 1-3 predictors. Third, stratification by geographic domain (PFT, geographic region) improved model performance in comparison to global models without stratification. Fourth, for the vast majority of strata, the best performing models were fit using square root transformation of field AGBD and/or height metrics. There was considerable variability in model performance across geographic strata, and areas with sparse training data and/or high AGBD values had the poorest performance. These models are used to produce global predictions of AGBD, but will be improved in the future as more and better training data become available.
| Original language | English |
|---|---|
| Article number | 112845 |
| Number of pages | 20 |
| Journal | Remote Sensing of Environment |
| Volume | 270 |
| Early online date | 7 Jan 2022 |
| DOIs | |
| Publication status | Published - 1 Mar 2022 |
Bibliographical note
Funding Information:Armston, Kellner, Hancock, and Dubayah were supported by NASA Contract #NNL 15AA03C to the University of Maryland for the development and execution of the GEDI mission. Duncanson and Minor were supported by a NASA GEDI Science Team Grant NNH20ZDA001N and a NASA Post Doctoral Program fellowship. Saarela was supported through NASA Carbon Monitoring System Grant 80HQTR18T0016 , and Healey and Patterson were funded by the GEDI mission through Interagency Agreement RPO201523 . We thank the NASA Terrestrial Ecology program for continued support of the GEDI mission, and the University of Maryland for providing independent financial support of the GEDI mission. We also thank NASA for contributing to several lidar data collections used in this study, including from the NASA Carbon Monitoring System (Grant number NNH13AW621 , to PI Cohen at the USFS Service). We also gratefully acknowledge the collection and provision of field and airborne data from a wide variety of other sources, including by the Sustainable Landscapes Brazil project supported by the Brazilian Agricultural Research Corporation (EMBRAPA) , the US Forest Service, the National Science Foundation ( DEB 0939907 ), Smithsonian Tropical Research Institute, USAID, and the US Department of State, among others. Additional data were acquired from the Terrestrial Ecosystem Research Network (TERN), an Australian Government NCRIS-enabled research infrastructure project, for provision of data used in this analysis, and from the National Ecological Observatory Network (NEON), a program sponsored by the National Science Foundation and operated under cooperative agreement by Battelle. We also thank the National Science and Engineering Research Council of Canada (NSERC), Discovery Grant Program (PI Sanchez-Azofeifa). We also thank the Spanish institutions and programs Instituto Geográfico Nacional, Organismo Autónomo de Parques Nacionales and Inventario Forestal Nacional for supporting this science with open data. The Council for Scientific and Industrial Research (CSIR) project "National Woody Vegetation Monitoring System for Ecosystem and Value-added Services" contributed to the collection of South African ALS and field data. We also thank the Sabie Sand Wildtuin, South African National Parks (SANPARKS), the Wits Rural Knowledge Hub and the Bushbuckridge Municipality in South Africa, for support in the South African field data collection. Additional Australian data were collected as part of the SMAPEx project funded by an Australian Research Council Discovery Project ( DP0984586 ). We thank Shell Gabon and the Smithsonian Conservation Biology Institute for funding the Rabi plot in Gabon, which is contribution No. 204 of the Gabon Biodiversity Program. We also acknowledge funding in French Guiana from CNES and "Investissement d'Avenir" grants managed by Agence Nationale de la Recherche (CEBA, ref. ANR-10-LABX-25-01). We thank the Project LIFE+ ForBioSensing PL “Comprehensive monitoring of stand dynamics in Białowieża Forest supported with remote sensing techniques" co-funded by Life Plus (contract number LIFE13 ENV/PL/000048) and Poland’s National Fund for Environmental Protection and Water Management (contract number 485/2014/WN10/OP-NM-LF/D) for funding the collection of the Polish data, and Rafał Sadkowski for helping with data preparation from the ForBioSensing project. We also thank The Silva Tarouca Research Institute (Czech Republic) for collecting and providing field reference data under an INTER-ACTION project ( LTAUSA18200 ). We also thank the former NERC Airborne Research Facility for their support with airborne data collection, and funding for airborne Lidar data provided by the Australian Department of Agriculture, Fisheries, and Forestry (DAFF). We also thank the Norwegian Agency for Development Cooperation (Norad), although the views expressed in this publication do not necessarily reflect the views of Norad. We also acknowledge DfID and UK Natural Environment Research Council ( NE/P004806/1 ) for collection of field data. The Tanzanian field work for this study was carried out as part of the project “Enhancing the measuring, reporting and verification (MRV) of forests in Tanzania through the application of advanced remote sensing techniques”, funded by the Royal Norwegian Embassy in Tanzania as part of the Norwegian International Climate and Forest Initiative. Finally, data from RAINFOR plots were supported by the Moore Foundation , and SERNANP (Peru) granted research permissions.
Funding Information:
Armston, Kellner, Hancock, and Dubayah were supported by NASA Contract #NNL 15AA03C to the University of Maryland for the development and execution of the GEDI mission. Duncanson and Minor were supported by a NASA GEDI Science Team Grant NNH20ZDA001N and a NASA Post Doctoral Program fellowship. Saarela was supported through NASA Carbon Monitoring System Grant 80HQTR18T0016, and Healey and Patterson were funded by the GEDI mission through Interagency Agreement RPO201523. We thank the NASA Terrestrial Ecology program for continued support of the GEDI mission, and the University of Maryland for providing independent financial support of the GEDI mission. We also thank NASA for contributing to several lidar data collections used in this study, including from the NASA Carbon Monitoring System (Grant number NNH13AW621, to PI Cohen at the USFS Service). We also gratefully acknowledge the collection and provision of field and airborne data from a wide variety of other sources, including by the Sustainable Landscapes Brazil project supported by the Brazilian Agricultural Research Corporation (EMBRAPA), the US Forest Service, the National Science Foundation (DEB 0939907), Smithsonian Tropical Research Institute, USAID, and the US Department of State, among others. Additional data were acquired from the Terrestrial Ecosystem Research Network (TERN), an Australian Government NCRIS-enabled research infrastructure project, for provision of data used in this analysis, and from the National Ecological Observatory Network (NEON), a program sponsored by the National Science Foundation and operated under cooperative agreement by Battelle. We also thank the National Science and Engineering Research Council of Canada (NSERC), Discovery Grant Program (PI Sanchez-Azofeifa). We also thank the Spanish institutions and programs Instituto Geogr?fico Nacional, Organismo Aut?nomo de Parques Nacionales and Inventario Forestal Nacional for supporting this science with open data. The Council for Scientific and Industrial Research (CSIR) project ?National Woody Vegetation Monitoring System for Ecosystem and Value-added Services? contributed to the collection of South African ALS and field data. We also thank the Sabie Sand Wildtuin, South African National Parks (SANPARKS), the Wits Rural Knowledge Hub and the Bushbuckridge Municipality in South Africa, for support in the South African field data collection. Additional Australian data were collected as part of the SMAPEx project funded by an Australian Research Council Discovery Project (DP0984586). We thank Shell Gabon and the Smithsonian Conservation Biology Institute for funding the Rabi plot in Gabon, which is contribution No. 204 of the Gabon Biodiversity Program. We also acknowledge funding in French Guiana from CNES and ?Investissement d'Avenir? grants managed by Agence Nationale de la Recherche (CEBA, ref. ANR-10-LABX-25-01). We thank the Project LIFE+ ForBioSensing PL ?Comprehensive monitoring of stand dynamics in Bia?owie?a Forest supported with remote sensing techniques" co-funded by Life Plus (contract number LIFE13 ENV/PL/000048) and Poland's National Fund for Environmental Protection and Water Management (contract number 485/2014/WN10/OP-NM-LF/D) for funding the collection of the Polish data, and Rafa? Sadkowski for helping with data preparation from the ForBioSensing project. We also thank The Silva Tarouca Research Institute (Czech Republic) for collecting and providing field reference data under an INTER-ACTION project (LTAUSA18200). We also thank the former NERC Airborne Research Facility for their support with airborne data collection, and funding for airborne Lidar data provided by the Australian Department of Agriculture, Fisheries, and Forestry (DAFF). We also thank the Norwegian Agency for Development Cooperation (Norad), although the views expressed in this publication do not necessarily reflect the views of Norad. We also acknowledge DfID and UK Natural Environment Research Council (NE/P004806/1) for collection of field data. The Tanzanian field work for this study was carried out as part of the project ?Enhancing the measuring, reporting and verification (MRV) of forests in Tanzania through the application of advanced remote sensing techniques?, funded by the Royal Norwegian Embassy in Tanzania as part of the Norwegian International Climate and Forest Initiative. Finally, data from RAINFOR plots were supported by the Moore Foundation, and SERNANP (Peru) granted research permissions.
Data Availability Statement
Supplementary data to this article can be found online at https://doi.org/10.1016/j.rse.2021.112845.Keywords
- Aboveground biomass
- Forest
- GEDI
- LiDAR
- Modeling
- Waveform