Improving pantropical forest carbon maps with airborne LiDAR sampling

Transcription

1 Mini Focus: Sustainable Landscapes in a World of Change: Tropical Forests, Land use and Implementation of REDD+ Improving pantropical forest carbon maps with airborne LiDAR sampling Carbon Management (213) 4(6), Alessandro Baccini*1 & Gregory P Asner2 Background: Countries interested in monitoring and quantifying the carbon stock of their tropical forests need cost-effective methodologies to map aboveground carbon density (ACD) at regional and national project levels, and with measurable precision and accuracy. This study reports on improvements made possible by the use of airborne high resolution LiDAR samples to regionally fine-tune freely available moderate resolution remote sensing data, and generate maps of ACD with greater detail and improved accuracy than existing pantropical data sets. Results: Regions in the Peruvian and Colombian Amazon indicate that, although existing pantropical data sets of ACD explain approximately 7% of the variance in ACD relative to high resolution LiDAR estimates, by fine calibration with airborne LiDAR samples, it is possible to reduce the relative root mean squared error from 25.2 and 31.4 MgC ha-1, to 15.7 and 17.6 MgC ha-1, respectively, in Colombia and Peru. Conclusion: Airborne LiDAR data can successfully be used for fine-tuning freely available moderate resolution remote sensing image data and significantly improving existing aboveground carbon density maps, to better meet the requirements for national and subnationl carbon density mapping. Global tropical forests store approximately 228 billion metric tons of carbon in aboveground live tissues, and are a source of CO2 through deforestation and degradation that currently equals or exceeds total transportation sector emissions worldwide [1]. As a result, there has been a rapid increase in efforts to conserve tropical forest carbon stocks as a contribution to stabilizing atmospheric CO2 concentrations [2]. One of the key initiatives supporting tropical forest carbon conservation is known as REDD+ [3]. To make REDD+ a possibility as a natural resource and international climate mitigation policy tool, the amount of carbon (approximately 48% of biomass by dry weight [4]) stored in forests must be repeatedly mapped and spatially monitored. In response to this need, remote sensing has taken on a role, along with new and pre-existing field plot inventory networks, in developing maps of aboveground carbon density (ACD); in units of MgC ha-1. At the pantropical scale, two particular approaches using satellite data have been put forward to enhance our understanding of ACD distributions [5,6]. Both approaches utilize moderate resolution, global satellite information with field data to estimate tropical forest carbon stocks. The gridded spatial resolution of these approaches ranges from 5 to 1 m, and the uncertainties reported on any given grid cell range from 1 to more than 1%, in other words, the uncertainty of the globally mapped ACD values within any given grid cell can be as large as the value reported. While global-scale carbon mapping approaches have advanced the readiness of the international community for REDD+ and similar partnership programs, much of the activity to reduce tropical carbon losses is advancing at national and subnational scales [7]. To support these activities, other work combining higher resolution satellite, airborne LiDAR and field plot data has focused on developing maps at grid cell resolutions of 3 1 m [8 1]. At 1-ha mapping resolution, uncertainties in mapped carbon stocks based on the LiDAR approach The Woods Hole Research Center, 149 Woods Hole Road, Falmouth, MA 254, USA Department of Global Ecology, Carnegie Institution for Science, 26 Panama Street, Stanford, CA 9435, USA *Author for correspondence: abaccini@whrc.org /CMT Future Science Ltd ISSN

2 Baccini & Asner are 1 2% [11,12], which is the critical level of accuracy supporting Tropical forests: Tropical forests contain jurisdictional carbon accounting vast carbon stocks, and harbor the majority of terrestrial plant and animals [13]. A current drawback with this species on Earth. approach, however, rests on the fact LiDAR: A technology utilizing emitted that fairly high spatial resolution laser pulses to quantify the distance satellite data such as from Landbetween objects, including the height sat sensors is required, and these and vertical structure of vegetation. data are not available for all tropical Carnegie Airborne Observatory: The regions, owing to persistent cloud world s first fully integrated laser and imaging spectrometer system used for cover [14]. large-scale biospheric exploration and On the one hand, the global carresearch. bon mapping approaches provide Moderate Resolution Imaging near wall-to-wall coverage of the Spectroradiometer: The NASA tropics, but their accuracy is necmoderate Resolution Imaging essarily lowered by their limited Spectroradiometer was launched on the TERRA and AQUA satellites in 2 sensitivity to vertical forest strucand 24, respectively. Both continue to tures. While regional- to nationaloperate today, providing global scale carbon mapping approaches coverage of land, oceans, ice and our provide high-resolution results atmosphere. with higher per-hectare accuracies, they are difficult to implement worldwide. We therefore seek to combine the best aspects of these two approaches by leveraging the global satellite coverage from the former method with the high precision airborne LiDAR-based ACD estimates of the latter approach. The benefit would be bi-directional: global maps can be improved on a region by region basis to better quantify global carbon stocks, and regional maps can be derived from the global data sets but with higher accuracy than that achieved by the original global mapping approach. Here, we report on the first study to fine-tune a pantropical carbon map to specific subnational regions, using freely available global satellite data with regionally sampled areas of forest structure from airborne LiDAR. We carried out the study in two regions of the Amazon basin, one in Colombia and one in Peru, to estimate the improved per-hectare accuracies achieved using the downscaling methodology presented. Key terms Methods & materials Study area The two study regions were defined in previous reports by Asner et al. in the southern Peruvian Amazon [8] and in the Colombian Amazon [9]. The Peru study area encompasses 4.3 Mha of forest in the department of Madre de Dios. The region is lowland humid tropical forest with a rapidly expanded land-use change process involving cattle ranching, agriculture and gold mining. The Colombia study area covers 16.5 Mha of lowland Amazonian forest with relatively little landuse change; however, the region contains widely varying carbon stocks based on floodplain inundation and 592 Carbon Management (213) 4(6) other hydrological processes. The Colombia study covers the government s REDD+ demonstration and research region. Together, the Peru and Colombia regions harbor a wide range of carbon stock levels matching the range found throughout the pantropics. Airborne LiDAR In both the Peru and Colombia study regions, LiDAR data were collected using the Carnegie Airborne Observatory (CAO) Alpha Sensor Package [1]. The CAO LiDAR was capable of both full waveform and discrete return measurements, but for this study only the data from the discrete return data were used, as described in detail by Asner et al. [8,9]. In Peru, the data were collected in 31 sampling blocks randomly distributed throughout the region. These sampling blocks ranged in size from approximately 8 to 22, ha per block, and total LiDAR coverage for Peru was 34,758 ha within the 4.3 Mha mapping region. In Colombia, the data were acquired in 39 sampling blocks, also randomly distributed throughout the region. The blocks ranged in size from approximately 37 to 28, ha, for a total LiDAR coverage of 465,622 ha. Details of the LiDAR instrument settings, data collection protocol and processing steps were previously detailed by Asner et al. [8,9]. Briefly, the data were collected from 2 m above ground-level in sampling blocks with 5% overlap among adjacent flightlines. This resulted in a laser spot density of approximately two laser shots m-2. The data were processed to laser point clouds, and then classified to derive digital terrain models and digital surface models. Mean canopy profile height (MCH) a commonly used LiDAR metric measuring the distance from ground (digital terrain models) to the approximate centroid of the tree crowns [15] was mapped at 1.1 m spatial resolution. The MCH maps were calibrated to field-estimated ACD values from plot inventory networks described by Asner et al. [16], and 3 m resolution ACD maps were derived for each LiDAR sampling block. Heavily documented previously [8,9,16], we provide a summary here of the method. First,.28 ha circular plots were established in the study regions, placed randomly in areas to capture the full range of aboveground carbon stock levels, and among a wide range of vegetation types and disturbance conditions. The median value of the MCH in each plot was computed from the CAO LiDAR data. Regressions were developed between plot-level median MCH and estimated ACD. The regression was then applied to LiDAR-based maps of median MCH at 3 m resolution. Moderate Resolution Imaging Spectroradiometer data We used canopy reflectance data produced from the Moderate Resolution Imaging Spectroradiometer (MODIS)

3 Improving pantropical forest carbon maps with airborne LiDAR sampling Band 2 (near-infrared) 78 9 nm; Band 4 (green) nm; Band 5 (near-infrared) nm; Band 6 (mid infrared) nm; Band 7 (mid infrared) nm; Variance in band 7; LST and variance in LST; Enhanced Vegetation Index (EVI2); Normalized Difference Infrared Index (NDII2); Elevation from Shuttle Radar Topography Mission. The remaining steps for this processing were detailed by Baccini et al. [6]. Analysis Approach airborne LiDAR assisted mapping at large scales Following the general approach of Baccini et al. [6], field protocols were developed to collect biometric data for tree stems located within Geoscience Laser Altimeter System (GLAS) footprints over broad geographical scales, and allometric relationships were used to estimate aboveground biomass density within those footprints [21]. Using these data, linear models were estimated to predict biomass within GLAS footprints based on GLAS LiDAR metrics. This relationship was then used to predict biomass at several thousand high-quality GLAS footprints distributed across the tropics. By intersecting the location of this sample with 5 m MODIS pixels, we identified locations where multiple GLAS footprints were available within MODIS pixels and used these to calibrate a machine learning model (RandomForest; [2]) that predicts biomass based on MODIS data. This model was then used to generate a wall-to-wall map of pantropical aboveground biomass from MODIS. In this article, we replaced the GLAS data with the CAO high resolution LiDAR ACD estimates to calibrate a RandomForest algorithm [2,22]. RandomForest is a data mining algorithm that performs recursive partitioning of the data and uses different bootstraps 17 Moderate Resolution Imaging Spectroradiometer aboveground carbon density estimates (MgC ha-1) sensors onboard the NASA TERRA and AQUA satellites. The 5 m spatial resolution MODIS reflectance measurements were screened for cloud cover, normalized for viewing and solar conditions and composited to an 8-day time interval [15,17]. This MODIS product assures the largest number of observations of the highest possible data quality. We further screened the reflectance data set to include only those solutions that make use of at least three observations within two 8-day time periods, taking the approach of using only the highest quality data results in an image product with a substantial number of spatial data gaps. To overcome this limitation, we used MODIS Nadir Bidirectional Reflectance Distribution FunctionAdjusted Reflectance MCD43A4 and MCD43A1 data spanning 2 years (27 28), including a total of 92 sensor observations [18]. Although, for the regions of this study, a 1-year window was sufficient to produce a cloudfree mosaic, for consistency with results reported previously [6], we used the 2-year time window. In addition to the pixel mean, we created a layer for the SD, Enhanced Vegetation Index and Normalized Difference Infrared Index vegetation indices [19]. For the same observations, the MODIS Land Surface Temperature (LST) was extracted and the average values were computed. NASA Shuttle Radar Topography Mission data at a resolution of 9 m were aggregated to the resolution of 5 m MODIS reflectance. The following variables were used to develop a RandomForest model that generates estimates of ACD in each MODIS pixel [2] : Band 1 (red) nm; Carnegie Airborne Observatory aboveground carbon density estimates (MgC ha-1) Figure 1. Pixel level (~5 m) averaged Carnegie Airborne Observatory aboveground carbon density estimates versus Moderate Resolution Imaging Spectroradiometer-based estimates, when the Moderate Resolution Imaging Spectroradiometer model is calibrated with 3% of Carnegie Airborne Observatory data (15, ha) and validated with a 1% subset not used in the model calibration process. The model explains 86% of the variance in biomass density with a root mean squared error = 16.5 MgC ha

4 Baccini & Asner samples to estimate a large number of tree-based models, resulting in more accurate and less sensitive to noise data sets. This approach generates a continuous range of ACD estimates for each grid cell of the data set described in the previous section. To do this, we first projected the CAO ACD in the MODIS Sinusoidal original projection and computed the average ACD within each of the 5 m grid cells overlapping with the CAO transects. The cells overlapping for at least 95% of the area were used as a training data set in the RandomForest algorithm. We assessed the results by cross-validation ana lysis, in which we set aside 1% of the data to test the RandomForest estimates versus the original CAO ACD estimates. Comparison of moderate & high resolution ACD We assessed potential improvements achieved in the new, regionally tuned carbon stocks estimates based on CAO ACD data (CACD) relative to the previously available moderate resolution data set, referred to here as the pantropical ACD (PACD). For each 5 m PACD cell overlapping with CACD sampling blocks (not including edges of dual coverage), we computed Root mean squared error based on 1% set aside data (biomass; Mg ha-1) 22 the average CACD carbon stocks. We also computed the average CACD for each flight block and compared it to the average carbon stocks derived from the 5 m PACD. We used the root mean squared error (RMSE) between CACD and PACD to quantify statistical goodness of fit. Results New approach The new, regionally tuned models calibrated using the CACD successfully estimated ACD with a RMSE of 15.7 and 17.6 Mg ha-1 in Colombia and Peru, respectively, and explained approximately 86% of the variance in ACD (Figure 1). The advantage of using CACD was threefold: The high sensitivity of the CAO instrument to forest structure provided a more accurate link to carbon stocks than the large footprint spaceborne GLAS sensor did [23] ; The 3 m spatial resolution of CACD provided wall- to-wall coverage of specific 5 m MODIS pixels, overcoming the limitations of sampling sensors such as GLAS that provide forest structural measurements for only few limited areas within MODIS pixels; Peru Colombia Carnegie Airborne Observatory area (1 ha) Figure 2. Decrease in root mean squared error as function of Carnegie Airborne Observatory area coverage used in model calibration. 594 Carbon Management (213) 4(6) It provided a much larger training data set for the calibration of the model to the specific study region of interest. There were 18,176 CAO training samples for Colombia and 12,137 for Peru. These factors led to an overall improved calibration of the optical satellite data to forest structure, particularly in pixels with a mosaic of land cover types [24]. The results indicated that a relatively limited coverage of CAOderived carbon stock estimates, for example 3% (or ~15, ha) of randomly selected CAO data, significantly improved regional carbon stocks estimates by reducing the overall RMSE (Figures 1 & 2). Spatially, the new regionally tuned model provided a more detailed characterization of the ACD distribution over the study region, revealing clearer spatial patterns in carbon stocks than that achieved using the original pantropical approach based on spaceborne GLAS LiDAR data

5 Improving pantropical forest carbon maps with airborne LiDAR sampling (Figure 3). The inferior performance of the GLAS-based approach is the result of sensor characteristics such as the large and variable footprint (~6 75 m), the limited number of GLAS shots in the study region (after filtering [6]) and the lack of field data available in the calibration process. In contrast, the airborne LiDAR data have a small and constant foot print of 1.1 m in this case, are derived from high-power laser emissions that yield detailed vertical structural information on the canopy and are well calibrated to field plots. Comparison of PACD & CACD maps A comparison of the original PACD and CACD maps revealed a general agreement between the two data sets, but important region-specific discrepancies were noted as well. Foremost, the PACD values tended to be higher than the CACD results, indicating a bias of approximately 5 MgC ha-1 in the PACD. We found A D that this discrepancy resulted from different allometric equations used to develop plot estimates of ACD in previous studies by Baccini et al. [6] and Asner et al. [8,9]. The CAO-based studies had used allometric equations incorporating both stem diameters and canopy height, while the PACD study used allometry based only on stem diameter [21]. These findings are consistent with Feldpausch et al., reporting that in the Amazon basin, the use of the tree height in the allometric equations results in lower estimates than allometry with only stem diameter [25]. A simple offset correction addressed this issue, but we emphasis here the importance of allometric models and the potential errors embedded in the construction of those models [26]. Overall, the pixel-level comparison of original PACD and CACD data indicated general agreement, with PACD explaining approximately 7% of the variance relative to CACD (Figure 4). We observed High: 156 MgC ha H igh : LoLow: w : MgC ha-1 B E C F HHigh: igh : MgC ha Low:: -18 Low MgC ha-1 Figure 3. Comparison of existing and new aboveground carbon density maps. Subset maps of ACD derived from pantropical mapping of (A) Colombia and (D) Peru, then (B) and (E) remapped after fine-tuning with CAO LiDAR data. The difference between the two maps is shown in panels (C) and (F). A subset of the CAO samples are shown in black lines. ACD: Aboveground carbon density; CAO: Carnegie Airborne Observatory

6 Geoscience Laser Altimeter System-Moderate Resolution Imaging Spectroradiometer aboveground carbon density (MgC ha-1) Baccini & Asner the calibration of the new model was based on allometry consistent with CACD, the previously observed allometry-based bias was now solved. Discussion By leveraging the capabilities of high-resolution airborne LiDAR ACD estimates, which are very accurate but limited in spatial coverage, with coarser resolution satellite observations, we are able to model 96 carbon stocks over large areas with demonstrable precision. Here, we show how freely available, global data such as from MODIS, when calibrated with CAO-estimated 48 ACD sampling blocks, can improve carbon stock maps in jurisdictional to national scales such as in Peru and Root mean squared error = 36.3 MgC ha-1 Colombia. Biomass estimates over large areas R2 =.7 are commonly developed by com bining ACD estimates from specific locations (forest inventory plots) with spatial layers such as land cover, Carnegie Airborne Observatory aboveground carbon density (MgC ha-1) vegetation maps and derivatives from these inputs. In essence, ACD data are used to assign a mean value Figure 4. Comparison of Carnegie Airborne Observatory and pantropical data sets of to the respective land cover class. aboveground carbon density aggregated to 5 m resolution. Each point in the figure An extension of this approach has shows the average CAO ACD within the Moderate Resolution Imaging Spectroradiometer pixel -1 been used to map ACD in a series versus the ACD derived from the pantropical data set after removing 5 MgC ha bias due to of tropical ecosystems by stratifying allometry. the landscape and using CAO estiacd: Aboveground carbon density; CAO: Carnegie Airborne Observatory. mates to assign the biomass densities that the ACD distribution in the Peru and Colombia to the strata [8,9]. Often, these methods assume that land study areas had a similar range, but with a different fre- cover classes or strata capture the geospatial variability quency distribution with Colombia (Figure 5). Colom- in ACD. In some cases it is a fair assumption, but in bia showed a binomial distribution, whereas Peru had other cases the mean ACD within the class is only partly the majority of carbon stocks values between 8 and representative of the ACD of the entire forest. 12 Mg ha-1. The lower variability of carbon stocks in The spatial distribution of both forest loss and carthe Peru data set likely approaches the limit in how bon density is important to calculate carbon emissions well the PACD data set captures small local variability from land cover land-use change. Mean carbon densiin ACD, thereby explaining the observed larger relative ties and mean rates of forest loss over large regions RMSE (Table 1). At the regional level, the compari- may yield accurate estimates of carbon emissions if son of PACD versus CACD aggregated at the flight the disturbances are distributed randomly. But if dissampling block (thousands of ha) indicated very good turbances, particularly from land-use and land-cover agreement, with a relative RMSE of 14.8 Mg ha-1 and change, affect forests that are systematically differr 2 =.89 (Figure 6). ent from the mean carbon density, that difference The comparison of the new regionally tuned PACD will bias emissions estimates. One way to address this and CACD maps indicated a more consistent agreement potential bias is to co-locate changes in areas with carbetween the two maps, with the new PACD explain- bon densities matching the spatial resolution of dising up to 86% of the variance relative to CACD turbance. The study presented here is a step forward (Figure 1). Furthermore, because the ACD data used in to better characterize the ACD of the vegetation that Carbon Management (213) 4(6)

7 Improving pantropical forest carbon maps with airborne LiDAR sampling Density may be disturbed. Although spatial disturbances at annual intervals are Peru in the order of 1 ha in size, here we.15 Colombia presented a new approach to better characterize ACD at the spatial resolution of 25 ha (5 5 m pixel size) for a specific region. While improved ACD estimates at 25 ha.1 are coarser than most disturbances, they are an improvement over using average values in large geographic strata, because ACD is constrained to an area of 25 ha instead of thou.5 sands of hectares. In the case of MODIS, we used the 5 m data, instead of the 25 m MODIS product, because the former contains the full seven-band land reflectance, which is needed for the model. ing. The 25 m MODIS product provides only two spectral bands, Aboveground carbon density (MgC ha-1) which has proven inadequate for use in monitoring carbon stocks [6]. Previous work used GLAS data Figure 5. Carnegie Airborne Observatory aboveground carbon density distribution to calibrate models of ACD as a (aggregated to 5 m) for Peru and Colombia. function of MODIS ref lectance and other variables [6]. One of the which, in the case of moderate resolution, is in the key aspects of the calibration of the GLAS data consisted in the collection of co-located field order of several hectares. Because of the high resolution measurements. Because GLAS stopped functioning in of CACD data and the wall-to-wall coverage of the the early part of 29, it is becoming challenging to MODIS pixel, CACD provided the solution to proper properly calibrate GLAS measurements with field data scaling of the ACD. Use of CACD data (field or LiDAR measurements) in for future use, unless accurate change ana lysis is performed to minimize temporal decorrelation. The use of the calibration of the original PACD provides a unique airborne ACD estimates for the calibration and updat- opportunity for an independent validation at the pixel ing of models based on globally available data represent level of the PACD. It also offers insight into the capabila solution to bridge the data gap created by a lack of ity and limitations of moderate resolution data to capture the local spatial variability of ACD in tropical regions. spaceborne LiDAR technology. The 3 m spatial resolution CACD data proved to be particularly important in this study as a way to scale-up Table 1. The average and maximum from the pantropical and Carnegie field data estimates to the coarser resolution MODIS Airborne Observatory aboveground carbon density data sets, and the data, and to fine-tune the RandomForest algorithm relative root mean squared error resulting from the comparison of the to better capture spatial variability in ACD specific original pantropical aboveground carbon density with Carnegie to our two study regions. It is important to note that Airborne Observatory aboveground carbon density (pantropical the RandomForest models for the PACD and CACD aboveground carbon density column). were based on the same predictor variables (see seccolombia Peru tion titled Moderate Resolution Imaging SpectroraPACD CAO PACD CAO diometer data ) but different ACD data, confirming the advantage of using CACD for the training of the Average ACD model. The use of field data or sampling measurements Maximum ACD always poses the challenge of how to solve the mismatch RMSE CAO RMSE indicates the RMSE in ACD estimates when the RandomForest model is calibrated in resolution between existing small forest inventory The using CAO ACD data set (CAO column). plots, often in the order of.1 ha and area covered by ACD: Aboveground carbon density; CAO: Carnegie Airborne Observatory; PACD: Pantropical the field-of-view of remote sensing instrumentation, aboveground carbon density; RMSE: Root mean squared error

8 Geoscience Laser Altimeter System-Moderate Resolution Imaging Spectroradiometer aboveground carbon density (MgC ha-1) Baccini & Asner 145 Root mean squared error = 14.8 MgC ha-1 R2 = However, costs were much higher in similar projects undertaken in more expensive regions, such as the US Hawaiian Islands [7]. Nonetheless, these cost ranges are far less than one would commit to on the ground doing field plots, which have averaged approximately $25 5 ha-1 in the same studies mentioned above. The value of well-calibrated airborne LiDAR for forest carbon estimation becomes obvious in these cases. Conclusion Developing countries are seeking 45 cost-effective approaches that best meet the requirements of mapping projects, from local to national scales [28]. Here, we combined the 2 best elements of moderate resolution spaceborne data and high-resolution airborne LiDAR ACD estimates to improve ACD maps originally calibrated at the continental scale. The use of CAO airborne-derived ACD Carnegie Airborne Observatory aboveground carbon density (MgC ha-1) to calibrate moderate resolution satellite data improved the mapping of Figure 6. The Carnegie Airborne Observatory and pantropical aboveground carbon density carbon stocks in two study regions, data sets aggregated at the scale of each airborne flight sample. Here we consider each flight reducing the relative RMSE from block as one observation and 31.4 MgC ha-1, to 15.7 and 17.6 MgC ha-1, in Colombia and However, it is important to note that the assessment of Peru, respectively. The comparison of the original PACD the new regionally calibrated PACD versus the CACD, with CAO high-resolution carbon estimates indicates that although based on a statistically valid cross-validation PACD provides a freely available (from national ACD approach with 1% of the data, is not an absolute inde- data) data set to be used over large areas and explains pendent validation. While a better characterization of approximately 7% of the variance relative to the CACD. the uncertainties is needed, the ability to significantly When PACD is finely tuned with regional CAO ACD reduce the uncertainty in ACD mapping by using CACD data, the variance explained increases to 86%. These suggests that, with proper methodological refinements, it results suggest that existing pantropical data sets can be could be possible to report on ACD trajectories over time, significantly improved by fine calibration with airborne at least for regions affected by large area deforestation LiDAR data to better meet the requirements for national processes. and subnational mapping of forest carbon stocks. Finally, the cost benefit trade-off in using airborne LiDAR data needs to be considered when working to fine- Future perspective tune course-resolution carbon maps to national and sub- By combining global-scale satellite mapping capabilities national scales. The published per-hectare costs of using with regional-scale airborne LiDAR sampling methods, LiDAR vary widely, and are based on an economy-of-scale subnational to national level carbon maps can be derived type of effect in which it is much more cost effective to do with demonstrably high fidelity. With global satellite larger projects with LiDAR samples data now freely available for the world and with airborne Key term distributed throughout a region [27]. LiDAR technology becoming much more commonplace, Forest degradation: Any activity that In the Peru and Colombian Amazon we believe these integrated approaches will pave the way reduces the carbon storage of a forest cases, the cost for the LiDAR data for a future of improved tropical forest monitoring. As over time. Examples include unsustainable logging and collection, processing and ana lysis a result, carbon emissions from tropical forests will be human-induced fire. ranged from US$.5 to.2 ha-1. more accurately estimated and, thus, the conservation and 598 Carbon Management (213) 4(6)

9 Improving pantropical forest carbon maps with airborne LiDAR sampling resource policy sectors will be empowered to reduce the negative effects of deforestation and forest degradation on our climate system. We also think these approaches will evolve quickly enough to greatly reduce reliance on highdensity, time-consuming and costly forest plot inventory networks. Financial & competing interests disclosure This ana lysis was made possible by the Gordon and Betty Moore Foundation (CA, USA). The Carnegie Airborne Observatory (CA, USA) is made possible by the Avatar Alliance Foundation (CA, USA), Grantham Foundation for the Protection of the Environment (MA, USA), John D and Catherine T MacArthur Foundation (IL, USA), Gordon and Betty Moore Foundation, WM Keck Foundation (CA, USA), Margaret A Cargill Foundation (MN, USA), MA Nyburg Baker and GL Baker Jr, and WR Hearst III. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript. Executive summary Background Remote sensing and plot inventory networks serve as a tool for monitoring and mapping tropical forest carbon density over time. One of the key initiatives supporting tropical forest carbon conservation is known as REDD. To make REDD+ a possibility, the amount of carbon stored in forests must be repeatedly mapped and spatially monitored. The use of airborne high-resolution LiDAR measurements improves existing pantropical carbon density models. In response to this need, remote sensing has taken on a role in developing maps of aboveground carbon density throughout the tropics, but these approaches are necessarily low in spatial and biophysical detail. Results Here we develop an approach to fine-tune pantropical carbon maps to specific subnational regions using airborne LiDAR. Globally available satellite data calibrated with airborne LiDAR data successfully mapped aboveground carbon density in Colombia and Peru, sharply increasing accuracy over global approaches. Combining NASA Moderate Resolution Imaging Spectroradiometer satellite data with Carnegie Airborne Observatory LiDAR measurements, two large regions of Colombia and Peru were mapped with a reduced root mean squared error of 15.7 and 17.6 MgC ha-1, respectively. Conclusion Existing pantropical data sets can be significantly improved by fine calibration with airborne LiDAR data to better meet the requirements for national and subnational mapping of forest carbon stocks. References Papers of special note have been highlighted as: of interest of considerable interest 5 n n n n 4 IPCC. Climate Change 27: Impacts, Adaptation and Vulnerability. Cambridge University Press, NY, USA (27). UNFCCC. Methodological Guidance for Activities Relating to Reducing Emissions From Deforestation and Forest Degradation and the Role of Conservation, Sustainable Management of Forests and Enhancement of Forest Carbon Stocks in Developing Countries. UNFCCC, Copenhagen, Denmark (29). 6 n n Saatchi SS, Harris NL, Brown S et al. Benchmark map of forest carbon stocks in tropical regions across three continents. Proc. Natl Acad. Sci. USA 18(24), (211). 9 Asner GP, Clark JK, Mascaro J et al. High-resolution mapping of forest carbon stocks in the Colombian Amazon. Biogeosciences 9, (212). 1 One of the key inputs into the current article, focusing on pantropical carbon mapping using moderate resolution satellite data. Asner GP, Knapp DE, Kennedy-Bowdoin T et al. Carnegie airborne observatory: in-flight fusion of hyperspectral imaging and waveform light detection and ranging for threedimensional studies of ecosystems. J. Appl. Remote Sens. 1(1), (27). 11 Mascaro J, Detto M, Asner GP, Muller-Landau HC. Evaluating uncertainty in mapping forest carbon with airborne LiDAR. Remote Sens. Environ. 115(12), (211). 12 Zolkos SG, Goetz SJ, Dubayah R. A metaana lysis of terrestrial aboveground biomass estimation using LiDAR remote sensing. Remote Sens. Environ. 128, (213). Baccini A, Goetz SJ, Walker WS et al. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nat. Clim. Chang. 2, (212). Angelsen A. Moving Ahead With REDD: Issues, Options and Implications. Center for International Forestry Research, Bogor, Indonesia (28). 7 Excellent basis calling for the need for forest carbon stock and emissions monitoring for REDD+. Asner GP, Hughes RF, Mascaro J et al. High-resolution carbon mapping on the million-hectare island of Hawaii. Front. Ecol. Environ. 9, (211). 8 Asner GP, Powell GVN, Mascaro J et al. High-resolution forest carbon stocks and emissions in the Amazon. Proc. Natl Acad. Sci. USA 17(38), (21). Martin AR, Thomas SC. A reassessment of carbon content in tropical trees. PLoS ONE 6(8), e23533 (211). The other key input to the current article, focusing on regional high resolution carbon estimation using airborne LiDAR. n n n Synthesizes LiDAR work, showing that uncertainties in LiDAR-based carbon 599

10 Baccini & Asner estimates can be reduced to approximately 1% at 1-ha spatial resolution. 13 Odigha O. GCF 21 Activities Report. Green Climate Fund, Cross River State, Nigeria (21). 14 De Sy V, Herold M, Achard F et al. Synergies of multiple remote sensing data sources for REDD+ monitoring. Curr. Opin. Environ. Sustain. 4(6), (212). 15 Lefsky MA, Cohen WB, Harding DJ, Parker GG, Acker SA, Gower ST. LiDAR remote sensing of above-ground biomass in three biomes. Glob. Ecol. Biogeogr. 11(5), (22) Ju J, Roy DP, Shuai Y, Schaaf C. Development of an approach for generation of temporally complete daily nadir MODIS reflectance time series. Remote Sens. Environ. 114(1), 1 2 ((21). 24 Baccini A, Friedl MA, Woodcock CE, Zhu Z. Scaling field data to calibrate and validate moderate spatial resolution remote sensing models. Photogramm. Eng. Remote Sens. 73(8), (27). 19 Hunt Jr, Raymond E, Barrett R. Detection of changes in leaf water content using near- and middle-infrared reflectances. Remote Sens. Environ. 3(1), (1989). 25 Feldpausch TR, Lloyd J, Lewis SL et al. Tree height integrated into pantropical forest biomass estimates. Biogeosciences 9, (212). 2 Breiman L. Random forests. Mach. Learn. 45(1), 5 32 (21) Chave J, Andalo C, Brown SL et al. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145(1), (25). Clark DB, Kellner JR. Tropical forest biomass estimation and the fallacy of misplaced concreteness. J. Veg. Sci. 23(6), (212). 27 Asner GP. Tropical forest carbon assessment: integrating satellite and airborne mapping approaches. Environ. Res. Lett. 4, 349 (29). 28 Asner GP. Painting the world REDD: addressing scientific barriers to monitoring emissions from tropical forests. Environ. Res. Lett. 6, 212 (211). Asner GP, Mascaro J, Muller-Landau HC. A universal airborne LiDAR approach for tropical forest carbon mapping. Oecologia 168(4), (212). 22 Schaaf CB, Gao F, Strahler AH et al. First operational BRDF, albedo nadir reflectance products from MODIS. Remote Sens. Environ. 83(1 2), , (22). 23 Carbon Management (213) 4(6) Baccini A, Friedl M, Woodcock CE, Warbinghton R. Forest biomass estimation over regional scales using multisource data. Geophys. Res. Lett. 31(1), L151 (24). Zwally HJ, Schutz B, Abdalati W et al. ICESat s laser measurements of polar ice, atmosphere, ocean, and land. J. Geodyn. 34(3 4), (22).