Search

In this study, we automate tree species classification and mapping using field-based training data, high spatial resolution airborne hyperspectral imagery, and a convolutional neural network classifier (CNN). We tested our methods by identifying seven dominant trees species as well as dead standing trees in a mixed-conifer forest in the Southern Sierra Nevada Mountains, CA (USA) using training, validation, and testing datasets composed of spatially-explicit transects and plots sampled across a single strip of imaging spectroscopy. We also used a three-band ‘Red-Green-Blue’ pseudo true-color subset of the hyperspectral imagery strip to test the classification accuracy of a CNN model without the additional non-visible spectral data provided in the hyperspectral imagery. Our classifier is pixel-based rather than object based, although we use three-dimensional structural information from airborne Light Detection and Ranging (LiDAR) to identify trees (points > 5 m above the ground) and the classifier was applied to image pixels that were thus identified as tree crowns. By training a CNN classifier using field data and hyperspectral imagery, we were able to accurately identify tree species and predict their distribution, as well as the distribution of tree mortality, across the landscape. Using a window size of 15 pixels and eight hidden convolutional layers, a CNN model classified the correct species of 713 individual trees from hyperspectral imagery with an average F-score of 0.87 and F-scores ranging from 0.67–0.95 depending on species. The CNN classification model performance increased from a combined F-score of 0.64 for the Red-Green-Blue model to a combined F-score of 0.87 for the hyperspectral model. The hyperspectral CNN model captures the species composition changes across ~700 meters (1935 to 2630 m) of elevation from a lower-elevation mixed oak conifer forest to a higher-elevation fir-dominated coniferous forest. High resolution tree species maps can support forest ecosystem monitoring and management, and identifying dead trees aids landscape assessment of forest mortality resulting from drought, insects and pathogens. We publicly provide our code to apply deep learning classifiers to tree species identification from geospatial imagery and field training data.

International audience; In mountain landscapes, surface temperatures vary over short distances due to interacting influences of topography and overstory vegetation on local energy and water balances. At two study landscapes in the Sierra Nevada of California, characterized by foothill oak savanna at 276–481 m elevation and montane conifer forest at 1977–2135 m, we deployed 288 near-surface (5 cm above the surface) temperature sensors to sample site-scale (30 m) temperature variation related to hillslope orientation and vegetation structure and microsite-scale (2–10 m) variation related to microtopography and tree overstory. Daily near-surface maximum and minimum temperatures for the 2013 calendar year were related to topographic factors and vegetation overstory characterized using small footprint LiDAR imagery acquired by the National Ecological Observatory Network (NEON) Airborne Observation Platform (AOP). At both landscapes we recorded large site and microsite spatial variation in daily maximum temperatures, and less absolute variation in daily minimum temperatures. Generalized boosted regression trees were estimated to measure the influence of tree canopy density, understory solar radiation, cold-air drainage and pooling, ground cover and microtopography on daily maximum and minimum temperatures at site and microsite scales. Site-scale models based on indices of understory solar radiation and landscape position explained an average of 61–65% of daily variation in maximum temperature; site-scale models based on tree canopy density and landscape position explained 65–83% of variation in minimum temperatures. Models explained

Tall trees and vertical forest structure are associated with increased productivity, biomass and wildlife habitat quality. While climate has been widely hypothesized to control forest structure at broad scales, other variables could be key at fine scales, and are associated with forest management. In this study we identify the environmental conditions (climate, topography, soils) associated with increased tree height across spatial scales using airborne Light Detection and Ranging (LiDAR) data to measure canopy height. The study was conducted over a large elevational gradient from 200 to 3000 m in the Sierra Nevada Mountains (CA, USA) spanning sparse oak woodlands to closed canopy conifer forests. We developed Generalized Boosted Models (GBMs) of forest height, ranking predictor variable importance against Maximum Canopy Height (CHMax) at six spatial scales (25, 50, 100, 250, 500, 1000 m). In our study area, climate variables such as the climatic water deficit and mean annual precipitation were more strongly correlated with CHmax (18–52% relative importance) than soil and topographic variables, and models at intermediate (50–500 m) scales explained the most variance in CHMax (R2 0.77–0.83). Certain soil variables such as soil bulk density and pH, as well as topographic variables such as the topographic wetness index, slope curvature and potential solar radiation, showed consistent, strong associations with canopy structure across the gradient, but these relationships were scale dependent. Topography played a greater role in predicting forest structure at fine spatial scales, while climate variables dominated our models, particularly at coarse scales. Our results indicate that multiple abiotic factors are associated with increased maximum tree height; climatic water balance is most strongly associated with this component of forest structure but varies across all spatial scales examined (6.9–54.8% relative importance), while variables related to topography also explain variance in tree height across the elevational gradient, particularly at finer spatial scales (37.15%, 20.26% relative importance at 25, 50 m scales respectively).

There has been an increasing interest in urban forests and the levels of biodiversity they contain. Currently there are no spatially explicit maps of tree species richness in urban areas. This research tests and identifies GIS and remote sensing metrics (climate, area, productivity, three-dimensional structure) hypothesized to be associated with species richness in native forests and identifies methods that can be applied to predict and map tree species richness in cities. We quantified tree species richness, floristic composition, and structure in 28 1-ha plots in the city of Los Angeles. Climate and remote sensing metrics from high-resolution aerial imagery (10cm), QuickBird (60cm), Landsat (30m), MODIS (250m), and airborne lidar (2m) were collected for each plot. There were 1208 individual stems and 108 trees identified to species. Species richness ranged from 2 to 31 species per ha and averaged 17 species per ha. Tree canopy cover from QuickBird explained the highest portion of variance (54%) in tree species richness followed by NDVI from Landsat (42%). Tree species richness can be higher in residential urban forests than native forests in the United States. Spatially explicit species richness maps at 1ha can be created and tested for cities in order to identify both hotspots and coldspots of tree species richness and changes in species richness over time.

Aim: To determine if habitat diversity, as estimated by climatic and topographic variables, can predict patterns of orchid diversity on different islands and archipelagos with similar explanatory power to biogeographical variables, such as area, isolation and age of an island. Location: Sixty-three islands on eight archipelagos (Solomon Islands, Vanuatu, New Caledonia, Fiji, Samoa, Tonga, Niue and the Cook Islands) in the south-west Pacific. Methods: For each island, we determined the orchid species present, age, area, isolation, and indicators of topographic heterogeneity and climatic variability. We then determined the power of various biogeographical, climatic and topographic variables to predict the number of indigenous and endemic species on archipelagos, and on islands within archipelagos, using generalized linear models (GLMs) and generalized linear mixed models (GLMMs) respectively. Results: We identified a total of 552 species in 110 genera. Area was the only significant biogeographical variable for predicting patterns of orchid species diversity on archipelagos and islands. However, climatic and topographic predictors of habitat diversity performed similarly well. The range in curvature was the best indicator of species richness from the topographic variables, while the range of temperature was the best climatic predictor. These key variables were often strongly correlated with area. Main conclusions: Climatic and topographic variables are useful indicators of habitat diversity. The high explanatory power of area and climatic and topographic predictors, and the strong correlation among these variables, suggests that increasing habitat diversity with increasing area may be the major driver of the species–area relationship. Using climatic and topographic variables as predictors of species richness therefore allows determining the key environmental factors and processes driving species diversity. Refereed/Peer-reviewed

Reducing uncertainty of terrestrial carbon cycle depends strongly on the accurate estimation of changes of global forest carbon stock. However, this is a challenging problem from either ground surveys or remote sensing techniques in tropical forests. Here, we examine the feasibility of estimating changes of tropical forest biomass from two airborne lidar measurements of forest height acquired about 10 yr apart over Barro Colorado Island (BCI), Panama. We used the forest inventory data from the 50 ha Center for Tropical Forest Science (CTFS) plot collected every 5 yr during the study period to calibrate the estimation. We compared two approaches for detecting changes in forest aboveground biomass (AGB): (1) relating changes in lidar height metrics from two sensors directly to changes in ground-estimated biomass; and (2) estimating biomass from each lidar sensor and then computing changes in biomass from the difference of two biomass estimates, using two models, namely one model based on five relative height metrics and the other based only on mean canopy height (MCH). We performed the analysis at different spatial scales from 0.04 ha to 10 ha. Method (1) had large uncertainty in directly detecting biomass changes at scales smaller than 10 ha, but provided detailed information about changes of forest structure. The magnitude of error associated with both the mean biomass stock and mean biomass change declined with increasing spatial scales. Method (2) was accurate at the 1 ha scale to estimate AGB stocks (R2 = 0.7 and RMSEmean = 27.6 Mg ha−1). However, to predict biomass changes, errors became comparable to ground estimates only at a spatial scale of about 10 ha or more. Biomass changes were in the same direction at the spatial scale of 1 ha in 60 to 64% of the subplots, corresponding to p values of respectively 0.1 and 0.033. Large errors in estimating biomass changes from lidar data resulted from the uncertainty in detecting changes at 1 ha from ground census data, differences of approximately one year between the ground census and lidar measurements, and differences in sensor characteristics. Our results indicate that the 50 ha BCI plot lost a significant amount of biomass (−0.8 Mg ha−1 yr−1 ± 2.2(SD)) over the past decade (2000–2010). Over the entire island and during the same period, mean AGB change was 0.2 ± 2.4 Mg ha−1 yr−1 with old growth forests losing −0.7 Mg ha−1 yr−1 ± 2.2 (SD), and secondary forests gaining +1.8 Mg ha yr−1 ± 3.4 (SD) biomass. Our analysis suggests that repeated lidar surveys, despite taking measurement with different sensors, can estimate biomass changes in old-growth tropical forests at landscape scales (>10 ha).