@InProceedings{FerrazSaaKelLonOme:2019:LiTrCr,
               author = "Ferraz, Antonio and Saatchi, Sassan and Keller, Michael and Longo, 
                         Marcos and Ometto, Jean Pierre Henry Balbaud",
          affiliation = "{NASA Jet Propulsion Laboratory} and {NASA Jet Propulsion 
                         Laboratory} and {USDA Forest Service} and {NASA Jet Propulsion 
                         Laboratory} and {Instituto Nacional de Pesquisas Espaciais 
                         (INPE)}",
                title = "Lidar tree crown detection reveals new patterns of tree density, 
                         height and volume in the Brazilian Amazon",
                 year = "2019",
         organization = "AGU Fall Meeting",
             abstract = "The spatial arrangement of tree crowns regulates light penetration 
                         and ecosystem properties and functions such as biological 
                         diversity, growth, competition, mortality, and recruitment. The 
                         variability of tree density and canopy structure across space and 
                         time are not well known, especially in tropical environments such 
                         as the Amazon forest where there is a large diversity of tree 
                         species with complex crown shapes. Field measurements provide 
                         important data on tree diversity and demographics but measurements 
                         of three-dimensional forest structure from the ground are 
                         extremely limited. Using extensive airborne lidar data, we 
                         extracted individual tree crowns (Ferraz et al., 2016) from 470 
                         samples (6.25ha) located in terra firme forests within the 
                         Brazilian Amazon. The samples were selected from a collection 558 
                         transects (15 km x 0.5km) of airborne lidar data acquired in 2016 
                         in a random design. For trees > 10 m tall, we found an average 
                         tree density of 403 trees ha-1 (range 86 to 630). The highest 
                         average tree density was located in the Japur{\'a}-Solimoes-Negro 
                         moist forest ecoregion (464 trees ha-1, range 341 to 621) with 
                         Guiana shield moist forest having the lowest average (315 trees 
                         ha-1, range 290-336). The lidar survey revealed extremely high 
                         trees in the Guiana shield where the tallest individual measured 
                         88.5 meters, a maximum for the Amazon forest. We modeled basal 
                         area (BA) based on lidar-derived height and crown area using a 
                         neotropical allometric equation. The average BA was 25.7 m2 ha-1 
                         (range 5.9 to 41.8). Guiana shield forests had the largest average 
                         BA (37.7 m2 ha-1, 33.7 to 41.8) with Mato Grosso seasonal forest 
                         showing the lowest BA with 15.59 m2 ha-1 with an average tree 
                         density of 402 trees ha-1. The BA in Mato Grosso ranged from 4.1 
                         to 26.2 m2 ha-1 for samples with tree densities from 156 to 517 
                         trees ha-1, respectively. We further examined tree size along 
                         edaphic, topographic and climate gradients across the Amazon.",
  conference-location = "San Francisco, CA",
      conference-year = "09-13 dec.",
             language = "en",
           targetfile = "ferraz_2019.pdf",
        urlaccessdate = "26 abr. 2024"
}