PS 33-15 - Development of non-destructive methods for estimating effects of woody plant encroachment in Texas savannas using allometry and remote sensing

Wednesday, August 14, 2019
Exhibit Hall, Kentucky International Convention Center
Ane C. F. da Silva1,2, Fernando J. Freire3, Nicole Havrilchak4, Nian-Wei KU5, Pedro A. M. Leite5, Lonesome Malambo6, Sorin Popescu5, Harrison Raub4, Toby SantaMaria4 and Jason West5, (1)Instituto Federal da Paraiba, Princesa Isabel, Brazil, (2)Ciencia Florestal, Universidade Federal Rural de Pernambuco, Recife, Brazil, (3)Agronomia, Universidade Federal Rural de Pernambuco, Recife, Brazil, (4)Ecology and Conservation Biology, Texas A&M University, College Station, TX, (5)Ecosystem Science and Management, Texas A&M University, College Station, TX, (6)Department of Ecology and Conservation Biology, Texas A&M University, College Station, TX
Background/Question/Methods

Woody plant encroachment continues to threaten the biodiversity of grasslands and open savannas globally. The rapidity with which savannas in particular are undergoing change has implications for primary production and ecosystem carbon balance at a range of scales. Estimation and monitoring of the herbaceous-woody biomass changes is required for the successful implementation of strategies for restoring savannas and the ecosystem services they provide, including C sequestration. Using vegetation inventory and remote sensing, we are developing site-specific allometric equations and non-destructive methodologies for biomass estimation of woody species in semiarid savanna in Sonora, Texas. We used 31 0.09-ha plots to describe vegetation structure and composition in 64.9 ha area. Using satellite imaging, we stratified the study area into three strata based on canopy cover, density, and composition. We measured all plants with stems ≥ 5 cm in diameter at breast height (DBH) and total height (Ht). For multi-stemmed plants, we measured all stems ≥ 5 cm and calculated the equivalent DBH. Mean Ht and DBH, total stem density, basal area (BA), and species number were calculated. Unmanned aerial vehicle was employed to fly and capture the aerial photos over study areas. Aerial photos were processed with structure from motion method (SfM) to generate the point cloud data. SfM point cloud data was processed for the comparisons of SfM point cloud data and field measurements. SfM point cloud variable will be compared with field survey to develop allometric equations for estimating the woody plant biomass.

Results/Conclusions

We measured 1424 plants. Plants belonged to eight different species, with 98% of them belonging to four species: Quercus fusiformis, Juniperus ashei, Prosopis glandulosa and Juniperus pinchotii. The DBH, Ht and BA ranged from 5.0 to 55.2 cm, 1.6 to 8.0 m and 0.02 to 2.66 m2.ha-1 across the three strata, respectively. The total stem density was higher in stratum 1 and 2 (p < 0.01). The mean DBH and height were higher in stratum 1 (p < 0.01). The mean BA was higher in strata 1 and 2. The species of greatest density and Ht in descending order were Q. fusiformis, J. ashei, P. glandulosa, and J. pinchotii. J. ashei presented higher mean DBH and mean BA. SfM point cloud data displayed the lowest point density at stratum1 and highest point density at stratum 3. The highest point was in stratum 3 (12.94 m). In addition, 30 SfM point cloud metrics were calculated for future comparisons.