Fly Another Day: Spying on Vegetation with Drones in Semi-Arid Rangelands
Abstract
Dryland ecosystems represent 41% of the global land surface and provide critical ecosystem services to 38% of the human population. Additionally, it has become increasingly clear that semi-arid ecosystems play a critical role in the interannual variability of the global carbon cycle. Semi-arid rangelands are heterogenous over fine spatial scales and the density and cover of vegetation influences the variability of the carbon cycle. Dynamic vegetation models lack information to represent accurate vegetation cover at fine spatial scales. Quantification of vegetation of the native landscape of semi-arid ecosystems faces challenges due to the small structural characteristics of these plants. Robust vegetation regime data sets will help elucidate the impact of anthropogenic influences, such as climate change, on dryland ecosystems. Sensors mounted on Uncrewed aerial systems (UAS) can provide measurements of vegetation on the scale at which land managers work and habitat is defined for wildlife. However, due to the heterogeneity of vegetation composition in drylands, we wanted to know: in which shrub community type is multispectral imagery most accurate for measurements of percent cover? During the summer of 2019, we collected multispectral imagery with the MicaSense Rededge, and used structure-from-motion with the DJI Phantom 4 at three distinct sagebrush communities in SW Idaho. Our goal is to determine which combination of structural and spectral measurements allows for the most accurate measurements of percent cover of vegetation. We expect the results from this study to provide detailed information to land owners and land managers who are interested in patterns of vegetation density and coverage for the purposes of ecological monitoring and sustainable land management or ownership, as well as incorporation into dynamic vegetation models. In addition, we aim to expand our analyses by integrating hyperspectral imagery from the Headwall Co-Aligned VNIR-SWIR; these findings will have direct application for monitoring campaigns and future research into the upscaling of UAS to satellite imagery.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.B31M2480R
- Keywords:
-
- 0402 Agricultural systems;
- BIOGEOSCIENCES;
- 0434 Data sets;
- BIOGEOSCIENCES;
- 0480 Remote sensing;
- BIOGEOSCIENCES;
- 1842 Irrigation;
- HYDROLOGY