Intercomparison of Remotely Sensed Vegetation Indices, Ground Spectroscopy, and Foliar Chemistry Data from NEON
Abstract
Hyperspectral imaging systems can be used to produce spectral reflectance curves giving rich information about composition, relative abundances of materials, mixes and combinations. Indices based on just a few spectral bands have been used for over 40 years to study vegetation health, mineral abundance, and more. These indices are much simpler to visualize and use than a full hyperspectral data set which may contain over 400 bands. Yet historically, it has been difficult to directly relate remotely sensed spectral indices to quantitative biophysical properties significant to forest ecology such as canopy nitrogen, lignin, and chlorophyll. This linkage is a critical piece in enabling the detection of high value ecological information, usually only available from labor-intensive canopy foliar chemistry sampling, to the geographic and temporal coverage available via remote sensing. Previous studies have shown some promising results linking ground-based data and remotely sensed indices, but are consistently limited in time, geographic extent, and land cover type. Moreover, previous studies are often focused on tuning linkage algorithms for the purpose of achieving good results for only one study site or one type of vegetation, precluding development of more generalized algorithms. The National Ecological Observatory Network (NEON) is a unique system of 47 terrestrial sites covering all of the major eco-climatic domains of the US, including AK, HI, and Puerto Rico. These sites are regularly monitored and sampled using uniform instrumentation and protocols, including both foliar chemistry sampling and remote sensing flights for high resolution hyperspectral, LiDAR, and digital camera data acquisition. In this study we compare the results of foliar chemistry analysis to the remote sensing vegetation indices and investigate possible sources for variance and difference through the use of the larger hyperspectral dataset as well as ground based spectrometer measurements of samples subsequently analyzed for foliar chemistry.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.B23E2125H
- Keywords:
-
- 0414 Biogeochemical cycles;
- processes;
- and modeling;
- BIOGEOSCIENCES;
- 0416 Biogeophysics;
- BIOGEOSCIENCES;
- 0426 Biosphere/atmosphere interactions;
- BIOGEOSCIENCES;
- 0476 Plant ecology;
- BIOGEOSCIENCES