TSWIFT – A scanning tower-based hyperspectral instrument to capture diurnal and seasonal physiological plant response
Francis A Ulep, University of California, Davis, faulep@ucdavis.edu (Presenter)
Troy Sehlin Magney, University of California, Davis, tmagney@ucdavis.edu
Christopher YS Wong, University of California, Davis, cyswong@ucdavis.edu
Huihui Zhang, USDA, huihui.zhang@usda.gov
Kevin Yemoto, USDA, kevin.yemoto@usda.gov
Andrew Schuh, Colorado State University, atmosschuh@gmail.com
Agricultural croplands are a crucial component of the terrestrial carbon budget. Understanding their contributions to the carbon cycle is challenged by uncertainties regarding plant response and recovery to episodic stress events (e.g., drought and heatwaves), which can impact crop productivity and influence seasonal carbon fluxes. Capturing crop response to short-term stresses is challenging to acquire from large-scale, broadband, or multi-spectral datasets. Additionally, remote sensing platforms such as satellites, planes, and UAVs face challenges due to coarse spatiotemporal resolution and cannot acquire plant response data from flash stress events. Here, we utilize a tower-based system that offers automated and continuous monitoring of hyperspectral reflectance (400 - 900 nm) with capabilities to retrieve solar-induced fluorescence (SIF). The Tower Spectrometer on Wheels for Investigations with Frequent Timeseries (TSWIFT) is a self-contained system on a trailer enabling rapid deployment. The instrument is equipped with a scanning RGB camera that pans 360° azimuth and tilts -45° to 90° zenith with a collocated telescope for radiance, enabling repeat sampling of user-specified targets within an area of 14 hectares.
TSWIFT was deployed at the USDA-ARS Limited Irrigation Research Farm in Greeley, Colorado (June – October 2022). We plan to use the SIF data to support our NASA CCS project focused on evaluating SIF as a predictor of semi-arid cropland productivity and refining mechanistic site-level modeling of the SIF – GPP relationship. One of our ongoing and main challenges is disentangling plant structure's complexities and phase angle-induced changes impacting TSWIFT's retrieval signals. We demonstrate an empirical approach to determine a phase angle correction coefficient that can be applied to each observation, rectifying changes due to sun angle and viewing geometry. A phase angle correction will improve the SIF signal to better represent dynamic variation in plant functions for applications in tracking vegetation stress response and assessing productivity.
Poster: Poster_Ulep_2-3_119_35.pdf
Associated Project(s):
Poster Location ID: 2-3
Presentation Type: Poster
Session: Poster Session 2
Session Date: Wed (May 10) 5:15-7:15 PM
CCE Program: TE