Improved Vegetation Photosynthetic Phenology Monitoring in the Northern Ecosystems Using Total Canopy Solar‐Induced Chlorophyll Fluorescence Derived From TROPOMI

Abstract

Solar‐Induced chlorophyll Fluorescence (SIF) from the TROPOspheric Monitoring Instrument (TROPOMI) with substantially improved spatiotemporal resolutions provides a new potential to improve satellite‐based phenology monitoring. The performance of TROPOMI SIF for tracking vegetation photosynthetic phenology, and how it compares to conventional vegetation indices (VIs)‐based approaches, however, have not been adequately assessed. Total canopy SIF, as a better proxy of Gross Primary Productivity (GPP) than original directional SIF, is a new SIF to estimate phenology while its performance has not been investigated. This study assesses the capability of TROPOMI SIF before and after canopy correction for phenology monitoring and improves our understanding of these questions. Benchmarked by tower‐based GPP, TROPOMI SIF generally performed better than VIs, especially for capturing the End Of Season (EOS) of vegetation photosynthetic activity at deciduous broadleaf forest (DBF), evergreen forest (ENF), and croplands (CRO) sites, but not for Start Of Season (SOS). This suggested that the advantage of SIF over VIs depended on phenological metrics. The total canopy SIF emission obtained through canopy correction generally performed better than the original SIF retrievals, especially in estimating the EOS of forest sites (DBF, MF, ENF), but soil correction did not further improve the accuracy of phenological monitoring. When comparing SIF‐ and VI‐based phenological metrics over northern terrestrial ecosystems, SIF showed earlier senescence date widely, while the differences in onset date were region dependent. These results indicate the necessity of canopy correction to convert directional SIF to canopy total SIF when using satellite SIF products to estimate phenological metrics.

Tracking Gross Primary Productivity Using Satellite Solar Induced Fluorescence: Insights Across Agricultural Ecosystems of India

“…Canopy structure also plays a crucial role in the seasonality of the GPP/ SIF ratio and compounds on the physiological factors (refer to Figures 8e-8g) (Fournier et al, 2012), as the fluorescence escape factor has temporal dynamics, that is, lower during peak growing season due to denser canopy (high LAI) and comparatively higher in early and late growing seasons, which leads to seasonal changes in GPP/SIF ratio. However, it's important to note that the seasonal dynamics of vegetation canopy structure (escape factor) are closely correlated with various climate variables (H. Liu et al, 2023;Miller & Stoner, 1979). This interconnectedness makes it challenging to distinctly isolate the potential influence of canopy structure from that of climate through correlation or machine learning analyses (Chen, Mao, Ricciuto, Lu, et al, 2021).…”

Identifying Crop Growth Stages from Solar-Induced Chlorophyll Fluorescence Data in Maize and Winter Wheat from Ground and Satellite Measurements

“…The Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the Terra (originally known as EOS AM-1) and Aqua (originally known as EOS PM-1) satellites have been widely employed in plant growth observation across extensive geographical regions [45,[63][64][65][66][67]. MODIS provides spatial resolutions of 250 or 500 m and a temporal cadence of 1-2 days, facilitating global-scale dynamic plant growth monitoring.…”

“…While prior research has explored the relationship between SIF and gross primary production, it predominantly focused on land surface phenology, typically limited to the start or end of the growing season [18,45,46]. In contrast, cropland systems involve complexities, demanding more precise and detailed information for crop growth monitoring.…”

“…Firstly, there is a discrepancy between phenological transition dates, such as the start of the season, and crop growth stages like the silking stage of maize and heading stage of winter wheat, creating a temporal gap [5,42]. Secondly, many studies on crop phenology estimation are constrained within the broader framework of plant phenology, focusing primarily on the start, peak, and end of the growing season [43][44][45]. This limitation hinders the alignment of phenological estimations with actual crop growth stage observations in agronomy.…”

“…This limitation hinders the alignment of phenological estimations with actual crop growth stage observations in agronomy. Thirdly, while SIF data has been widely proven effective in tracking GPP and crop growth observation, its potential for identifying specific crop growth stages remains unexplored [45,46].…”