Improving the monitoring of deciduous broadleaf phenology using the Geostationary Operational Environmental Satellite (GOES) 16 and 17

Abstract

Abstract. Monitoring leaf phenology tracks the progression of climate change and seasonal variations in a variety of organismal and ecosystem processes. Networks of finite-scale remote sensing, such as the PhenoCam network, provide valuable information on phenological state at high temporal resolution, but they have limited coverage. Satellite-based data with lower temporal resolution have primarily been used to more broadly measure phenology (e.g., 16 d MODIS normalized difference vegetation index (NDVI) product). Recent versions of the Geostationary Operational Environmental Satellites (GOES-16 and GOES-17) can monitor NDVI at temporal scales comparable to that of PhenoCam throughout most of the western hemisphere. Here we begin to examine the current capacity of these new data to measure the phenology of deciduous broadleaf forests for the first 2 full calendar years of data (2018 and 2019) by fitting double-logistic Bayesian models and comparing the transition dates of the start, middle, and end of the season to those obtained from PhenoCam and MODIS 16 d NDVI and enhanced vegetation index (EVI) products. Compared to these MODIS products, GOES was more correlated with PhenoCam at the start and middle of spring but had a larger bias (3.35 ± 0.03 d later than PhenoCam) at the end of spring. Satellite-based autumn transition dates were mostly uncorrelated with those of PhenoCam. PhenoCam data produced significantly more certain (all p values ≤0.013) estimates of all transition dates than any of the satellite sources did. GOES transition date uncertainties were significantly smaller than those of MODIS EVI for all transition dates (all p values ≤0.026), but they were only smaller (based on p value <0.05) than those from MODIS NDVI for the estimates of the beginning and middle of spring. GOES will improve the monitoring of phenology at large spatial coverages and provides real-time indicators of phenological change even when the entire spring transition period occurs within the 16 d resolution of these MODIS products.

Fusing Geostationary Satellite Observations with Harmonized Landsat-8 and Sentinel-2 Time Series for Monitoring Field-Scale Land Surface Phenology

“…Observations from geostationary satellites have much higher effectiveness than MODIS data to track the seasonality of vegetation growth and phenology detections [39,40,43,66]. As revealed in the previous studies, cloud contaminations are the major impacts on the quality of EVI2 time series for phenology detections [16,36].…”

Remote sensing of phenology: Towards the comprehensive indicators of plant community dynamics from species to regional scales

“…Progress in geostationary satellite systems offers critical advantages of monitoring specific locations over extended time frames and different times of the day, which becomes especially important for regions unreachable by frequent field surveys or ground‐based phenological observation networks (Hashimoto et al, 2021; Wheeler & Dietze, 2021). For example, the application of GOES Advanced Baseline Imager over Amazon tropical forests (Hashimoto et al, 2021) revealed presence of seasonality in 85% of the region's evergreen forests, which would not be possible to detect with traditional sun‐synchronous sensing.…”

“…Geostationary systems can also substantially reduce the impact of clouds on estimation of phenological transition dates and improve the assessments of phenological trends and shifts over vast regions and multi‐year periods. To better understand potential uncertainties due to still coarse spatial resolutions (~1 km) of such products, their time series can be cross‐compared with high frequency in situ observations such as phenocam images (Wheeler & Dietze, 2021).…”

Reducing model uncertainty of climate change impacts on high latitude carbon assimilation

“…In addition, because they provide a synoptic view over multiple decades, satellite‐derived phenology metrics can also be used to track changes in plant seasonality (Zeng et al, 2011), which allows us to determine the controls of plant seasonality on carbon cycling and storage (Pulliainen et al, 2017; Richardson et al, 2010; Zhu et al, 2013). More recently, the characterization of vegetation phenology has become more routine, particularly with the creation of standard data products (Ganguly et al, 2010), while novel platforms now enable higher temporal resolutions (Wheeler & Dietze, 2021). Similarly, biophysical products, including LAI, are often used to capture seasonality of plants, but also provide a more meaningful metric for evaluating carbon cycle models (Li et al, 2018; Viskari et al, 2015).…”