Monitoring drought impacts on crop productivity of the U.S. Midwest with solar-induced fluorescence: GOSIF outperforms GOME-2 SIF and MODIS NDVI, EVI, and NIRv
Introduction
Drought is one of the most widespread disasters in the world that endanger natural and managed ecosystems. Drought can substantially reduce gross primary production (GPP) and thereby the carbon sink of terrestrial ecosystems. Droughts can bring irreversible loss to the fragile and sensitive agro-ecosystems (Lobell et al., 2014; Wang et al., 2016; Xu et al., 2019), and can also cause widespread crop mortality and an increase in the frequency of pests and diseases, further reducing crop yield. For example, the 2012 flash drought in the U.S. Central Great Plains, the most severe drought since 1930 (Otkin et al., 2016; Wolf et al., 2016), led to agricultural losses amounting to $20 billion (Kam et al., 2014). What's more, global warming also increases the frequency and severity of drought (Boisier et al., 2015; Cook et al., 2015). Therefore, it is essential to monitor the responses of crop growth condition and crop yield to drought at regional to global scales (Wang et al., 2019).
The eddy covariance (EC) technique is generally considered as the most accurate method for estimating GPP at the ecosystem scale (Baldocchi et al., 2001). The EC flux towers directly observe NEE (Net Ecosystem Exchange), and GPP can be derived from NEE using nighttime-based or daytime-based partitioning method (Reichstein et al., 2005). However, the EC techniquet can only estimate GPP at the ecosystem scale and reveal the effects of drought on vegetation within the footprint of the EC tower. Uneven and sparse distributions of EC flux towers make it challenging to monitor drought effects on ecosystem productivity at large spatial scales (Qiu et al., 2020b). In contrast, the remote sensing technique is well suited for monitoring the effects of drought on ecosystems over broad spatial domains (West et al., 2019). The occurrence of drought is generally accompanied by a decrease in precipitation and an increase in temperature. When plants are subjected to water and/or heat stress, photosynthesis or productivity usually declines. The changes in plant productivity can be detected by satellite-derived vegetation indices (VIs). Numerous studies have shown that VIs such as the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) are able to monitor the growth status of vegetation and estimate crop yield (Huang et al., 2019; Labus et al., 2002; Lee et al., 2017; Potgieter et al., 2007; Satir and Berberoglu, 2016). However, VIs usually indicate the greenness or chlorophyll content of vegetation canopy and do not accurately capture rapid changes in vegetation photosynthesis due to water and heat stresses (Smith et al., 2018). A recently proposed new VI, near-infrared reflectance of terrestrial vegetation (NIRv, the product of NDVI and NIR reflectance) (Badgley et al., 2017) was found to be strongly correlated with GPP over different spatial and temporal scales (Wang et al., 2021; Wu et al., 2020). Several studies have also demonstrated the ability of NIRv to estimate crop yield. In contrast to NDVI, both EVI and NIRv do not saturate in areas with high density vegetation cover (Baldocchi et al., 2020). It is worth noting that NIRv estimates the global GPP with high accuracy without the input of additional environmental variables (Badgley et al., 2019). Hence, NIRv can be used as a proxy for GPP to explore photosynthesis of vegetation.
Recent satellite observations of solar-induced chlorophyll fluorescence (SIF) have provided a promising technique to measure plant photosynthesis from space (Frankenberg et al., 2011; Frankenberg et al., 2014; Joiner et al., 2013; Joiner et al., 2011; Li and Xiao, 2022; Li et al., 2018a; Sun et al., 2017). Sunlight absorbed by plants is partly used for photosynthesis and partly dissipated as heat via non-photochemical quenching (NPQ), while a very small fraction of the absorbed energy is emitted as SIF (Baker, 2008). Therefore, SIF is a by-product of the photosynthesis process of vegetation and is physiologically related to GPP (Joiner et al., 2011; Li et al., 2018b; Yang et al., 2018). This also means that SIF may be sensitive to water and heat stresses which affect photosynthesis. Currently, several satellite missions, including SCIAMACHY (Joiner et al., 2016), GOSAT (Greenhouse gases Observing SATellite) (Frankenberg et al., 2011; Joiner et al., 2011), GOME-2 (The Global Ozone Monitoring Experiment-2) (Joiner et al., 2013), OCO-2 (Orbiting Carbon Observatory-2) (Frankenberg et al., 2014) and TROPOMI (TROPOspheric Monitoring Instrument) (Kohler et al., 2018) provide global SIF data at different spatial and temporal resolutions. SIF is increasingly used to estimate GPP, and has been shown as a stronger proxy of GPP than VIs (Damm et al., 2010; Frankenberg et al., 2011; Guanter et al., 2012; Li et al., 2018b; Parazoo et al., 2014; Smith et al., 2018). More and more studies have also shown that SIF has a high sensitivity to environmental stress, and therefore has a strong potential for detecting vegetation phenology and diagnosing the responses of ecosystems to water and heat stresses.
Previous studies have shown that satellite-derived SIF from GOSAT and GOME-2 captured the changes in photosynthesis under drought conditions for different regions such as the Amazon (Lee et al., 2013), Europe (Wang et al., 2020) and U.S. Great Plains (He et al., 2020; Sun et al., 2015). These studies have demonstrated the potential of SIF in monitoring vegetation photosynthesis and its response to drought. However, the SIF data used in these studies have coarse spatial and temporal resolutions (e.g., 0.5° and monthly), which may lead to a weak ability to capture the timely photosynthetic changes in vegetation, especially for heterogeneous regions (Qiu et al., 2020a). For example, GOME-2 provides SIF data with a spatial resolution of 40 km × 80 km (40 km × 40 km after July 2013) (Koehler et al., 2015), while the spatial resolution of GOSAT SIF data is 10 km in diameter (Joiner et al., 2012). OCO-2 provides SIF observations with much smaller footprints (1.3 km × 2.25 km), but has sparse coverage across the globe (Qiu et al., 2020a; Shi et al., 2021). These disadvantages to some extent hinder the application of these SIF data in carbon cycle studies from the ecosystem scale to the global scale. To address this issue, researchers have developed SIF data with high temporal and spatial resolutions using machine learning method (Li and Xiao, 2019; Zhang et al., 2018). For example, the global, OCO-2 based SIF product (GOSIF) (Li and Xiao, 2019) has finer resolutions (0.05°, 8-day) compared to the original OCO-2 SIF data, and it also has continuous global coverage and a longer time period (2000 to present), which allows us to better explore the responses of vegetation photosynthesis to drought at different spatial and temporal scales (Li et al., 2020). In addition, the sensitivity of crop NDVI, EVI, NIRv, and GOSIF to drought is not clear at present. Moreover, the yield of crops is closely related to photosynthesis. When subjected to prolonged water and heat stresses, the photosynthetic capacity of crops decreases, which may also lead to lower crop yield. Previous studies have shown that coarse-resolution SIF has a great potential in detecting heat stress in wheat in a timely manner and in assessing the impact of drought on wheat yield (Song et al., 2018), while it is not clear to what extent the fine-resolution SIF data can help.
In this study, we explored the performance of the GOSIF data with finer spatial and temporal resolution in monitoring the variations of crop productivity in response to drought at different spatial scales. We selected the widely reported 2012 drought in the U.S. Midwest as the representative drought event. We compared the performance of GOSIF with satellite-derived VIs (NDVI, EVI, and NIRv) and coarse-resolution GOME-2 SIF in capturing the changes of crop productivity over the course of the seasonal cycle under drought conditions. We hypothesized that (1) SIF performs better than VIs because SIF is considered to contain environmental information related to photosynthetic light use efficiency (Li et al., 2018b; Yang et al., 2015); (2) GOSIF performs better than the GOME-2 SIF because GOSIF is based on high-quality OCO-2 SIF data and also has finer spatial and temporal resolution than GOME-2 SIF.
Section snippets
Study area
Our study area is the U.S. Midwest, including eight states (Nebraska, Kansas, Iowa, Missouri, Illinois, Wisconsin, Indiana, and Ohio) (Fig. 1). Corn and soybeans are widely planted in this region, which accounts for a large portion of the Corn Belt. It is one of the world's major crop-producing areas. The planting area of corn and soybeans accounts for about 41% of the study area. The average annual production of soybeans and corn is about 2 billion bushels and 8.9 billion bushels,
Characterization of the 2012 drought
As shown in Fig. 2a, the sc_PDSI was substantially lower throughout the year in 2012 than the multiyear average. The precipitation decreased sharply from March 2012 and did not return to the normal level until December (Fig. 2b). The average monthly precipitation was 81 mm in the reference years and decreased by 28.4% in 2012. In particular, the precipitation in the growing season decreased by about 76 mm and 52 mm for soybeans and corn, respectively. The average temperature from January to
Discussion
Our results showed that SIF from the GOSIF product performed better in monitoring the responses of crop productivity to drought than vegetation indices derived from MODIS. Due to the close relation of SIF with vegetation photosynthesis, SIF can reveal how the photosynthesis of vegetation is affected by environmental stresses (Li et al., 2020). Previous studies showed that SIF was sensitive to water stress and heat stresses (Buerling et al., 2013; Ni et al., 2015; Rahbarian et al., 2011;
Conclusions
We used SIF (GOSIF and GOME-2 SIF), NDVI, EVI, and NIRv data to evaluate the impact of the 2012 drought on crop productivity in the U.S. Midwest. We compared the seasonal cycles and spatial anomalies of SIF and VIs data in the drought year relative to the reference years. We also evaluated the performance of SIF and VIs for estimating yields of corn and soybean. Overall, GOSIF is more sensitive to the response of precipitation and temperature compared to VIs and coarse spatial resolution GOME-2
Declaration of Competing Interest
I hereby certify that this paper consists of original, unpublished work which is not under consideration for publication elsewhere, and all the authors listed have approved the manuscript that is enclosed. We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant No.41971283, 41801261, 41827801, 41901274, 41971352). J.X. was supported by University of New Hampshire. We thank the GOME-2 and MODIS research teams for producing GOME-2 SIF and MODIS reflectance and VI data, and USDA-NASS for providing crop yield data. We thank AmeriFlux for making the flux data publicly available, and thank Dr. Andy Suyker for providing the flux data.
References (82)
Effect of water stress at different development stages on vegetative and reproductive growth of corn
Field Crops Res.
(2004)Prospects for chlorophyll fluorescence remote sensing from the Orbiting Carbon Observatory-2
Remote Sens. Environ.
(2014)Retrieval and global assessment of terrestrial chlorophyll fluorescence from GOSAT space measurements
Remote Sens. Environ.
(2012)- et al.
Synergistic use of SMAP and OCO-2 data in assessing the responses of ecosystem productivity to the 2018 U.S. drought
Remote Sens. Environ.
(2020) - et al.
Chlorophyll fluorescence observed by OCO-2 is strongly related to gross primary productivity estimated from flux towers in temperate forests
Remote Sens. Environ.
(2018) - et al.
Retrospective droughts in the crop growing season: implications to corn and soybean yield in the Midwestern United States
Agric. For. Meteorol.
(2010) Assessing the evolution of soil moisture and vegetation conditions during the 2012 United States flash drought
Agric. For. Meteorol.
(2016)Assessing the benefit of satellite-based Solar-Induced Chlorophyll Fluorescence in crop yield prediction
Int. J. Appl. Earth Observ. Geoinform.
(2020)- et al.
Crop yield prediction under soil salinity using satellite derived vegetation indices
Field Crops Res.
(2016) - et al.
Gross primary production and ecosystem respiration of irrigated maize and irrigated soybean during a growing season
Agric. For. Meteorol.
(2005)
Comparison of ROS formation and antioxidant enzymes in Cleome gynandra (C-4) and Cleome spinosa (C-3) under drought stress
Plant Sci.
Warmer spring alleviated the impacts of 2018 European summer heatwave and drought on vegetation photosynthesis
Agric. For. Meteorol.
Sun-induced chlorophyll fluorescence is more strongly related to absorbed light than to photosynthesis at half-hourly resolution in a rice paddy
Remote Sens. Environ.
Fluorescence, temperature and narrow-band indices acquired from a UAV platform for water stress detection using a micro-hyperspectral imager and a thermal camera
Remote Sens. Environ.
Terrestrial gross primary production: using NIRV to scale from site to globe
Global Change Biol.
Canopy near-infrared reflectance and terrestrial photosynthesis
Sci. Adv.
Chlorophyll fluorescence: a probe of photosynthesis in vivo
Annu. Rev. Plant Biol.
FLUXNET: a new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities
Bull. Am. Meteorol. Soc.
Outgoing Near-Infrared radiation from vegetation scales with canopy photosynthesis across a spectrum of function, structure, physiological capacity, and weather
J.Geophys. Res.-Biogeosci.
Projected strengthening of Amazonian dry season by constrained climate model simulations
Nat. Climate Change
Fluorescence-based sensing of drought-induced stress in the vegetative phase of four contrasting wheat genotypes
Environ. Exp. Bot.
Unprecedented 21st century drought risk in the American Southwest and Central Plains
Sci. Adv.
Remote sensing of sun-induced fluorescence to improve modeling of diurnal courses of gross primary production (GPP)
Global Change Biol.
Water-use efficiency in Flaveria species under drought-stress conditions
Photosynthetica
The fluorescence explorer mission concept-ESA's earth explorer 8
IEEE Trans. Geosci. Remote Sens.
Meteorological, agricultural and socioeconomic drought in the Duhok Governorate, Iraqi Kurdistan
Nat. Hazards
New global observations of the terrestrial carbon cycle from GOSAT: patterns of plant fluorescence with gross primary productivity
Geophys. Res. Lett.
The ability of sun-induced chlorophyll fluorescence from OCO-2 and MODIS-EVI to monitor spatial variations of soybean and maize yields in the Midwestern USA
Remote Sens.
Chapter 2. High-Yield Maize–Soybean Cropping Systems in the US Corn Belt
Chapter 2. High-Yield Maize–Soybean Cropping Systems in the US Corn Belt
Light saturated RuBP oxygenation by Rubisco is a robust predictor of light inhibition of respiration in Triticum aestivum L
Plant Biol.
Improving the monitoring of crop productivity using spaceborne solar-induced fluorescence
Global Change Biol.
Global and time-resolved monitoring of crop photosynthesis with chlorophyll fluorescence
Proc. Natl. Acad. Sci. U. S. A.
Tracking seasonal and interannual variability in photosynthetic downregulation in response to water stress at a temperate deciduous forest
J. Geophys. Res. Biogeosci.
Causes and predictability of the 2012 great plains drought
Bull. Am. Meteorol. Soc.
Evaluating the performance of satellite-derived vegetation indices for estimating gross primary productivity using FLUXNET observations across the globe
Remote Sens.
The 2012 flash drought threatened US midwest agroecosystems
Chin. Geogr. Sci.
Global monitoring of terrestrial chlorophyll fluorescence from moderate-spectral-resolution near-infrared satellite measurements: methodology, simulations, and application to GOME-2
Atmosp. Meas. Tech.
New methods for the retrieval of chlorophyll red fluorescence from hyperspectral satellite instruments: simulations and application to GOME-2 and SCIAMACHY
Atmosp. Meas. Tech.
Filling-in of near-infrared solar lines by terrestrial fluorescence and other geophysical effects: simulations and space-based observations from SCIAMACHY and GOSAT
Atmosp. Meas. Tech.
First observations of global and seasonal terrestrial chlorophyll fluorescence from space
Biogeosciences
Did a skillful prediction of sea surface temperatures help or hinder forecasting of the 2012 Midwestern US drought?
Environ. Res. Lett.
Cited by (56)
Evident influence of water availability on the relationship between solar-induced chlorophyll fluorescence and gross primary productivity in the alpine grasslands of the Tibetan Plateau
2024, International Journal of Applied Earth Observation and GeoinformationSoil moisture retrieval by a novel hybrid model based on CYGNSS and Sun-induced fluorescence data
2024, Journal of HydrologyContrasting responses of relationship between solar-induced fluorescence and gross primary production to drought across aridity gradients
2024, Remote Sensing of EnvironmentWater availability and atmospheric dryness controls on spaceborne sun-induced chlorophyll fluorescence yield
2024, Remote Sensing of EnvironmentDetecting drought stress occurrence using synergies between Sun induced fluorescence and vegetation surface temperature spatial records
2024, Science of the Total Environment