Toward a Comprehensive Understanding of Global Vegetation CO 2 Assimilation from Space
Sujong Jeong,Hayoung Park
DOI: https://doi.org/10.1111/gcb.15475
IF: 11.6
2020-01-01
Global Change Biology
Abstract:Large-scale global analysis of the relationship between growing season solar-induced chlorophyll fluorescence (SIF) and gross primary productivity (GPP), indicated by the GPP/SIF ratio, varied greatly with higher values found in wet-and-cold climate regions and lower values found in hot-and-dry climate regions. Such pattern has been shown to be most influenced by the environmental factor of moisture availability. The recent emergence of satellite remote sensing of solar-induced chlorophyll fluorescence (SIF) has opened up new opportunities in continuous large-scale measurements of gross primary productivity (GPP). However, uncertainties remain in the contributing factors that determine the relationship between SIF and GPP. Analyzing climatic factors to explain large-scale variations in the relationship between SIF and GPP reveals moisture availability could be the primary mechanism that determines the spatial variation of the GPP/SIF ratio on a global scale. Terrestrial ecosystems reduce the increase in atmospheric CO2 concentrations by absorbing approximately 33% of total anthropogenic carbon emissions, including fossil fuel combustion and land-use. The amount of CO2 fixation by terrestrial vegetation per unit time as a result of the photosynthetic reduction of CO2 into organic components is GPP. Many recent studies have shown various changes in vegetation structure, growing season, and functions associated with climate change (Park et al., 2020; Piao et al., 2019). These changes in vegetation properties imply that there are also concurrent GPP changes across the world. An accurate understanding of GPP is fundamental to a comprehensive assessment of the role of terrestrial ecosystems in the global carbon cycle, as well as in climate change, but global GPP is hard to measure. On a global-scale perspective, it is difficult to have a clear estimate of GPP and to understand how GPP is altered by climate factors due to the lack of relevant information and spatial and temporal variations in measurements of factors which are used to estimate GPP. Previous studies utilized eddy covariance tower measurements, process-based models, and light-use efficiency models to estimate GPP. Process-based models are useful for estimating responses of GPP to environmental changes, but they still have inherent uncertainties in the model structure and parameters. Flux tower estimates of GPP directly observe the carbon exchange between vegetation and the atmosphere; however, the number of measurement sites is insufficient for representing global-scale variations because of the large spatial heterogeneity between sites. Light-use efficiency models incorporating satellite data also serve as quasi-observations of GPP of the global land surface. However, models using the light-use efficiency approach are limited by the availability of sufficient long-term satellite information to operate the models. In addition, GPP from light-use efficiency models could be strongly constrained by atmospheric correction of data. To overcome the weaknesses in previous methods, satellite remote sensing of SIF has emerged as an alternative method for estimating GPP (Mohammed et al., 2019). SIF is the re-emission of a small fraction of absorbed radiation in the red to near infra-red spectral regions from excited chlorophyll-a molecules during photosynthesis. SIF is a reliable physiological measure of GPP as it circumvents the need for light-use efficiency parameters and offers a more direct physiologically based proxy to plant photosynthetic activity than other methods of estimating GPP (He et al., 2020). Compared to traditional reflectance-based vegetation indices, which provide information on vegetation structure and greenness, SIF provides insight into the physiological functioning of plants and thereby allows for a better mechanistic understanding of vegetation-atmosphere carbon exchange of terrestrial ecosystems (Frankenberg et al., 2011). Recent empirical studies from an ecosystem to a global level show SIF displays strong agreements with diurnal, seasonal, and annual variations of GPP (Köhler et al., 2018). Although relationships between SIF and GPP on short temporal scales have shown large uncertainties, satellite-measured SIF has a strong positive correlation with GPP across several different biomes on longer temporal and larger spatial scales (Li et al., 2018). However, even with satellite measurements of SIF for direct estimates of global GPP, there are still limitations in understanding the contributing factors that determine large-scale spatial and temporal variations in the relationship between SIF and GPP. The recent study of Chen et al. (2020) found climatic factors explained large-scale variations in the relationships between SIF and GPP, using several different satellite measurements of SIF and flux tower- and satellite-based estimates of GPP. Using the ratio of SIF to GPP, they found that GPP/SIF showed a bi-modal hump-shaped latitudinal distribution, with a higher ratio of GPP/SIF in northern high latitudes and in tropical rainforests in the Amazon, Congo basin, and Southeast Asia. In contrast, a relatively low ratio of GPP/SIF was observed in dry regions such as central Asia, Southwest America, Australia, and Southwest Africa. The observed regional differences of GPP/SIF are clearly different from the prominent regional patterns that both SIF and GPP each show by themselves. Analysis of GPP/SIF with climate variables such as temperature, precipitation, and shortwave radiation using machine-learning methods revealed the climatic constraints on the ratio of GPP/SIF. In their study, Chen et al. (2020) found that the ratio of GPP/SIF showed a decreasing trend from regions with cold and wet climates to regions with hot and dry climates. This climatic gradient of GPP/SIF indicates the importance of moisture availability on spatial variations in GPP/SIF. In addition, at a given temperature, an increase in precipitation is related to an increase in GPP/SIF. To verify the climatic factors determining spatial variations of the GPP/SIF, Chen et al. (2020) performed an estimation of the GPP/SIF ratio, using machine learning with climatic variables. While regional differences exist in the main climatic factors that control machine-learning estimations of GPP/SIF variations, precipitation was denoted as the primary factor for the GPP/SIF ratio variations on a global scale. Chen et al. (2020) suggested a possible mechanism for the role of moisture availability in the GPP/SIF variations. If vegetation has sufficient moisture for physiological activity, temperature and radiation could equally affect vegetation productivity and fluorescence emissions. However, if vegetation has insufficient water, stomatal closure could have a larger impact on GPP than SIF. This asymmetric effect related to moisture availability is supported by a newly developed photosynthesis model (Gu et al., 2019). The model, constructed with SIF and a few physiological and structural parameters, demonstrated that moisture stress could lead to a decreasing GPP/SIF ratio by changing CO2 partial pressure inside the chloroplast stroma. In addition, the moisture limitations on GPP/SIF have strong implications for climate change. With increasing greenhouse gases in the atmosphere, the climate will continue to become much warmer and drier, eventually enhancing atmospheric aridity (Park et al., 2018). As a result, the importance of moisture stress, emphasized by Chen et al. (2020), will be enhanced in future climate predictions of ecosystem change and CO2 flux estimations from plant productivity. In addition, the model also demonstrated that the chlorophyll fluorescence emission ratio, which is the ratio of the fluorescence yield at the two fluorescence-emission maxima in the 690 and 730 nm regions, could be another factor influencing the GPP/SIF ratio, especially when the canopy structure is complex. A study by Dechant et al. (2020), which studied the physiological and structural components of SIF and their relation to photosynthesis using in-situ datasets, found that canopy structure plays a dominant role in the relationship between GPP and SIF. More studies on climatic variables and other parameters such as fluorescence emittance and yield and how these factors affect GPP can further improve remote sensing estimations of GPP. The finding that moisture constraints are the key environmental control on GPP/SIF variation will be useful for using satellite-measured SIF to estimate GPP across the globe. Findings from this study provide a somewhat different conclusion than the results from a site-level study (Li et al., 2018); however, such differences are expected as many different aspects of measurements (such as time, space, and frequency of observations) could affect the response of GPP/SIF to climate variables. Nonetheless, through this study, we now have a better understanding of spatial variations on the ratio of GPP/SIF, which brings us a step further toward a comprehensive understanding of global-scale vegetation CO2 assimilation from space. Sujong's laboratory is funded by the Creative-Pioneering Researchers Program through Seoul National University (SNU). Data sharing not applicable – no new data generated.