Historically inconsistent productivity and respiration fluxes in the global terrestrial carbon cycle

The terrestrial carbon cycle is a major source of uncertainty in climate projections. Its dominant fluxes, gross primary productivity (GPP), and respiration (in particular soil respiration, RS), are typically estimated from independent satellite-driven models and upscaled in situ measurements, respectively. We combine carbon-cycle flux estimates and partitioning coefficients to show that historical estimates of global GPP and RS are irreconcilable. When we estimate GPP based on RS measurements and some assumptions about RS:GPP ratios, we found the resulted global GPP values (bootstrap mean \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${149}_{-23}^{+29}$$\end{document}149−23+29 Pg C yr−1) are significantly higher than most GPP estimates reported in the literature (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${113}_{-18}^{+18}$$\end{document}113−18+18 Pg C yr−1). Similarly, historical GPP estimates imply a soil respiration flux (RsGPP, bootstrap mean of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${68}_{-8}^{+10}$$\end{document}68−8+10 Pg C yr−1) statistically inconsistent with most published RS values (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${87}_{-8}^{+9}$$\end{document}87−8+9 Pg C yr−1), although recent, higher, GPP estimates are narrowing this gap. Furthermore, global RS:GPP ratios are inconsistent with spatial averages of this ratio calculated from individual sites as well as CMIP6 model results. This discrepancy has implications for our understanding of carbon turnover times and the terrestrial sensitivity to climate change. Future efforts should reconcile the discrepancies associated with calculations for GPP and Rs to improve estimates of the global carbon budget.

. Distribution of global gross primary productivity (GPP) reported from the literature. (a) Violin plots of GPP estimated by different methods; (b) distributions of global GPP estimates bootstrapped from the raw data (GPPlit) or aggregated by GPP groups before bootstrap resampling (GPPlit-group) compared with the GPP implied by soil respiration (GPPRs). FLUXNET (GPP estimated based on FLUXNET sites data and upscaling approaches), Isotope (GPP estimated based on atmospheric isotope data (i.e., 18 O and 13 C, ref 1 and 2, respectively)), Mixed (GPP estimated by mixing satellite and site measurements), MODIS (GPP results using models driven by MODIS remote sensing images), SBU (in-situ based upscaling approaches), SIF (GPP estimated based on solar-induced chlorophyll fluorescence).   Method 1 does not use error information when resampling. Methods 2-4 use errors, but handle missing values differently: method 2 replaces missing errors with values calculated from the median coefficient of variability (CV) of non-missing values; method 3 replaces missing errors with values calculated from the maximum CV across the dataset; and method 4 sets missing errors to zero. We used method 3 in the main analysis, which is the most conservative (produces the widest distribution for both RS and GPP).  Table 1 in the main text. Black dots are those for which GPPRs was below the intersection point (127.6 Pg C yr -1 ), while the red dots are above the intersection in Figure 1a in the main text. Rroot (root respiration), RA (autotrophic respiration); NPP (net primary productivity).

Figure 12.
Relationship between RsGPP (global soil respiration as driven by gross primary productivity (GPP) estimates from the literature) and the partitioning variables, as defined in Table 1 in the main text. Black dots are those RsGPP below the intersection point (78.2 Pg C yr -1 ) in Figure 1b in the main text, while the red dots are above it. Rroot (root respiration), RA (autotrophic respiration).  We used the International Geosphere-Biosphere Programme (IGBP) classification layer in the MCD12Q1 to mask all pixels with the land cover types of snow/ice, water, and barren. Table 1. Global soil respiration estimates from the literature (Rslit, Pg C yr -1 ), with any reported 95% confidence interval or stadard deviation (CI or SD, Pg C yr -1 ) and trend (Pg C yr -2 ). Note that n.a. means not available; a Confidence interval; b Standard deviation; † from ref 9 .  Table 2. Global gross primary productivity collected from the literature (GPPlit, Pg C yr -1 ), with any accompanying standard deviation (SD, Pg C yr -1 ) and trend (Pg C yr -2 ). Note that n.a. means not available.

Year
Year Period GPPlit SD Trend Notes Ref.

RS measurement frequency and measurement time
Overall, measurement time and frequency causes no significant bias on annual RS, possibly due to canceling effects 46 .

Estimate global GPP based on remote sensing technology
Remote sensing image related • Remote sensing signals become less reliable over time due to sensor degradation 49,50 . • Cloud contamination [51][52][53] .
• In areas with sparse vegetation, soil background albedo influences reflectivity 54 .

Eddy covariance related
• Products such as FLUXCOM do not account for all C loss pathways or CO2 fertilization effects 3 . • Uncertainties and mismatches in the algorithms that partition towers' net ecosystem exchange into GPP and respiration 55 .  Annual † Root respiration estimated based on soil respiration from ref. 101 .