An abrupt shift in gross primary productivity over Eastern China-Mongolia

doi:10.21203/rs.3.rs-3010327/v1

Download PDF

Article

An abrupt shift in gross primary productivity over Eastern China-Mongolia

https://doi.org/10.21203/rs.3.rs-3010327/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The terrestrial ecosystem in East Asia exhibits significant variability in the gross primary productivity (GPP), especially in semi-arid regions that are vulnerable to climate change. This GPP variation significantly modulates the local carbon budget, but our understanding of this and the underlying mechanism is still lacking. Here, we examine the GPP variability in East Asia and its relationship with climate change over the last three decades. We identify an abrupt decrease in GPP over the Eastern China-Mongolia region around the year 2000. This is attributed to an abrupt decrease in precipitation associated with the phase shift of the Pacific decadal oscillation. Of the twelve offline land surface models, eight could simulate this abrupt response, while the others fail due to the combination of exaggerated CO₂ fertilization effect and underrated climate impact. For accurate prediction, it is necessary to improve the sensitivity of the GPP to changes in CO₂ concentrations and the climate system.

Earth and environmental sciences/Biogeochemistry/Carbon cycle

Earth and environmental sciences/Climate sciences/Biogeochemistry/Carbon cycle

The land carbon sink offsets a considerable amount of anthropogenic CO₂ emissions; it sequestered a total of 3.1 ± 0.6 GtC yr^− 1, from 2012 to 2021, which accounts for 29% of total CO₂ emissions¹. Variability of the land carbon sink is a major driver of the interannual variability of atmospheric CO₂ concentrations^2–4. Gross Primary Productivity (GPP), the organic carbon sequestrated by terrestrial plants through photosynthesis, has a significant impact on the variability of the land carbon sink⁵. The variability of GPP is significantly influenced by global environmental changes, such as CO₂ concentrations and meteorological factors⁶. In detail, GPP increases in response to rising atmospheric CO₂ concentrations, which is referred to as the CO₂ fertilization effect^7,8. In addition, GPP depends on climate conditions, as abiotic changes, such as precipitation and temperature, significantly affect the productivity of terrestrial ecosystems^4,9,10. Thus, the amount of carbon sequestration varies with these factors, and GPP could exhibit strong inter-annual and intra-seasonal variability¹.

As meteorological factors vary under a normal distribution, it is expected that the variation in GPP is within a certain range. However, strong abiotic forcings can cause abrupt changes in GPP. Specifically, abrupt changes in terrestrial carbon flux, which are non-linear and non-stationary, can result from extreme climate events such as heat waves, cold surges, floods, and droughts¹¹. These changes can deteriorate the local carbon uptake. Therefore, to avoid underestimating the potential risks of degradation to terrestrial ecosystems, it is essential to understand the response of GPP to climate change, especially extreme events, and the underlying mechanisms.

Previous studies have reported abrupt changes in vegetation productivity and their mechanisms. Ma et al.¹² have examined how climate extremes impact terrestrial ecosystems in southeastern Australia, with abrupt changes in phenology and vegetation productivity. Berdugo et al.¹³ have investigated the ubiquity and causes of abrupt vegetation shifts in global drylands. It has been shown that abrupt changes in productivity losses due to climatic, topographic, edaphic, and human factors are commonly detected in dryland dynamics. Notably, semi-arid forests have shown dominant abrupt decreasing shifts in the Normalized Difference Vegetation Index (NDVI) over the last two decades¹³. As semi-arid ecosystems are highly sensitive to hydroclimatic variations, this could lead to a loss of ecosystem resilience¹².

Fossil fuel CO₂ emissions in East Asia are about 1.5 PgC yr^− 1, and about 13–27% of these emissions have been offset by its terrestrial ecosystem from 1990 to 2009⁷. Therefore, East Asia is known to be a significant contributor to the global carbon cycle^7,14,15. Based on the prevailing semi-arid regions and historical extreme climate events in Inner East Asia, it is possible that there have been abrupt changes in GPP over East Asia¹⁶. In this regard, understanding the abrupt changes in terrestrial carbon flux in East Asia is important to consider the local carbon budget for achieving carbon neutrality. However, the abrupt changes in GPP and their mechanisms in this region remain unclear. Although several studies have examined the change points of GPP in East Asia^5,6,17, they have mostly focused on the Chinese region, and the detailed mechanisms of the abrupt changes in terrestrial carbon flux have not yet been fully examined.

The primary objective of this study is to identify the major variability and abrupt changes in GPP in East Asia over the last three decades. We examine the dominant spatio-temporal characteristics of GPP variability, and identify the significant abrupt changes in GPP and their main drivers. Furthermore, we investigate the consistency in detecting abrupt changes with multiple GPP datasets to validate their simulation capabilities for abrupt changes in GPP under greenhouse gas warming and relevant climate changes.

The abrupt shift in GPP over Eastern China-Mongolia

We examine the first leading mode of GPP variability over East Asia for the period of 1982–2013 by using the Empirical Orthogonal Function (EOF) (Fig. 1). The first eigenvectors and principal components (PCs) of GPP_FLUXCOM and GPP_NIRv explain 20.9% and 14.1% of the total variance, respectively. The maximum variation of GPP_FLUXCOM and GPP_NIRv is consistently located in the Eastern China-Mongolia region (40°–52°N and 110°–124°E). In addition, the PCs of both first leading modes consistently show a distinctive phase shift around the year 2000. Specifically, there is a positive phase until the 2000s and a negative phase afterward, suggesting a decrease in vegetation productivity around the year 2000 in the Eastern China-Mongolia region. Note that the phase shift in GPP with decadal time scale is comparable in magnitude to their interannual variability, and even larger for GPP_NIRv in particular.

Eastern China-Mongolia is a semi-arid temperate grassland region with limited precipitation^18,19. Previous studies have reported that vegetation productivity is strongly dependent on water availability in semi-arid regions across Northern China^17,19, suggesting the possibility that changes in GPP are caused by changes in precipitation over the Eastern China-Mongolia region. Therefore, we further examined the changes in GPP and precipitation in this region and their relationship (Fig. 2). There is a prominent decreasing shift in GPP in the late 1990s; a positive anomaly before the late 1990s, and a negative anomaly afterwards. This result is consistent with the PCs of the first leading mode shown in Fig. 1c and d. The pronounced decreasing shift in precipitation in the late 1990s is also observed. From 1999 to 2007, the average GPP_NIRv (GPP_FLUXCOM) decreased by 11% (6%) compared to the period of 1990 to 1998, and precipitation also decreased by 28% during the same period. This is consistent with previous studies that found a decreasing shift in summer precipitation in Northeastern and North China in the late 1990s^20,21.

It has been reported that the phase change of the Pacific decadal oscillation (PDO) can affect the regional precipitation in East Asia through atmospheric teleconnections^21–24. After the late 1990s, the Pacific entered a negative phase of the PDO (Fig. 2), and the upper-level atmospheric circulation was changed accordingly. The East Asian westerly jet stream (EAWJS) is weakened and poleward shifted. The poleward-shifted EAWJS changes the jet-related secondary meridional-vertical circulation²³, making an anomalous descending motion to the northern parts of East Asia. Subsequently, the anticyclonic circulation is dominant over the target region in a negative PDO phase, resulting in less precipitation in the late 1990s.

The GPP_FLUXCOM and GPP_NIRv anomalies are positively correlated with the precipitation anomaly in Eastern China-Mongolia (r = 0.92 and 0.63, P < 0.01). These results are consistent with a previous study that showed a positive relationship between decreasing precipitation and GPP over Northern China from 1999 to 2011¹⁷. On the contrary, temperature and solar radiation have relatively low correlation coefficients (r = − 0.45, P < 0.05 and r = − 0.31, P < 0.1) with GPP_FLUXCOM anomaly and no statistically significant correlation with GPP_NIRv anomaly at the 95% confidence level. These results suggest that the decreasing shift in precipitation associated with the PDO phase shift is probably responsible for that of GPP in the late 1990s. However, it should be noted that the GPP and precipitation show more dramatic changes than the PDO index, suggesting that other processes can be involved in the dramatic shift. For example, changes in the strength of the land-atmosphere coupling can lead to rapid changes in precipitation²⁵, which will be discussed in more detail in the Discussion section.

We further examine whether these decreasing shifts of GPP and precipitation are statistically significant based on the Lepage test. The phase shifts of GPP_FLUXCOM, GPP_NIRv, and precipitation around 1999 are all statistically significant (Fig. 3). The GPP_FLUXCOM and GPP_NIRv show the most significant abrupt shifts in 1999–2000 and 2000–2001, respectively (P < 0.05). Precipitation also shows the most significant abrupt shift in 1999–2000. This is consistent with the results of previous studies^20,21, which showed the abrupt change point of summer precipitation around 1999 over Northeastern China. These highest HK values are largely determined by the standardized Wilcoxon rank sum statistic, indicating that these statistically significant abrupt shifts mostly come from the large changes in the mean state. To test the robustness of the Lepage test, we vary the length of the moving windows, and similar results are found using window lengths of 8- to 11-years, which show the same significant abrupt shift year of GPP and precipitation (Supplementary Fig. 2).

Vegetation productivity is significantly regulated by temperature and CO₂ concentrations, as well as precipitation¹⁰. For example, the abrupt changes in NDVI in Southwest China were mainly driven by changes in temperature during 1982–2015²⁶. Therefore, to confirm whether precipitation is a major cause of the regime shift in GPP around 1999, we evaluate the relative contributions of changes in temperature, precipitation, and CO₂ concentrations to the GPP shift. We apply multiple linear regression to the GPP with respect to normalized climate factors (precipitation and temperature) and CO₂ concentrations. Based on the reconstructed GPP, we estimate the contributions of the climate factors and CO₂ to the GPP shift between period 1 (P1: 1990 to 1998) and period 2 (P2: 1999 to 2007). It is evident that there is a decrease in the reconstructed GPP_FLUXCOM (− 1.44) and GPP_NIRv (− 0.93) between P1 and P2, indicating a significant decrease in GPP during P2 due to climate factors and CO₂ (Fig. 4). Among them, it is clear that the decreased in GPP mostly due to the change in precipitation (GPP_FLUXCOM:−1.49, GPP_NIRv:−1.19). Temperature also contributes to the decrease, but its impact is minor (GPP_FLUXCOM:−0.045, GPP_NIRv:−0.064). Increasing CO₂ leads to an increase in GPP in P2 due to the fertilization effect^7,27. These results indicate that the change in precipitation is mainly responsible for the statistically significant regime shift in GPP around 1999. In summary, there are significant abrupt changes in GPP in Eastern China-Mongolia over the past three decades, which can be attributed to the abrupt decrease in precipitation.

The inter-model diversity of the abrupt shift in GPP over Eastern China-Mongolia GPP

To support the robustness of the regime shift in GPP, we further analyze the output of offline land surface models participating in TRENDY. Using the simulations provided by TRENDY, we isolate the time-varying GPP_{climate−forcing} and the GPP_{CO2−forcing} (details in Section 2.1). As shown in Fig. 5a, an abrupt shift in GPP around the year 2000 is well captured in the multi-model ensemble (MME) mean of the TRENDY models. This change is statistically significant based on the Lepage test (Fig. 5b). However, there is significant diversity and spread among the individual TRENDY models, especially in their HK values (Fig. 5b). Of the twelve models, only eight models show a significant abrupt shift in GPP in 2000–2001 (Strong shift group: CLASS-CTEM, CLM5.0, JULES-ES, JSBACH, ORCHIDEE-CNP, ORCHIDEE, SDGVM, and VISIT), while the other models do not simulate abrupt shifts (Weak shift group: LPJ-GUESS, CABLE-POP, DLEM, and ISBA-CTRIP).

We further investigate the differences in GPP between the strong and weak shift groups (Fig. 5a). The red and blue horizontal bars show the mean values of GPP for the two groups during period 1 (P1': 1991 to 1999) and period 2 (P2': 2000 to 2008). Although the two groups show quite similar interannual variability, their decadal variability is different. The weak shift group shows a smaller mean state change (5.64 TgC month^− 1) in GPP between the two periods than that of the strong shift group (17.4 TgC month^− 1).

To examine the driver of the abrupt shift in GPP and the cause of the differences between the groups, we quantify the contributions of the climate and CO₂ forcings to the mean state change in GPP from the TRENDY models. We calculated the differences in normalized GPP separated by each forcing by subtracting the mean of P1' from P2' (Fig. 6). In all the models, GPP decreases from P1' to P2' and is mostly driven by the GPP_{climate−forcing} (Fig. 6a). The decrease in GPP_{climate−forcing} is mostly driven by changes in precipitation rather than temperature and solar radiation (Supplementary Fig. 3). Additionally, in terms of GPP_{climate−forcing}, the TRENDY models also present the aforementioned precipitation-induced regime shift in GPP, which is consistent with satellite and reanalysis data.

The effect of CO₂ forcing shows the opposite of climate forcing, but the impact is relatively small: the MME mean of GPP_{CO2−forcing} differences (0.26) is smaller in magnitude compared to that of GPP_{climate−forcing} (− 1.40). In particular, the four models in the weak shift group (LPJ-GUESS, CABLE-POP, DLEM, and ISBA-CTRIP) have a higher sensitivity of GPP to CO₂ and a lower sensitivity to climate compared to the other group (Fig. 6a). These characteristics are more obvious in the group means (Fig. 6b): the MME means of the GPP_{CO2−forcing} and GPP_{climate−forcing} differences for the weak shift group (the strong shift group) are 0.42 (0.18) and − 0.98 (-1.60), respectively. The differences in the sensitivity of GPP to CO₂ and climate forcing between the models lead to significant differences in the changes in GPP between the two groups.

The characteristics of abrupt changes in GPP in Eastern China-Mongolia

We have investigated the first EOF mode of GPP variability and found significant abrupt changes in GPP in East Asia over the past three decades. The PCs of the first leading modes of GPP_FLUXCOM and GPP_NIRv consistently show a distinctive phase shift around the year 2000 in Eastern China-Mongolia. The high correlation between GPP and water availability and the abrupt decrease in precipitation suggests that the abrupt decrease in GPP resulted from the changes in precipitation. Because the precipitation shift is known to be related to the phase shift of the PDO, our results suggest that the occurrence of abrupt changes in GPP in the region depends on the phase change of the PDO. The abrupt change around the year 2000 is consistently simulated by satellite and reanalysis GPP, while some offline land surface models fail to capture it. Our results suggest that the inter-model diversity in the TRENDY models can be attributed to the diverse sensitivities of CO₂ and climate forcing, which could be constrained by observational data. Improving the sensitivity in each model would be helpful to reduce uncertainties in the TRENDY models and to investigate the more accurate variability of GPP.

It is interesting that GPP and precipitation show more dramatic changes than the PDO index, suggesting that other factors may play a role in accelerating the changes. It is possible that land-atmosphere coupling is involved in the abrupt change in precipitation over Eastern China-Mongolia. For example, D’Odorico et al.²⁵ showed that there is a positive feedback between vegetation and precipitation recycling. Decreased precipitation in water-limited regions causes a decline in vegetation, which leads to decreased evapotranspiration. This further reduces precipitation and exacerbates the loss of vegetation, creating positive feedback. In addition, the response of vegetation to environmental change is non-linear¹⁰. Therefore, the positive feedback and non-linear response of the vegetation could lead to dramatic changes in the terrestrial carbon cycle, and changes that exceed a certain threshold could possibly lead to catastrophic outcomes. Further studies are needed to understand the role of land-atmosphere coupling in abrupt changes in vegetation productivity.

The necessity for further exploring abrupt changes in semi-arid vegetation productivity

The results of our study suggest the need for further investigation in semi-arid regions. Firstly, based on a previous study investigating the prevalence of abrupt vegetation shifts in global drylands¹³, the aforementioned relationship between abrupt changes in GPP and precipitation could be extended to other semi-arid regions. Secondly, projections based on the representative concentration pathways (RCPs) RCP8.5 and RCP4.5 indicate that Northeastern China is one of the regions that will experience the most significant expansion of drylands by the end of the twenty-first century compared to the 1961–1990 baseline²⁸. As a result, there is a possibility of more frequent abrupt changes in vegetation productivity in East Asia in the future. Lastly, a previous study indicates that the interannual variability of the land CO₂ sink is primarily influenced by semi-arid ecosystems²⁹. Therefore, the intense occurrence of abrupt changes in semi-arid ecosystems over East Asia could exacerbate the interannual variation of the land sink, posing a challenge to the development of a carbon-neutral strategy. In this regard, it is essential to conduct further research on the abrupt changes and their mechanisms of global vegetation productivity in semi-arid regions in the future.

Terrestrial photosynthesis is a highly uncertain process due to the lack of direct observations of GPP on a global scale. To obtain robust results on GPP responses to climate variability, we used both data-driven and process-based GPP datasets with over 30 years of records for the East Asian region.

FLUXCOM

FLUXCOM provides an ensemble of upscaled GPP products at a global scale (www.fluxcom.org) for the period of 1980 to 2013. FLUXCOM combines FLUXNET site-level observations, satellite remote sensing, and meteorological data based on three machine learning techniques: Random Forest (RF), Artificial Neural Network (ANN), and Multivariate Adaptive Regression Splines (MARS) to scale up these fluxes to the global scale^30,31. In this study, we use the ensemble mean of GPP datasets generated by the three machine learning techniques.

Near-infrared reflectance (NIRv)

We used NIRv as a proxy for GPP. Based on the close relationship between NIRv and solar-induced chlorophyll fluorescence (SIF), NIRv is a reliable indicator of changes in GPP^32,33. The NIRv data are derived from the observational datasets of the Advanced Very High Resolution Radiometer (AVHRR) sensors and flux stations (https://data.tpdc.ac.cn) for the period of 1982 to 2018. The NIRv is calculated as a function of monthly NDVI and near-infrared reflection of the total pixel (NIRT)^32,33 (Eq. (1)).

$$\text{N}\text{I}\text{R}\text{v}=(\text{N}\text{D}\text{V}\text{I}-0.08)\times \text{N}\text{I}\text{R}\text{T}$$

TRENDY v8

We analyze the output from 12 Dynamic Global Vegetation Models (DGVMs) that are part of a recent model intercomparison project, TRENDY v8 and the Global Carbon Budget 2019 (Table 1)³⁴. These models are forced with the same meteorological conditions, atmospheric CO₂ concentrations, and land use datasets^35–37.

Table 1

Summary of the TRENDYv8 models used in this study.
Model	Spatial resolution	References
CABLE-POP	$1^\circ \times 1^\circ$	Haverd et al., 2018
CLASS-CTEM	$2.8125^\circ \times 2.8125^\circ$	Melton and Arora, 2016
CLM5.0	$1.875^\circ \times 0.625^\circ$	Lawrence et al., 2019
DLEM	$0.5^\circ \times 0.5^\circ$	Tian et al., 2015
ISBA-CTRIP	$1^\circ \times 1.2^\circ$	Decharme et al., 2019
JSBACH	$1.875^\circ \times 1.875^\circ$	Mauritsen et al., 2019
JULES-ES	$1.875^\circ \times 1.25^\circ$	Sellar et al., 2019
LPJ-GUESS	$0.5^\circ \times 0.5^\circ$	Smith et al., 2014
ORCHIDEE	$0.5^\circ \times 0.5^\circ$	Krinner et al., 2005
ORCHIDEE-CNP	$2^\circ \times 2^\circ$	Goll et al., 2017
SDGVM	$0.5^\circ \times 0.5^\circ$	Walker et al., 2017
VISIT	$0.5^\circ \times 0.5^\circ$	Kato et al., 2013

TRENDYv8 provides a suite of simulations that allow the response of land to climate, CO₂, and land use forcing to be separated. The “S0” simulation is a baseline simulation with time-invariant pre-industrial CO₂ concentrations, climate, and land use. The “S1” simulation is forced by time-varying atmospheric CO₂ concentrations, but with fixed climate and land-use information. The “S2” simulation is forced by time-varying atmospheric CO₂ concentrations and climate with constant land-use change. Based on the experimental designs, we utilize the experimental simulations to evaluate the relative contributions of CO₂ fertilization and climate forcing to the change in GPP. For each TRENDY model, the CO₂ fertilization effect-driven change in GPP (GPP_{CO2−forcing}) is calculated from the differences between S1 and S0, while the climate change-driven change in GPP (GPP_{climate−forcing}) is calculated from the difference between S2 and S1. In addition, the response of GPP to total forcing (GPP_TRENDY) is calculated as the difference between S2 and S0, which is the sum of CO₂ fertilization and climate change-driven GPP. Note that this study excludes the effects of land use change on GPP change, as land use change does not appear to be a significant driver of GPP change over the study period (Supplementary Fig. 1). This study utilizes the TRENDY GPP dataset spanning from 1980 to 2018, which was regridded to a spatial resolution of 1° × 1° for consistent analysis in the multi-model domain.

We focused on averaged GPP in the boreal summer months of June–July–August (JJA), the season with the largest portion of carbon sequestration amount, to examine the spatio-temporal variations across the East Asia region (24°–52°N and 100°–149°E).

Meteorological data

We used precipitation, temperature, and solar radiation from CRUNCEPv7^38,39. Atmospheric CO₂ concentrations from the NOAA GML Carbon Cycle Cooperative Global Air Sampling Network were also used^38,39.

Lepage test for abrupt change detection

There are several methods available for detecting change points in adjacent data on a decadal scale. Among these methods, the Lepage test⁴⁰ has few underlying assumptions, which makes it adopted for detecting various types of climate changes, including linear trends, cyclical variations, step-like changes, and discontinuous changes²⁰. This method has been extensively used in numerous studies to detect abrupt changes^20,21,41,42. The Lepage statistic (HK) is calculated as the sum of the squares of the standardized Wilcoxon’s and Ansari-Bradley’s statistics, which represent the comparison of the median and variability between two samples, respectively⁴⁰. If the HK is greater than 5.99, the difference between the two samples is significant at the 95% confidence level.

Data availability

All data used in this study are publicly available and can be downloaded from the corresponding websites (FLUXCOM: https://fluxcom.org/; NIRv: https://data.tpdc.ac.cn/; Access to TRENDY v.8 data can be obtained by contacting Stephen Sitch ([email protected]); Meteorological datasets (precipitation, temperature, and solar radiation): https://rda.ucar.edu/datasets/ds314.3/citation/; CO₂ concentration: https://gml.noaa.gov/).

Code availability

The computer codes that support the analysis within this paper are available from the corresponding author on request.

Acknowledgement

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (NRF-2022R1A3B1077622). Jin-Soo Kim was supported by Low-Carbon and Climate Impact Research Centre at the School of Energy and Environment, City University of Hong Kong and CityU Start-up Grant for New Faculty (No. 9610581).

Author contributions

D.-B.L compiled the data, conducted analyses, prepared the figures, and wrote the manuscript. J.-S.K, S.-W.P, and J.-S.K designed the research and wrote the majority of the manuscript content. All of the authors discussed the study results and reviewed the manuscript.

Competing interests

The authors declare no competing interests.

Friedlingstein, P. et al. Global Carbon Budget 2022. Earth Syst. Sci. Data 14, 4811-4900 (2022).
Zhao, H., Jia, G., Wang, H., Zhang, A. & Xu, X. Seasonal and interannual variations in carbon fluxes in East Asia semi-arid grasslands. Science of The Total Environment 668, 1128-1138 (2019).
Kim, J.-S., Kug, J.-S., Yoon, J.-H. & Jeong, S.-J. Increased Atmospheric CO₂ Growth Rate during El Niño Driven by Reduced Terrestrial Productivity in the CMIP5 ESMs. Journal of Climate 29, 8783-8805 (2016).
Park, S.-W. et al. Two Aspects of Decadal ENSO Variability Modulating the Long-Term Global Carbon Cycle. Geophysical Research Letters 47, e2019GL086390 (2020).
He, P. et al. Spatiotemporal evolutionary and mechanism analysis of grassland GPP in China. Ecological Indicators 143, 109323 (2022).
Ma, J. et al. Trends and controls of terrestrial gross primary productivity of China during 2000–2016. Environmental Research Letters 14, 084032 (2019).
Piao, S. L. et al. The carbon budget of terrestrial ecosystems in East Asia over the last two decades. Biogeosciences 9, 3571-3586 (2012).
Park, S.-W., Kug, J.-S., Jun, S.-Y., Jeong, S.-J. & Kim, J.-S. Role of Cloud Feedback in Continental Warming Response to CO₂ Physiological Forcing. Journal of Climate 34, 8813-8828 (2021).
Sakalli, A., Cescatti, A., Dosio, A. & Gücel, M. U. Impacts of 2°C global warming on primary production and soil carbon storage capacity at pan-European level. Climate Services 7, 64-77 (2017).
Liu, Z. et al. Global divergent responses of primary productivity to water, energy, and CO₂. Environmental Research Letters 14, 124044 (2019).
Wan, L., Liu, H., Gong, H. & Ren, Y. Effects of Climate and Land Use changes on Vegetation Dynamics in the Yangtze River Delta, China Based on Abrupt Change Analysis. Sustainability 12, 1955 (2020).
Ma, X., Huete, A., Moran, S., Ponce-Campos, G. & Eamus, D. Abrupt shifts in phenology and vegetation productivity under climate extremes. Journal of Geophysical Research: Biogeosciences 120, 2036-2052 (2015).
Berdugo, M., Gaitán, J. J., Delgado-Baquerizo, M., Crowther, T. W. & Dakos, V. Prevalence and drivers of abrupt vegetation shifts in global drylands. Proceedings of the National Academy of Sciences 119, e2123393119 (2022).
Piao, S. et al. Contribution of climate change and rising CO₂ to terrestrial carbon balance in East Asia: A multi-model analysis. Global and Planetary Change 75, 133-142 (2011).
Kim, J.-S. et al. Arctic warming-induced cold damage to East Asian terrestrial ecosystems. Communications Earth & Environment 3, 16 (2022).
Zhang, P. et al. Abrupt shift to hotter and drier climate over inner East Asia beyond the tipping point. Science 370, 1095-1099 (2020).
Yuan, W. et al. Multiyear precipitation reduction strongly decreases carbon uptake over northern China. Journal of Geophysical Research: Biogeosciences 119, 881-896 (2014).
Kang, X. et al. Modeling impacts of climate change on carbon dynamics in a steppe ecosystem in Inner Mongolia, China. Journal of Soils and Sediments 11, 562-576 (2011).
Xu, H.-j., Wang, X.-p., Zhao, C.-y. & Yang, X.-m. Diverse responses of vegetation growth to meteorological drought across climate zones and land biomes in northern China from 1981 to 2014. Agricultural and Forest Meteorology 262, 1-13 (2018).
Liu, Y., Huang, G. & Huang, R. Inter-decadal variability of summer rainfall in Eastern China detected by the Lepage test. Theoretical and applied climatology 106, 481-488 (2011).
Xu, Z., Fan, K. & Wang, H. Decadal variation of summer precipitation over China and associated atmospheric circulation after the late 1990s. Journal of Climate 28, 4086-4106 (2015).
Qian, C. & Zhou, T. Multidecadal Variability of North China Aridity and Its Relationship to PDO during 1900–2010. Journal of Climate 27, 1210-1222 (2014).
Zhu, Y., Wang, H., Ma, J., Wang, T. & Sun, J. Contribution of the phase transition of Pacific Decadal Oscillation to the late 1990s' shift in East China summer rainfall. Journal of Geophysical Research: Atmospheres 120, 8817-8827 (2015).
Piao, J., Chen, W. & Chen, S. Water vapour transport changes associated with the interdecadal decrease in the summer rainfall over Northeast Asia around the late-1990s. International Journal of Climatology 41, E1469-E1482 (2021).
D’Odorico, P., Bhattachan, A., Davis, K. F., Ravi, S. & Runyan, C. W. Global desertification: Drivers and feedbacks. Advances in Water Resources 51, 326-344 (2013).
Xu, X., Liu, H., Lin, Z., Jiao, F. & Gong, H. Relationship of Abrupt Vegetation Change to Climate Change and Ecological Engineering with Multi-Timescale Analysis in the Karst Region, Southwest China. Remote Sensing 11, 1564 (2019).
Park, S.-W., Kim, J.-S. & Kug, J.-S. The intensification of Arctic warming as a result of CO₂ physiological forcing. Nature Communications 11, 2098 (2020).
Huang, J., Yu, H., Guan, X., Wang, G. & Guo, R. Accelerated dryland expansion under climate change. Nature Climate Change 6, 166-171 (2016).
Ahlström, A. et al. The dominant role of semi-arid ecosystems in the trend and variability of the land CO₂ sink. Science 348, 895-899 (2015).
Jung, M. et al. The FLUXCOM ensemble of global land-atmosphere energy fluxes. Scientific Data 6, 74 (2019).
Deng, M. et al. The Response of Vegetation to Regional Climate Change on the Tibetan Plateau Based on Remote Sensing Products and the Dynamic Global Vegetation Model. Remote Sensing 14, 3337 (2022).
Badgley, G., Field, C. B. & Berry, J. A. Canopy near-infrared reflectance and terrestrial photosynthesis. Science Advances 3, e1602244 (2017).
Yang, R. et al. Divergent historical GPP trends among state-of-the-art multi-model simulations and satellite-based products. Earth Syst. Dynam. 13, 833-849 (2022).
Friedlingstein, P. et al. Global carbon budget 2019. Earth System Science Data 11, 1783-1838 (2019).
Sitch, S. et al. Recent trends and drivers of regional sources and sinks of carbon dioxide. Biogeosciences 12, 653-679 (2015).
Wang, S., Zhang, Y., Ju, W., Qiu, B. & Zhang, Z. Tracking the seasonal and inter-annual variations of global gross primary production during last four decades using satellite near-infrared reflectance data. Science of the Total Environment 755, 142569 (2021).
O’Sullivan, M. et al. Process-oriented analysis of dominant sources of uncertainty in the land carbon sink. Nature Communications 13, 4781 (2022).
Viovy, N. CRUNCEP Version 7 - Atmospheric Forcing Data for the Community Land Model. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory, Boulder, CO (2018). Accessed 23 Aug 2021.
Dlugokencky et al. Atmospheric Carbon Dioxide Dry Air Mole Fractions from the NOAA GML Carbon Cycle Cooperative Global Air Sampling Network, 1968-2020. In (2021-07-30 ed.).
Lepage, Y. A combination of Wilcoxon's and Ansari-Bradley's statistics. Biometrika 58, 213-217 (1971).
Yonetani, T. Detection of long term trend, cyclic variation and step-like change by the Lepage test. Journal of the Meteorological Society of Japan. Ser. II 71, 415-418 (1993).
Kwon, M., Jhun, J.-G., Wang, B., An, S.-I. & Kug, J.-S. Decadal change in relationship between east Asian and WNP summer monsoons. Geophysical Research Letters 32 (2005).
Mauritsen, T. et al. Developments in the MPI-M Earth System Model version 1.2 (MPI-ESM1.2) and Its Response to Increasing CO₂. Journal of Advances in Modeling Earth Systems 11, 998-1038 (2019).

There is NO Competing Interest.

SICEE.docx
Supplementary Information

Download PDF

Version 1

posted

You are reading this latest preprint version

An abrupt shift in gross primary productivity over Eastern China-Mongolia

Status:

Version 1

Abstract

Figures

Introduction

Results

The abrupt shift in GPP over Eastern China-Mongolia

The inter-model diversity of the abrupt shift in GPP over Eastern China-Mongolia GPP

Discussion

The characteristics of abrupt changes in GPP in Eastern China-Mongolia

The necessity for further exploring abrupt changes in semi-arid vegetation productivity

Data and Methods

FLUXCOM

Near-infrared reflectance (NIRv)

TRENDY v8

Meteorological data

Lepage test for abrupt change detection

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1

Model	Spatial resolution	References
CABLE-POP	\(1^\circ \times 1^\circ\)	Haverd et al., 2018
CLASS-CTEM	\(2.8125^\circ \times 2.8125^\circ\)	Melton and Arora, 2016
CLM5.0	\(1.875^\circ \times 0.625^\circ\)	Lawrence et al., 2019
DLEM	\(0.5^\circ \times 0.5^\circ\)	Tian et al., 2015
ISBA-CTRIP	\(1^\circ \times 1.2^\circ\)	Decharme et al., 2019
JSBACH	\(1.875^\circ \times 1.875^\circ\)	Mauritsen et al., 2019
JULES-ES	\(1.875^\circ \times 1.25^\circ\)	Sellar et al., 2019
LPJ-GUESS	\(0.5^\circ \times 0.5^\circ\)	Smith et al., 2014
ORCHIDEE	\(0.5^\circ \times 0.5^\circ\)	Krinner et al., 2005
ORCHIDEE-CNP	\(2^\circ \times 2^\circ\)	Goll et al., 2017
SDGVM	\(0.5^\circ \times 0.5^\circ\)	Walker et al., 2017
VISIT	\(0.5^\circ \times 0.5^\circ\)	Kato et al., 2013