Climatic warming strengthening water constraint on seasonal peak photosynthesis in northern ecosystems

doi:10.21203/rs.3.rs-3577873/v1

Download PDF

Article

Climatic warming strengthening water constraint on seasonal peak photosynthesis in northern ecosystems

https://doi.org/10.21203/rs.3.rs-3577873/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The seasonal peak of vegetation photosynthesis is a key indicator of terrestrial ecosystem productivity, bearing significant impacts on atmospheric CO₂ concentrations. Recent remote sensing observations have indicated an escalating trend in peak photosynthesis, yet the sustained trajectory of this increase under warming remains uncertain. Notably, water availability and temperature emerge as paramount limiting factors governing vegetation photosynthesis. To probe these constraints, here we employ solar-induced chlorophyll fluorescence as a surrogate for photosynthesis, focusing on the examination of temperature and water-related restrictions on peak photosynthesis for northern ecosystems. Our findings unveil a discernible expansion of water-constrained zones in northern ecosystems over the past two decades. A general relationship is identified between mean annual temperature and precipitation that effectively distinguishes temperature-constrained and water-constrained ecosystems, which can be used to predict future changes in the constraint status in various climate change scenarios. Our results show that water-constrained regions will further expand by 4.51% and 11.13% by 2100 under the SSP245 and SSP585 scenarios.

Earth and environmental sciences/Ecology

Earth and environmental sciences/Ecology/Ecosystem ecology

Vegetation is a key component of the Earth system, absorbing approximately one-third of anthropogenic CO₂ emissions^1,2. Global climate change has substantially extended growing-season length (GSL) of vegetation and amplified seasonal peak photosynthesis (GPP_max) in rencent decades^3–6, increasing terrestrial carbon uptake. Prevailing evidence suggests that warming alleviates climate constraints on vegetative photosynthetic activity in the northern regions^7,8. However, recent studies also indicate that negative effects of warming on ecosystem carbon sequestration are emerging, casting uncertainty on the sustainability of the observed increase in terrestrial carbon uptake over the past few decades^9,10. Interannual variations of gross primary productivity (GPP) are tightly related to GSL and GPP_max but GPP_max accounts for more than does GSL^11,12. While a continuous upswing in peak vegetation growth remains ecologically tenable¹³, this positive trajectory may not be perpetual, owing to shifts in resource availability. Consequently, achieving a comprehensive understanding of the mechanisms governing peak vegetation growth assumes paramount importance in forecasting vegetative responses under forthcoming climate scenarios.

The magnitude of vegetative photosynthesis is intricately linked to the synchronization of seasonal vegetation growth with resource availability¹⁴. An optimal photosynthetic capacity is postulated to be achieved when the timing of peak photosynthetic activity (DOY_GPP) aligns with the timing of seasonal peak of radiation (DOY_PAR)¹⁵. However, DOY_GPP may shift from DOY_PAR due to constraints imposed by water availability and temperature^16–18. Following the law of minimum¹⁹, DOY_GPP would adjust towards those time intervals when the limiting resource, whether it be temperature or water, is more abundant.

Here, we used a recent machine-learning-derived solar-induced chlorophyll fluorescence (SIF) dataset, specifically the Reconstructed TROPOMI SIF (RTSIF)²⁰, as a proxy for vegetation photosynthesis. The DOY_RTSIF retrieved from RTSIF showed very high consistency with the DOY_GPP retrieved from the GPP estimates from eddy covariance flux towers (Fig. S1). We investigated the synchrony of DOY_RTSIF and DOY_PAR and used their differences (ΔDOY_RTSIF,PAR) as representatives of temperature and water constraints to assess climatic constraints in northern ecosystems (> 30^oN)⁵. Our findings unveiled a notable expansion of regions constrained by limited water resources during the past two decades. We further established a general relationship predicated on mean annual temperature and precipitation, which effectively demarcated regions affected by either water or temperature constraints. Leveraging this analytical framework, we projected forthcoming alterations in water and temperature constraints and their potential repercussions on vegetation growth under various carbon emission scenarios.

Seasonal peak photosynthesis and potential climatic constraints

We first estimate the seasonal peak vegetation photosynthesis (RTSIF_max) and its corresponding timing (DOY_RTSIF) using the RTSIF dataset based on a weighted spline fitting algorithm with a fixed per-pixel threshold (see Methods section). Subsequently, we assess the synchronization between DOY_RTSIF and the peak radiation timing (DOY_PAR) within northern ecosystems, spanning latitudes greater than 30°N, during the years 2001 to 2020, to discern the prevailing constraints imposed by temperature and water availability in these ecosystems (Methods, Extended Fig. 1). We find that the majority of northern ecosystems, comprising over 69.38% of the total, experience constraints primarily due to temperature, while the remainder are predominantly restricted by water availability (Fig. 1a). The temperature-constrained regions predominantly encompass the Arctic and cold boreal regions, while water-constrained regions are mainly located in the northern forests of North America and certain temperate regions. Similar spatial patterns of ΔDOY_RTSIF,PAR are obtained using an independent dataset from the Global Ozone Monitoring Experiment-2 (GOME-2) SIF (Fig. S2). It is important to emphasize that the magnitude of seasonal peak photosynthesis is closely intertwined with the climatic constraints governing vegetation growth, wherein substantial RTSIF_max values are typically observed in close proximity to ΔDOY_RTSIF,PAR equating to zero. Ecosystems operating under either temperature constraints (ΔDOY_RTSIF,PAR > 0) or water constraints (ΔDOY_RTSIF,PAR < 0) exhibit diminished photosynthetic capacity, reflecting limitations in resource availability (Fig. S3).

Changes in DOY_RTSIF have regionally varying impacts on RTSIF_max (reflected by the opposite signs of the correlation coefficient between RTSIF_max and DOY_RTSIF) which are closely related to dominant climatic constraints (Fig. 1b). Similar patterns can also be found in the partial correlations between RTSIF_max and DOY_RTSIF (Fig. S4). An "early peak – larger peak" pattern occurs mainly in temperature-constrained areas because more favorable thermal conditions enable vegetation to enhance its synchronization with radiation, consequently augmenting plant growth²¹. Conversely, in water-constrained regions, an "early peak-smaller peak" pattern prevails. This phenomenon can be mainly attributed to summer water deficits triggered by warming-induced water constraints, which impede peak vegetation growth^22,23. This is consistent with the opposite trends observed between RTSIF_max and ΔDOY_RTSIF,PAR for ΔDOY_RTSIF,PAR<0 and ΔDOY_RTSIF,PAR>0 (Fig. S3).

Changes in water and temperature constraints in the past 20 years

A widespread shift in DOY_RTSIF toward earlier in the growing season is found in 50.58% of the northern vegetated area during 2001–2020 (Fig. 2a). With a relatively stable DOY_PAR (Table S2), a significant decrease of ΔDOY_RTSIF,PAR is detected (-0.13 days/year, p < 0.05). However, distinct trends materialize in various climatic constrained regions: a pronounced decline is noted in temperature-constrained areas (-0.08 days/year, p < 0.1), whereas water-constrained regions exhibit no statistically significant temporal trend (Fig. S5). Over the past two decades, certain boreal ecosystems have undergone a transition from positive to negative ΔDOY_RTSIF,PAR, indicating a transition from temperature-constrained to water constrained status^24,25 (Fig. S6). These alterations underscore the influence of climate change, which has notably diminished the relative significance of thermal constraints in boreal vegetated regions and concurrently accentuated the role of water availability in temperate regions^26–30.

SIF_max increased during the past 20 years in both temperature and water-constrained regions. However, it is noteworthy that temperature-constrained regions exhibit a steeper positive slope (Fig. 2b), with a rate of change of 4.003*10^− 4 mWm^− 2um^− 1sr^− 1/yr, compared to 1.529*10^− 4 mWm^− 2um^− 1sr^− 1/yr for water-constrained regions, as determined by a two-tailed t-test (p < 0.1). These findings align with previous observations employing different vegetation indices^31–35. Given the expected ongoing warming and drying trends^36,37, the continuous monitoring of ecosystems situated in water-constrained regions assumes paramount importance, as these trends could amplify water stress and, consequently, impact peak photosynthesis in these ecosystems^38–40.

To appraise the capacity of contemporary terrestrial ecosystem models in capturing the ramifications of seasonal climatic constraints on vegetation's seasonal peak photosynthesis, we further conducted an assessment of the performance of 13 dynamic global vegetation models (DGVMs) that are participants in the "Trends and drivers of the regional scale sources and sinks of carbon dioxide" (TRENDY v.8)⁴¹. In congruence with empirical observations, it was found that modelled GPP_max exhibited an increasing trend in both temperature-constrained and water-constrained regions, corroborating the enhanced peak photosynthesis in northern ecosystems (Fig S7). Notably, no significant difference in the upward trends of modelled GPP_max was observed between temperature-constrained and water-constrained regions (two-tailed t-test, p = 0.79). Compared with observations, most models accurately captured the negative correlation between GPP_max and DOY_GPP in temperature-constrained regions (Fig. 2c). However, four of the models (DLEM, SDGVM, CLM5.0, and ORCHIDEE) failed to reproduce the positive correlation between GPP_max and DOY_GPP in water-constrained regions. This divergence in model performance implies that current terrestrial carbon cycle models may underestimate the adverse consequences of warming⁴².

A general relationship between precipitation and temperature in separating water and temperature limited zones

Using ΔDOY_RTSIF,PAR as an indicator of the combined temperature and water constraints affecting ecosystem photosynthesis, we elucidate the interplay of mean annual precipitation (MAP) and temperature (MAT) on constraint status, as illustrated in Fig. 3a. The temperature and water constrained regions can be effectively separated using a support vector machine (SVM, see Methods section)⁴³. Notably, the solid black line in Fig. 3a indicates the non-constrained status (ΔDOY_RTSIF,PAR=0) where the relationship between mean annual temperature (^oC) and precipitation (mm) is

$$\begin{array}{c}MAT=0.0138\times MAP+3.7687\#\left(1\right)\end{array}$$

The fidelity of this relationship is substantiated by calculated ΔDOY_GPP,PAR and DOY_PAR derived from Fluxnet site observations, further corroborating the effectiveness of the proposed MAT and MAP relationship in distinguishing water and temperature-constrained ecosystems (Fig. S8). Using this simple relationship, the predicted water- and temperature-constrained regions are almost identical to those estimated using ΔDOY_RTSIF,PAR (Fig. S9). It is noteworthy that the majority of vegetation types currently persist within temperature-constrained regions, except for CSH, DBF and EBF (Fig. 3b). Over the past two decades, precipitation trends have been positive for half of the vegetation types (indicated by green arrows in Fig. 3b) and negative for the remaining half (denoted by red arrows). Notably, for vegetation types situated below the optimal line defined by Eq. (1), changes in temperature and precipitation over the last two decades appear to be converging toward the optimal line. Conversely, for CSH, DBF, and EBF, it is evident that vegetation growth is increasingly constrained by water availability.

Future changes of water and temperature constraints

Equation 1 provides a straightforward approach with robustness through time for evaluating potential alterations in the constraint status of vegetation photosynthesis in forthcoming scenarios (Table S3). To examine prospective changes in the spatial distribution of temperature and water-constrained regions by the end of the 21st century (2100), we have incorporated simulated temperature and precipitation data from 13 climate models participating in phase 6 of the Coupled Model Intercomparison Project (CMIP6) under representative scenarios SSP245 and SSP585 (Fig. 4a, 4b). Our results indicate that the reduction in temperature-constrained regions is anticipated to persist as a consequence of ongoing warming, indicating a continued amelioration in the temperature constraint on vegetation photosynthetic activity. This aligns with the conclusions drawn in several recent studies regarding the reduction of temperature-limitation on vegetation growth^8,44. Conversely, the shift in water-constrained regions predominantly occurs in regions such as Central and Eastern North America, the Mediterranean, Western and Southern Asia, and East Asia. Under SSP245 and SSP585, we estimate that the water-constrained regions will expand by 4.51% and 11.13%, respectively, by 2100 (Fig. 4c).

In future scenarios, MAT is projected to rise in most terrestrial areas, contributing to an expanded occurrence of water-constrained regions. To delineate the respective influences of MAT and MAP, we have maintained either MAT or MAP at the average levels observed during the current period (2016 to 2020). This allows for the quantification of alterations in water-constrained regions attributed to variations in the other variable. As expected, the combined effects of warming and diminished precipitation play pivotal roles in the transition of regions from temperature-constrained to water-constrained status. Notably, these shifts are predominantly influenced by substantial increases in MAT, which exert a more prominent effect than changes in MAP (Fig. S10).

In summary, our study has introduced a pragmatic approach for identifying the influence of temperature and water constraints on vegetation growth, hinging on the timing of peak RTSIF and peak PAR. The analysis of ΔDOY_GPP,PAR over the last two decades underscores an expansion of regions with limited water resources within northern ecosystems. Furthermore, leveraging an established relationship between MAT and MAP, capable of distinguishing between conditions of water and temperature constraints, we project the ongoing expansion of water-constrained regions throughout this century, under varying carbon emission scenarios. This evolution highlights a potential underestimation of the role of water availability constraints on vegetation photosynthesis within existing terrestrial ecosystem models, thereby underscoring the need for further research to incorporate these effects into model representations.

Datasets and study area

We used the RTSIF dataset with 8-day temporal and 0.05° spatial resolutions to derive the annual peak timing of vegetation photosynthesis (DOY_RTSIF) from 2001 to 2020 in northern ecosystems (> 30° N). The RTSIF dataset uses surface reflectance from the MODIS Collection 6 (C6, MCD43C1), climate information and landcover types as inputs and trains machine-learning algorithms on daily SIF observations from TROPOMI²⁰. It was demonstrated to capture well the seasonal dynamics of satellite-observed SIF, showed high consistency with ecosystem GPP and therefore was suitable for vegetation phenology retrievals as a proxy for GPP^45,46. To test the robustness of peak timing of vegetation photosynthesis derived from RTSIF, we also used an independent SIF dataset from the GOME-2 dataset.

Our analysis used photosynthetically active radiation (PAR) to evaluate climatic constraints on vegetation photosynthesis in northern ecosystems. PAR was obtained from the Clouds and Earth’s Radiant Energy System (CERES) with 1-day temporal and 1° spatial resolutions⁴⁷. The surface air temperature (TEM) and precipitation (PRE) dataset were provided by the Global Land Data Assimilation System (GLDAS v.5) with 3-hour temporal and 0.25° spatial resolutions⁴⁸.

We used the FLUXNET2015 Tier 1 dataset to derive GPP_max and DOY_GPP and analyze temperature and water limitations⁴⁹. Daily GPP estimates from the night-time partitioning method were used for the analysis⁵⁰. We applied rigorous data quality checks and site selection to eliminate artefacts due to data quality and human interference. We first selected site years with valid observations during the growing season (less than 50% gap-filled observations for each month between April and October). We then removed all sites of cropland and intensively managed grasslands since their phenology and carbon balance were disturbed by human interference⁵¹. In addition, carbon fluxes in wetlands are influenced by upstream organic matter and nutrient inputs and fluctuating groundwater levels, and these sites were also removed⁵². Mean annual temperature (MAT) and mean annual precipitation (MAP) are calculated from the measurements on site. To be consistent with the RTSIF analysis, we calculated ΔDOY_GPP,PAR using the PAR retrieval from CERES, since not all stations have radiation observations of good quality.

The vegetation map was derived from MODIS C6 (MCD12Q1) with the International Geosphere-Biosphere Programme (IGBP) classification scheme⁵³. We only considered natural vegetated areas > 30° N.

Retrieval of peak timing

The weighted harmonic analysis of time series-maximum rate method (wHANTS-MR) was used to extract phenology from RTSIF⁵⁴. This method first fits the normalized yearly observations with weights into a series of harmonic functions with different frequencies, then fits the model to predict the RTSIF for each day of the year. The predicted largest RTSIF is RTSIF_max, and the corresponding time is DOY_RTSIF. This method has been widely used and proved to be an effective tool for capturing large-scale vegetation phenology^55,56. We assigned each observation a weight value that represents the confidence in data quality, so that the fitted smoothed curves are more dependent on high-quality observations. For RTSIF, since initial quality check and gap filling were applied to the reflectance before the dataset production, we only checked the RTSIF values using two criteria: (1) the RTSIF values smaller than zero were replaced with zero, and the corresponding weights were set to 0.5; and (2) for each year, observations outside the 3σ limits were linearly interpolated, and the corresponding weights were set to 0.2 (the mean seasonal average and standard deviation were calculated using 20 years of RTSIF between 2001 and 2020). During model optimization, we used the weighted root mean square error as the cost function, so the observations with smaller weights have a smaller contribution to the model fitness.

Definition of climatic constraints

We investigated the impacts of climatic constraints on vegetation photosynthesis in northern ecosystems on the basis of the framework proposed by ref⁵. This framework is based on two fundamental principles. First, maximum vegetation photosynthesis and radiation occur at the same day of year when there are no climatic limitations, i.e. DOY_RTSIF=DOY_PAR. Second, DOY_RTSIF will tend to be closer to the peak timing of the dominant limiting factor, DOY_Tmax or DOY_Wmax. The key idea is that, because the timing of maximum water availability (DOY_Wmax) is usually before the timing of maximum radiation (DOY_PAR), and the timing of peak temperature (DOY_Tmax) is after DOY_PAR, an earlier DOY_RTSIF than DOY_PAR (ΔDOY_RTSIF,PAR<0) suggests that the ecosystem is water constrained, while a later DOY_RTSIF than DOY_PAR (ΔDOY_RTSIF,PAR>0) suggests that the ecosystem is temperature constrained (Extended Fig. 1).

SVM Model for predicting ΔDOY _RTSIF,PAR

We used the SVM model to separate the data points with positive and negative ΔDOY_RTSIF,PAR values (Fig. 2). SVM is a widely used supervised learning method for classification problems⁴³. The labeled samples are first projected to a multidimensional (two in our case, MAT and MAP) space, with each dimension corresponding to one predictor variable. The essence of this algorithm is to find an optimum hyperplane that separates the labeled samples into different categories, so that the total distance between the hyperplane and data points is maximized. In our study, the classification was conducted for temperature and precipitation using MAT and MAP as predictors; positive and negative ΔDOY_RTSIF,PAR are the two groups that need to be separated. The resultant SVM is a function of the two predictor variables and can be used to predict whether the seasonal peak photosynthesis is limited by water or temperature.

CMIP6 models

We used 13 CMIP6 models to predict the future regions limited by precipitation or temperature. These models were obtained from the Climate Model Diagnostics and Earth System Grid Intercomparison Program (ESG) (https://esgf-node.llnl.gov/search/cmip6). We analyzed the SSP245 and SSP585 scenarios and calculated the future regions limited by precipitation or temperature based on the SVM model developed in this study (details of individual models in Supplementary Table 4).

Global vegetation models

We used GPP outputs from 13 DGVMs from the TRENDY S3 simulations (dynamic CO₂, climate and land use) to evaluate the performances of recent DGVMs to simulate peak vegetation growth⁴¹. These DGVMs included CABLE-POP, CLASS-CTEM, CLM5.0, DLEM, ISAM, JSBACH, JULES, LPJ-GUESS, LPX, OCN, ORCHIDEE, ORCHIDEE-CNP, SDGVM. We retrieved peak timings of vegetation photosynthesis (DOY_GPP) and seasonal peak photosynthesis (GPP_max) from 2000 to 2018 and calculated their trends and correlations. All model outputs were linearly interpolated to a daily temporal resolution and aggregated to a 0.5° spatial resolution.

Data availability

The RTSIF dataset is from https://doi.org/10.6084/m9.fgshare.19336346.v2. The FLUXNET2015 dataset is from https://fluxnet.org/data/fluxnet2015-dataset/. The CERES PAR product is from https://ceres.larc.nasa.gov/data/. The precipitation and surface air temperature are from https://disc.gsfc.nasa.gov/datasets/GLDAS_NOAH025_3H_2.1/. The GOME-2 SIF dataset is from https://avdc.gsfc.nasa.gov/pub/data/satellite/MetOp/GOME_F/.

Code availability

All computer codes for the analysis of the data are available from the corresponding author on reasonable request.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 91847301and No.51809007) and Major Science and Technology Programs, Qinghai Province (2021-SF-A7-1-3).

Author contributions

Yuefei Huang and Guangqian Wang designed the study. Xingan Chen performed the analysis. Xingan Chen and Shuo Zhang drafted this paper. Yuting Yang and Shen Tan contributed to the interpretation of the results. Chong Nie and Zhichao Chen processed the data.

Competing interests

The authors declare no competing interests.

Terrer, C. et al. A trade-off between plant and soil carbon storage under elevated CO2. Nature 591, 599-603 (2021).
Liu, Y., Wu, C., Wang, X. & Zhang, Y. Contrasting responses of peak vegetation growth to asymmetric warming: Evidences from FLUXNET and satellite observations. Global. Change. Biol. 29, 2363-2379 (2023).
Zhu, Z. et al. Greening of the Earth and its drivers. Nature climate change 6, 791-795 (2016).
Huang, K. et al. Enhanced peak growth of global vegetation and its key mechanisms. Nature Ecology & Evolution 2, 1897-1905 (2018).
Park, T. et al. Changes in timing of seasonal peak photosynthetic activity in northern ecosystems. Global. Change. Biol. 25, 2382-2395 (2019).
Xu, C., Liu, H., Williams, A. P., Yin, Y. & Wu, X. Trends toward an earlier peak of the growing season in Northern Hemisphere mid-latitudes. Global. Change. Biol. 22, 2852-2860 (2016).
Zhang, K. et al. Satellite-based model detection of recent climate-driven changes in northern high-latitude vegetation productivity. J. Geophys. Res. 113 (2008).
Keenan, T. F. & Riley, W. J. Greening of the land surface in the world’s cold regions consistent with recent warming. Nature Climate Change 8, 825-828 (2018).
Wang, T. et al. Emerging negative impact of warming on summer carbon uptake in northern ecosystems. Nat. Commun. 9, 1-7 (2018).
Gampe, D. et al. Increasing impact of warm droughts on northern ecosystem productivity over recent decades. Nature Climate Change (2021).
Xia, J. et al. Joint control of terrestrial gross primary productivity by plant phenology and physiology. Proc. Natl. Acad. Sci. 112, 2788-2793 (2015).
Zhou, S. et al. Dominant role of plant physiology in trend and variability of gross primary productivity in North America. Sci. Rep. 7, 41366 (2017).
Kattge, J. et al. TRY - a global database of plant traits. Global. Change. Biol. 17, 2905-2935 (2011).
Bauerle, W. L. et al. Photoperiodic regulation of the seasonal pattern of photosynthetic capacity and the implications for carbon cycling. Proc. Natl. Acad. Sci. 109, 8612-8617 (2012).
Wang, H. et al. Towards a universal model for carbon dioxide uptake by plants. Nature Plants 3, 734-741 (2017).
Yang, J. et al. Divergent shifts in peak photosynthesis timing of temperate and alpine grasslands in China. Remote Sens. Environ. 233, 111395 (2019).
Zhang, F. et al. Precipitation temporal repackaging into fewer, larger storms delayed seasonal timing of peak photosynthesis in a semi‐arid grassland. Funct. Ecol. 36, 646-658 (2022).
Zhang, T. et al. Joint control of alpine meadow productivity by plant phenology and photosynthetic capacity. Agr. Forest. Meteorol. 325, 109135 (2022).
Blackman, F. F. Optima and limiting factors. Ann. Bot-london. 19, 281-295 (1905).
Chen, X. et al. A long-term reconstructed TROPOMI solar-induced fluorescence dataset using machine learning algorithms. Sci. Data 9 (2022).
Luo, X. & Keenan, T. F. Global evidence for the acclimation of ecosystem photosynthesis to light. Nature Ecology & Evolution 4, 1351-1357 (2020).
Buermann, W., Bikash, P. R., Jung, M., Burn, D. H. & Reichstein, M. Earlier springs decrease peak summer productivity in North American boreal forests. Environ. Res. Lett. 8, 024027 (2013).
Novick, K. A. et al. The increasing importance of atmospheric demand for ecosystem water and carbon fluxes. Nature Climate Change 6, 1023-1027 (2016).
Yuan, W. et al. Increased atmospheric vapor pressure deficit reduces global vegetation growth. Sci. Adv. 5, eaax1396 (2019).
Geng, X. et al. Climate warming increases spring phenological differences among temperate trees. Global. Change. Biol. 26, 5979-5987 (2020).
Piao, S. et al. Weakening temperature control on the interannual variations of spring carbon uptake across northern lands. Nature Climate Change 7, 359-363 (2017).
Fu, Y. H. et al. Declining global warming effects on the phenology of spring leaf unfolding. Nature 526, 104-107 (2015).
Wang, K. et al. Regional and seasonal partitioning of water and temperature controls on global land carbon uptake variability. Nat. Commun. 13 (2022).
Xu, L. et al. Temperature and vegetation seasonality diminishment over northern lands. Nature Climate Change 3, 581-586 (2013).
Jiao, W. et al. Observed increasing water constraint on vegetation growth over the last three decades. Nat. Commun. 12 (2021).
Feng, X. et al. Recent leveling off of vegetation greenness and primary production reveals the increasing soil water limitations on the greening Earth. Sci. Bull. 66, 1462-1471 (2021).
Piao, S. et al. Characteristics, drivers and feedbacks of global greening. Nature Reviews Earth & Environment 1, 14-27 (2019).
Higgins, S. I., Conradi, T. & Muhoko, E. Shifts in vegetation activity of terrestrial ecosystems attributable to climate trends. Nature Geoscience 16, 147-153 (2023).
Gao, S. et al. An earlier start of the thermal growing season enhances tree growth in cold humid areas but not in dry areas. Nature Ecology & Evolution 6, 397-404 (2022).
Wang, X. et al. A two-fold increase of carbon cycle sensitivity to tropical temperature variations. Nature 506, 212-215 (2014).
Balting, D. F., Aghakouchak, A., Lohmann, G. & Ionita, M. Northern Hemisphere drought risk in a warming climate. npj Climate and Atmospheric Science 4 (2021).
Dai, A. Increasing drought under global warming in observations and models. Nature climate change 3, 52-58 (2013).
Zani, D., Crowther, T. W., Mo, L., Renner, S. S. & Zohner, C. M. Increased growing-season productivity drives earlier autumn leaf senescence in temperate trees. Science 370, 1066-1071 (2020).
Lian, X. et al. Seasonal biological carryover dominates northern vegetation growth. Nat. Commun. 12 (2021).
Park Williams, A. et al. Temperature as a potent driver of regional forest drought stress and tree mortality. Nature Climate Change 3, 292-297 (2013).
Sitch, S. et al. Recent trends and drivers of regional sources and sinks of carbon dioxide. Biogeosciences 12, 653-679 (2015).
Buermann, W. et al. Widespread seasonal compensation effects of spring warming on northern plant productivity. Nature 562, 110-114 (2018).
Suykens, J. A. K. & Vandewalle, J. Least Squares Support Vector Machine Classifiers. Neural. Process. Lett. 9, 293-300 (1999).
Zhang, Y. et al. Future reversal of warming-enhanced vegetation productivity in the Northern Hemisphere. Nature Climate Change 12, 581-586 (2022).
Magney, T. S. et al. Mechanistic evidence for tracking the seasonality of photosynthesis with solar-induced fluorescence. Proc. Natl. Acad. Sci., 201900278 (2019).
Köhler, P. et al. Global Retrievals of Solar‐Induced Chlorophyll Fluorescence With TROPOMI: First Results and Intersensor Comparison to OCO‐2. Geophys. Res. Lett. 45, 10,456-410,463 (2018).
Su, W., Charlock, T. P., Rose, F. G. & Rutan, D. Photosynthetically active radiation from Clouds and the Earth's Radiant Energy System (CERES) products. J. Geophys. Res. 112 (2007).
Rodell, M. et al. The Global Land Data Assimilation System. B. Am. Meteorol. Soc. 85, 381-394 (2004).
Pastorello, G. et al. The FLUXNET2015 dataset and the ONEFlux processing pipeline for eddy covariance data. Sci. Data 7 (2020).
Reichstein, M. et al. On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Global. Change. Biol. 11, 1424-1439 (2005).
Dušek, J., Čížková, H., Stellner, S., Czerný, R. & Květ, J. Fluctuating water table affects gross ecosystem production and gross radiation use efficiency in a sedge-grass marsh. Hydrobiologia 692, 57-66 (2012).
Pugh, C. A., Reed, D. E., Desai, A. R. & Sulman, B. N. Wetland flux controls: how does interacting water table levels and temperature influence carbon dioxide and methane fluxes in northern Wisconsin? Biogeochemistry 137, 15-25 (2018).
Sulla-Menashe, D., Gray, J. M., Abercrombie, S. P. & Friedl, M. A. Hierarchical mapping of annual global land cover 2001 to present: The MODIS Collection 6 Land Cover product. Remote Sens. Environ. 222, 183-194 (2019).
Jakubauskas, M. E., Legates, D. R. & Kastens, J. H. Harmonic analysis of time-series AVHRR NDVI data. Photogramm. Eng. Rem. S 67, 461-470 (2001).
Liu, Q. et al. Temperature, precipitation, and insolation effects on autumn vegetation phenology in temperate China. Global. Change. Biol. 22, 644-655 (2016).
Zhang, Y., Commane, R., Zhou, S., Williams, A. P. & Gentine, P. Light limitation regulates the response of autumn terrestrial carbon uptake to warming. Nature Climate Change 10, 739-743 (2020).

There is NO Competing Interest.

SupplementaryInformation.pdf
Supporting Information
FigE1.png
Extended Fig.1 | Illustration of the optimal GPP_max conceptual model. Colored solid curves indicate the seasonal cycles of temperature (T, blue), radiation (PAR, green), and water (W, red). Vertical lines indicate when each variable reaches a maximum state, and the colored dashed line marks the GPP_max reached in this state. The black solid lines represent the envelope of the potential GPP_max in different states. The seasonal peak timing of the photosynthesis is adjusted to match the radiation (DOY_GPP = DOY_PAR), and therefore the optimized maximum seasonal GPP was achieved.

Download PDF

Version 1

posted

You are reading this latest preprint version

Climatic warming strengthening water constraint on seasonal peak photosynthesis in northern ecosystems

Status:

Version 1

Abstract

Figures

Main

Methods

Declarations

Data availability

Code availability

Acknowledgements

Author contributions

Competing interests

References

Additional Declarations

Supplementary Files

Status:

Version 1