Elsevier

Ecological Modelling

Volume 456, 15 September 2021, 109663
Ecological Modelling

The impacts of vegetation on the soil surface freezing-thawing processes at permafrost southern edge simulated by an improved process-based ecosystem model

https://doi.org/10.1016/j.ecolmodel.2021.109663Get rights and content

Highlights

  • The better parameterizaitons of buffer zone in the BEPS model can improve the simulations of soil temperature.

  • The improved BEPS model performed better in simulating soil FT processes.

  • Forests have smaller amplitudes of diurnal soil FT cycles and delayed soil thaw timing compared to grasslands.

Abstract

Permafrost degradation due to climate warming would potentially increase the release of previously frozen soil carbon and change the carbon budget of the cold region ecosystem. The underlying permafrost degradation would be effectively mediated by soil surface freezing-thawing (FT) processes. Aboveground vegetation can regulate soil FT processes, however its effects on ground thermal transfer have not been well represented by ecosystem models. In this study, we improved the hydrothermal module of the Boreal Ecosystem Productivity Simulator (BEPS) through more careful parameterization of snowpack density, puddled water, soil organic matter and super-cooled soil water. The impacts of vegetation on the soil surface FT processes have also been investigated using the improved BEPS model and the measured soil temperature data at forest and grassland sites on the southern edge of permafrost region in Mongolia and northeastern China. The improved BEPS model performs better than the original model in simulations of soil temperature and soil FT processes. Smaller amplitudes of soil diurnal FT cycles were found in forest sites compared to grassland sites. Forest sites have delayed soil thaw timing and similar soil freezing time compared to grassland sites. Differences in snow depths and soil organic matter content due to distinct vegetation community structures have considerable influences on the disparity in soil FT processes. Thus, it is important to improve the simulation of the impacts of vegetation on soil surface FT processes for better forecasting the permafrost degradation.

Introduction

Permafrost (the ground that remains below 0 °C for at least two consecutive years) underlies approximately 23.9% of the exposed land surface of the Northern Hemisphere and it contains twice as much carbon as the atmosphere (Grosse et al., 2016; Schädel et al., 2016). The large quantities of carbon stored in the frozen soil could be released into the atmosphere due to permafrost degradation caused by climatic warming. This degradation would potentially change the global carbon budget and accelerate global climate warming (Chaudhary et al., 2020; Hollesen et al., 2011). The most dramatic permafrost degradation mainly occurs at the southern edge of permafrost area due to the significant poleward movement of permafrost caused by climate warming (Guo et al., 2018; Yue et al., 2020). The soil surface freezing-thawing (FT) processes, which are measured by the amplitude and duration of the diurnal soil FT cycle as well as seasonal FT timing, would effectively mediate the underlying permafrost degradation (Guo et al., 2018). The soil FT processes during the transition period from cold-to-warm or warm-to-cold seasons can last from a few days to a few weeks and they are susceptible to change because of the rapid climate warming (Bao et al., 2021; Pachauri et al., 2014).

There have been many researches investigating the soil FT processes in response to climatic warming, mainly using near-surface soil temperature data from in situ measurements (Bao et al., 2021; Cao et al., 2020; Wang et al., 2020; Yoshikawa et al., 2020), land surface temperature data from satellites (Baltzer et al., 2014; Kim et al., 2014) as well as ecosystem model simulations (Foster et al., 2019; Guo and Wang, 2014). However, there are still uncertainties in these studies due to the sparse and short-term field measurements (Tao et al., 2017), coarse resolution of the available remote sensing data (Grosse et al., 2016; Tao et al., 2017) or the poor representation of soil FT processes in ecosystem models (Barman and Jain, 2016; Jan et al., 2020; Nitzbon et al., 2020). To better understand the pattern and drivers of the soil FT processes at regional scale, the more realistic representation and parameterizations of soil FT processes are needed to be incorporated into ecosystem models.

The impacts of vegetation on soil FT processes have been examined by many studies. Field measurements in south Siberia (Hu et al., 2013), Zhangbei county of Hebei province, China (Chen et al., 2020) and Qinghai-Tibetan Plateau (Hu et al., 2020a) found that vegetation could exert buffer effect on soil temperature, and consequently affect the soil FT processes. Guo et al. (2018) found that soil FT cycles would be larger in amplitude and longer in duration at steppe sites than at forest sites. Soils in the forest sites and steppe sites froze almost simultaneously, but experience a delay in thawing for forest sites (Guo et al., 2018). In addition, the hydrologic and thermal regimes of soil surface and soil FT processes could be affected by vegetation through intercepting snowfall and solar radiation (Chang et al., 2015; French, 2007; Karjalainen et al., 2019) and through buffer effect of litter layer (Chaudhary et al., 2020; Lawrence and Slater, 2008). Thus, it is essential to improve the ecosystem model for better simulating the impacts of vegetation on soil FT processes.

Three forest sites and two grassland sites investigated in this study were located along a vast latitudinal (around 10°) and longitudinal (around 20°) gradient at the south edge of the permafrost region of Mongolia and northeastern China. Soil temperature and meteorological data were recorded at these sites during 2003 to 2014. The soil-vegetation thermal and hydraulic modules of the Boreal Ecosystems Productivity Simulator (BEPS) have been improved and tested at the studied sites. Using the measured soil temperature data and those simulated by the BEPS model, we investigated the impacts of vegetation types on soil FT processes.

Section snippets

Site description

Three forest sites and two grassland sites at the southern edge of a permafrost region in Mongolia and northeastern China were investigated in this study (Fig. 1 and Table 1). Three sites (SKT, DXF, and KBU) are in the permafrost region, and two sites (CHB and CNG) belong to the region of seasonally frozen ground. These sites were selected fulfilling the following conditions in data quality and availability: (1) soil data including soil texture and soil organic carbon (SOC) content are

Simulation of soil temperature

The hourly soil temperature simulated by the original and improved BEPS model were compared with measured data to evaluate the performance of the original and improved BEPS model in simulating soil temperature at the studied sites (Supplementary materials, Fig. S1). The regression statistics are summarized in Table 2. The original and improved BEPS models are both able to explain >90% of the variance in soil temperature. The root mean square error (RMSE) of soil temperature simulated by the

The impacts of different improvement schemes on the simulations of soil temperature

The Gaussian kernel density estimation (KDE) of the differences between the simulated soil temperatures from the experimental model runs and the observations of the five studied sites was conducted to compare the contribution of the four modifications of BEPS model in simulating soil temperature. Considering the impacts of canopy and wind speed on the snowpack density reduces the MAE of simulated soil temperature from 3.23 to 2.90 °C (Fig. 7). The incorporation of puddled water into the

Conclusions

In this study, the BEPS model has been improved through more careful treatments of the impacts of thermal “buffer zone” (i.e. snowpack, puddled water and SOM) on subsurface heat transfer. The different patterns of soil surface FT processes between forest and grassland and the regulation of vegetation to soil FT processes were analyzed based on the measured soil temperature data and those simulated by the BEPS model. The major conclusions are drawn as follows:

  • (1)

    Compared to the original BEPS model,

CRediT

Zhenhai Liu: Conceptualization, Software, Writing - Original draft preparation.

Bin Chen: Conceptualization, Methodology, Writing - Review and Editing.

Shaoqiang Wang: Writing - Review and Editing, Supervision, Project administration.

Qinyi Wang: Validation, Data Curation.

Jinghua Chen: Software.

Weibo Shi: Software.

Xiaobo Wang: Software.

Yuanyuan Liu: Visualization.

Yongkai Tu: Visualization.

Mei Huang: Supervision.

Junbang Wang: Supervision.

Zhaosheng Wang: Supervision.

Hui Li: Data Curation.

Tongtong

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors sincerely appreciate Prof. Jingming Chen for sharing the codes of the BEPS model. This study used eddy covariance data acquired and shared by ChinaFLUX, AsiaFlux and FLUXNET. The GLOBMAP-V2 LAI datasets and SoilGrids dataset are provided by Yang Liu and Tomislav Hengl. We greatly offer our profound appreciation to all the providers of the freely available data. This research was supported by the Science and Technology Strategic Pilot of the Chinese Academy of Sciences (XDA20030203).

References (67)

  • Y. Kim et al.

    Response of vegetation growth and productivity to spring climate indicators in the conterminous united states derived from satellite remote sensing data fusion

    Agric. For. Meteorol.

    (2014)
  • P. Kumar et al.

    Assimilation of near-surface temperature using extended kalman filter

    Adv. Wat. Resour.

    (2003)
  • S. Li et al.

    Energy partitioning and its biophysical controls above a grazing steppe in central mongolia

    Agric. For. Meteorol.

    (2006)
  • J. Liu et al.

    A process-based boreal ecosystem productivity simulator using remote sensing inputs

    Rem. Sens. Environ.

    (1997)
  • S. Wang et al.

    New calculation methods of diurnal distribution of solar radiation and its interception by canopy over complex terrain

    Ecol. Modell.

    (2002)
  • Y. Yue et al.

    Ecological indicators of near-surface permafrost habitat at the southern margin of the boreal forest in china

    Ecol. Indic.

    (2020)
  • S. Zhang et al.

    Evaluation and improvement of the daily boreal ecosystem productivity simulator in simulating gross primary productivity at 41 flux sites across Europe

    Ecol. Modell.

    (2018)
  • X. Zhang et al.

    Species-specific tree growth and intrinsic water-use efficiency of dahurian larch (larix gmelinii) and mongolian pine (pinus sylvestris var. mongolica) growing in a boreal permafrost region of the greater hinggan mountains, northeastern China

    Agric. For. Meteorol.

    (2018)
  • J.L. Baltzer et al.

    Forests on thawing permafrost: fragmentation, edge effects, and net forest loss

    Glob. Chang. Biol.

    (2014)
  • T. Bao et al.

    Much stronger tundra methane emissions during autumn freeze than spring thaw

    Glob. Chang. Biol.

    (2021)
  • R. Barman et al.

    Comparison of effects of cold-region soil/snow processes and the uncertainties from model forcing data on permafrost physical characteristics

    J. Adv. Model. Ear. Syst.

    (2016)
  • J. Brown et al.

    Circum-arctic Map of Permafrost and Ground-Ice conditions. Version 2. Boulder, Color.

    (2002)
  • B. Cao et al.

    The era5-land soil-temperature bias in permafrost regions

    Cryosph.

    (2020)
  • X. Chang et al.

    Thermal impacts of boreal forest vegetation on active layer and permafrost soils in northern da xing'anling (hinggan) mountains, northeast china

    Arct., Antarct. Alp. Res.

    (2015)
  • N. Chaudhary et al.

    Modelling past and future peatland carbon dynamics across the pan-arctic

    Glob. Chang. Biol.

    (2020)
  • T. Che et al.

    Snow depth derived from passive microwave remote-sensing data in china

    Ann. Glaciol.

    (2008)
  • B. Chen et al.

    Including soil water stress in process-based ecosystem models by scaling down maximum carboxylation rate using accumulated soil water deficit

    Agric. For. Meteorol.

    (2019)
  • B. Chen et al.

    Characteristics of spatial and temporal variation of seasonal and short-term frozen soil in china in recent 50 years

    Chines. J. Atmos. Sci.

    (2008)
  • J.M. Chen et al.

    Vegetation structural change since 1981 significantly enhanced the terrestrial carbon sink

    Nat. Commun.

    (2019)
  • L. Chen et al.

    Influences of forest cover on soil freeze-thaw dynamics and greenhouse gas emissions through the regulation of snow regimes: a comparison study of the farmland and forest plantation

    Sci. Tot. Environ.

    (2020)
  • R.B. Clapp et al.

    Empirical equations for some soil hydraulic properties

    Wat. Resour. Res.

    (1978)
  • D.A. de Vries

    Thermal properties of soils

    Phys. Plant Environ.

    (1963)
  • G. Dong et al.

    Effects of spring drought on carbon sequestration, evapotranspiration and water use efficiency in the songnen meadow steppe in northeast china

    Ecohydrol.

    (2011)
  • Cited by (0)

    View full text