Elsevier

Ecological Modelling

Volume 369, 10 February 2018, Pages 88-100
Ecological Modelling

Modeling the forest phosphorus nutrition in a southwestern Swedish forest site

https://doi.org/10.1016/j.ecolmodel.2017.12.018Get rights and content

Highlights

  • Klintaskogen is likely P limited today, but might return to be N limited in future under low N deposition projection.

  • Plant P uptake is dominated by mineralization, but inputs still contribute ca. 20%.

  • P weathering accounts for about 2/3 of these forest P inputs; P is accumulating in SOM.

Abstract

In this study, a phosphorus (P) module containing the biogeochemical P cycle has been developed and integrated into the forest ecosystem model ForSAFE. The model was able to adequately reproduce the measured soil water chemistry, tree biomass (wood and foliage), and the biomass nutrient concentrations at a spruce site in southern Sweden. Both model and measurements indicated that the site showed signs of P limitation at the time of the study, but the model predicted that it may return to an N-limited state in the future if N deposition declines strongly. It is implied by the model that at present time, the plant takes up 0.50 g P m−2 y−1, of which 80% comes from mineralization and the remainder comes from net inputs, i.e. deposition and weathering. The sorption/desorption equilibrium of P contributed marginally to the supply of bioavailable P, but acted as a buffer, particularly during disturbances.

Introduction

Forests are among the most important ecosystems on the planet, providing and regulating multiple important services such as timber production, biodiversity conservation, carbon (C) sequestration, bioenergy supply and potable water supply (Nelson et al., 2011; COM, 2013). Nitrogen (N) is often reported to limit growth in northern forest ecosystems (Tamm, 1991; Jonard et al., 2015). The increase in atmospheric N deposition due to anthropogenic activities has shifted forest ecosystems from being N-limited towards being N-saturated (Aber et al., 1989; Aber et al., 1998), causing N leaching in some forest ecosystems (Gundersen et al., 2006; Kreutzer et al., 2009; Yu et al., 2016). An increased nitrogen pool in the forest can affect other nutrient pools and thereby forest nutrition. It can compromise the availability of base cations by depleting soil base cations through N leaching, which causes acidification and eutrophication (Driscoll et al., 2003; Eriksson et al., 1992; Likens et al., 1996). Another effect of increased N status is a stimulated forest growth (Reich et al., 2006; Ciais et al., 2013), leading to the limitation of other nutrients such as phosphorus (P) (Aber et al., 1989; Akselsson et al., 2008). A switch from N limitation to P limitation over recent decades has been found in many forest studies and experimental studies in Europe (Braun et al., 2010; Jonard et al., 2015; Talkner et al., 2015; Flückiger and Braun, 1999) and North America (Crowley et al., 2012; Tessier and Raynal, 2003; Gress et al., 2007). Such a transition, from N limitation to P limitation, is not commonly detected in Swedish forests (Ingerslev et al., 2001; Högberg et al., 2006) due to the generally relatively low atmospheric N deposition in Sweden (Simpson et al., 2011). However, in southwestern Sweden, where current and historical N deposition is highest (Akselsson et al., 2010), it is suspected that P limitation might already occur (Rosengren-Brinck and Nihlgård, 1995; Akselsson et al., 2008).

Forest P limitation has long been evaluated by the needle N/P ratio (Linder, 1995; Rosengren-Brinck and Nihlgård, 1995; Mellert and Göttlein, 2012; Jonard et al., 2015; Braun et al., 2010), partly because foliar nutrient concentrations and ratios are well-established indicators of nutrient limitation in forest trees (Mellert and Ewald, 2014; Jonard et al., 2015). Particularly because P nutrition in forests is more challenging to evaluate than other nutrients, due to the high uncertainties in quantifying the biogeochemical P processes (Frossard et al., 2011; Shen et al., 2011; Fox et al., 2011; Jones and Oburger, 2011), and measuring the soil P availability (Shen et al., 2011; Hinsinger, 2001). Due to the high uncertainties in P processes measurement, the forest P cycle has not been much quantitatively investigated in field studies (Yanai, 1992; Yanai, 1998; Jonard et al., 2009), nor has it been soundly evaluated in modeling studies (Jonard et al., 2010; Achat et al., 2009; Wang et al., 2010; Yang et al., 2014; Müller and Bünemann, 2014). These highly uncertain P processes include atmospheric deposition (Newman, 1995; Tipping et al., 2014), weathering (Newman, 1995; Smits et al., 2012), sorption/desorption (McGechan and Lewis, 2002; Frossard et al., 2011), mineralization (Bünemann, 2015), and rhizosphere processes (Hinsinger, 2001; Hinsinger et al., 2011). Nevertheless, forest P cycle and its impacts on other cycles (e.g. C and N) are much less investigated in modeling studies simply due to the absence of P cycle in most forest/terrestrial ecosystem models (Fontes et al., 2010; Flato et al., 2013).

In this paper, we first implemented a P module, which contains the biogeochemical processes of the full P cycle, into the integrated dynamic forest model, ForSAFE. We then tested the model at a southwestern Swedish forest site, which is at high risk of P limitation. The aims of this study were: 1) to evaluate the forest nutrition (N and P) at the study site, 2) to quantify the forest P cycle, especially the biogeochemical P processes, from a modeling perspective.

Section snippets

The ForSAFE model

ForSAFE is a mechanistic biogeochemical model of the forest ecosystem and was designed to simulate the dynamic responses of the forest ecosystem to environmental changes (Zanchi et al., 2014; Yu et al., 2016). The model aggregates independent processes—chemical, physical, and physiological—based on empirical evidence (Belyazid, 2006; Wallman et al., 2005). These independent but mutually interacting processes bring together three basic material and energy cycles to form a single integrated

Model evaluation

The model overestimated the wood biomass by 17% and underestimated the foliage biomass by 13%, but the modeled trends for the change in wood biomass and leaf biomass agreed well with the measurements (Fig. 3). The modeled needle N concentrations were about 5% lower than the forest inventory data. Although less noticeable, the model still captured the decrease in needle N concentration after 2000. The modeled needle P concentrations were within the range of the forest inventory data, but the

Forest nutrition at the study site

Swedish forests are mostly limited by N due to low N deposition (Akselsson et al., 2007; Akselsson et al., 2008). Previous forest studies in Sweden have shown that Swedish forests usually have either relatively high needle P concentrations (Vestin et al., 2013; Rothpfeffer and Karltun, 2007) or low needle N/P ratios due to very low needle N concentrations (Bauer et al., 1997; Anonymous, 2003). However, low needle P concentrations or high needle N/P ratios (>10) have been reported in southern

Conclusions

The ForSAFE model was implemented with a P module containing the biogeochemical P cycle and tested at Klintaskogen forest site. Both the forest inventory data and the model results supported the suspicion that Klintaskogen is already limited by P. The model simulations showed that between1900 and 2000 the site switched from being N-limited to being P-limited, and in future the site may return to N limitation, however, this prediction is heavily dependent on the low future projection of N

Acknowledgments

The authors wish to express their gratitude for the financial support granted by the project “Biodiversity and Ecosystem services in a Changing Climate” (BECC). We also thank Gregory van der Heijden, Nicholas Rosenstock, Ann-Mari Fransson, Christian Morel and Pål Axel Olsson for their valuable advice on model development.

References (100)

  • C.T. Driscoll

    Effects of acidic deposition on forest and aquatic ecosystems in New York State

    Environ. Pollut.

    (2003)
  • M. Jonard

    Modeling forest floor contribution to phosphorus supply to maritime pine seedlings in two-layered forest soils

    Ecol. Modell.

    (2010)
  • C. Müller et al.

    A 33P tracing model for quantifying gross P transformation rates in soil

    Soil Biol. Biochem.

    (2014)
  • S. Manzoni et al.

    Soil carbon and nitrogen mineralization: theory and models across scales

    Soil Biol. Biochem.

    (2009)
  • M.B. McGechan et al.

    Sorption of phosphorus by soil, part 1: principles, equations and models

    Biosyst. Eng.

    (2002)
  • W.B. McGill et al.

    Comparative aspects of cycling of organic C, N, S and P through soil organic matter

    Geoderma

    (1981)
  • C. Rothpfeffer et al.

    Inorganic elements in tree compartments of Picea abies-Concentrations versus stem diameter in wood and bark and concentrations in needles and branches

    Biomass Bioenergy

    (2007)
  • C.W. Runyan et al.

    Hydrologic controls on phosphorus dynamics: a modeling framework

    Adv. Water Resour.

    (2012)
  • J.P. Schimel et al.

    The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model

    Soil Biol. Biochem.

    (2003)
  • C. Stroia et al.

    Dynamics of diffusive soil phosphorus in two grassland experiments determined both in field and laboratory conditions Agriculture

    Ecosyst. Environ.

    (2007)
  • H. Wallander et al.

    The stimulating effect of apatite on ectomycorrhizal growth diminishes after PK fertilization

    Soil Biol. Biochem.

    (2008)
  • P. Wallman

    ForSAFE—an integrated process-oriented forest model for long-term sustainability assessments

    For. Ecol. Manage.

    (2005)
  • P. Wallman

    DECOMP –a semi-mechanistic model of litter decomposition

    Environ. Modell. Softw

    (2006)
  • F. Werner

    Small-scale spatial distribution of phosphorus fractions in soils from silicate parent material with different degree of podzolization

    Geoderma

    (2017)
  • R.D. Yanai

    The effect of whole-tree harvest on phosphorus cycling in a northern hardwood forest

    For. Ecol. Manage.

    (1998)
  • L. Yu

    Storm disturbances in a Swedish forest-A case study comparing monitoring and modelling

    Ecol. Modell.

    (2016)
  • G. Zanchi

    Modelling the effects of management intensification on multiple forest services: a Swedish case study

    Ecol. Modell.

    (2014)
  • J.D. Aber et al.

    A generalized, lumped-parameter model of photosynthesis, evaporation and net primary production in temperate and boreal forest ecosystems

    Oecologia

    (1992)
  • J.D. Aber

    Nitrogen saturation in northern forest ecosystems

    Bioscience

    (1989)
  • J. Aber

    Nitrogen saturation in temperate forest ecosystems − Hypotheses revisited

    Bioscience

    (1998)
  • D.L. Achat et al.

    Process-Based assessment of phosphorus availability in a low phosphorus sorbing forest soil using isotopic dilution methods

    Soil Sci. Soc. Am. J.

    (2009)
  • M. Alveteg

    Dynamics of Forest Soil Chemistry

    (1998)
  • Anonymous, 2003. Ecocyclic pulp mill –KAM. Final report 1996–2002.,...
  • P.M. Attiwill et al.

    Nutrient cycling in forests

    New Phytol.

    (1993)
  • G. Bauer et al.

    Nutrient contents and concentrations in relation to growth of Picea abies and Fagus sylvatica along a European transect

    Tree Physiol.

    (1997)
  • S. Belyazid

    Dynamic Modelling of Biogeochemical Processes in Forest Ecosystems

    (2006)
  • B. Berg et al.

    Plant Litter: Decomposition, Humus Formation, Carbon Sequestration

    (2008)
  • P. Ciais

    Carbon and other biogeochemical cycles

  • C.C. Cleveland et al.

    C:N:P stoichiometry in soil: is there a redfield ratio for the microbial biomass?

    Biogeochemistry

    (2007)
  • COM
  • K.F. Crowley

    Do nutrient limitation patterns shift from nitrogen toward phosphorus with increasing nitrogen deposition across the northeastern United States?

    Ecosystems

    (2012)
  • A. Ericsson

    Concentrations of mineral nutrients and arginine in needles of Picea abies trees from different areas in southern Sweden in relation to nitrogen deposition and humus form

    Ecol. Bull.

    (1995)
  • E. Eriksson et al.

    Acidification of in Sweden forest

    Ambio

    (1992)
  • J.C. Fardeau

    Dynamics of phosphate in soils: an isotopic outlook

    Fertil. Res.

    (1995)
  • R. Fischer et al.

    Forest Condition in Europe 2011 Technical Report of ICP Forests and FutMon

    (2011)
  • W. Flückiger et al.

    Nitrogen and its effect on growth, nutrient status and parasite attacks in beech and Norway spruce

    Water Air Soil Pollut.

    (1999)
  • G. Flato

    Evaluation of climate models

  • T.D. Fletcher

    Package QuantPsyc.

    (2015)
  • L. Fontes

    Models for supporting forest management in a changing environment

    Forest Systems

    (2010)
  • D. Fowler

    The global nitrogen cycle in the twenty-first century

    Philos. Trans. R. Soc. B

    (2013)
  • Cited by (22)

    • Protection of forest ecosystems in the eastern United States from elevated atmospheric deposition of sulfur and nitrogen: A comparison of steady-state and dynamic model results

      2023, Environmental Pollution
      Citation Excerpt :

      However, dynamic biogeochemical models used for estimating TLs remain under development, particularly with regard to weathering and microbial process that govern nutrient and carbon dynamics in soils. For example, developments to ForSAFE have been made to explicitly simulate the microbial community with linkages to soil decomposition and nutrient uptake (Yu et al., 2018) and to improve soil mineral weathering estimates (Sverdrup et al., 2019). Therefore, it may also be possible to further develop the CL/TL models to better represent mechanisms of nutrient N effects on terrestrial species, such as competitive interactions in mixed species forests.

    • Phosphorus speciation in the organic layer of two Swedish forest soils 13–24 years after wood ash and nitrogen application

      2022, Forest Ecology and Management
      Citation Excerpt :

      In the mineral soil, a large amount of phosphate (PO4) is adsorbed to poorly ordered, surface aluminum (Al) and iron (Fe) mineral phases (Prietzel et al., 2016; Tuyishime et al., 2022), which act as a long-term buffer of the equilibrium P concentration (Wood et al. 1984). Forest harvests remove a large amount of nutrients, which might decrease the long-term availability of P (Akselsson et al., 2008; Yanai, 1998; Yu et al., 2018). Co-limitation of P caused by excessive nitrogen (N) deposition (Hedwall et al., 2017) and increased leaching/erosion due to wildfire (Lagerström et al., 2009) may also contribute to long-term P deficiency.

    • Phosphorus abundance and speciation in acid forest Podzols – Effect of postglacial weathering

      2022, Geoderma
      Citation Excerpt :

      At a later stage of pedogenesis, the bioavailability of P is decreased, and P occluded within Al and Fe oxides and P associated with organic matter (OM) are the main pools of P (Turner and Condron, 2013; Walker and Syers, 1976). However, it has been argued that some forest soils in central and northern Europe, although at an early stage of soil development, are characterized by a low P nutritional status and P-limited conditions (Ilg et al., 2009; Yu et al., 2018), which may be affected by a low P availability of the P adsorbed to Fe and Al phases in the subsoil. Further, a high N availability due to high atmospheric deposition rates, enhanced forest productivity and intensified use of forest resources may contribute to an increased plant P demand (Akselsson et al., 2008; Heuck et al., 2018; Jonard et al., 2015; Mohren et al., 1986; Yanai, 1998).

    • Simulation of water and chemical transport of chloride from the forest ecosystem to the stream

      2021, Environmental Modelling and Software
      Citation Excerpt :

      The model simulates the most relevant chemical cycles in natural ecosystems: carbon, nitrogen, base cations (sodium, calcium, magnesium, potassium), sulphur, chlorine and aluminium. More recently, the phosphorous cycle was also added to the model (Yu et al., 2018). As a result of the changes in hydrology and soil chemistry, the two-dimensional version of the model, ForSAFE-2D (Fig. 1), can simulate water flows and the transport of chemical elements from the forest ecosystem to the stream on a daily basis.

    • Phosphorus in 2D: Spatially resolved P speciation in two Swedish forest soils as influenced by apatite weathering and podzolization

      2020, Geoderma
      Citation Excerpt :

      The primary production of these ecosystems is usually limited by the availability of nitrogen (N), and currently there are few indications of P limitation (Akselsson et al., 2008; Binkley and Högberg, 2016). However, the increased atmospheric N deposition in the last decades have rendered some forest ecosystems more N-enriched, which may increase problems with N leaching (Gundersen et al., 2006), and thereby make P a more critical nutrient (Yu et al., 2018). In addition, today’s forest management methods, with harvesting of biomass, could lead to a successive depletion of bioavailable P in forest soils.

    View all citing articles on Scopus
    View full text