Modelling impacts of atmospheric deposition and temperature on long-term DOC trends
Graphical abstract
Introduction
Long-term monitoring of surface water quality has revealed increasing concentrations of dissolved organic carbon (DOC) across large parts of the Northern Hemisphere, particularly close to industrialised regions (Skjelkvale et al., 2001, Driscoll et al., 2003, Evans et al., 2005, Monteith et al., 2007, Erlandsson et al., 2008). These observations have raised concerns over increasing water treatment costs (Ritson et al., 2014b) and possible destabilisation of terrestrial carbon stocks (Freeman et al., 2001). A debate has ensued over the possible causes of observed increases (Clark et al., 2010), that have included climate change (Freeman et al., 2001), changes in land management and use (Yallop and Clutterbuck, 2009), nitrogen (N) deposition (Findlay, 2005), CO2 enrichment (Freeman et al., 2004) and declines in acid deposition (Evans et al., 2006, Monteith et al., 2007). Analyses of surface water data (Evans et al., 2006, De Wit et al., 2007, Oulehle and Hruška, 2009, Erlandsson et al., 2010, Monteith et al., 2014), supported by evidence from laboratory (Clark et al., 2006, Clark et al., 2011) and field studies (Clark et al., 2005, Ekström et al., 2011, Evans et al., 2012) have pointed to effects of declining sulphur deposition as the major cause, but do not exclude the possibility that other drivers have also exerted influence on DOC trends.
Decreases in acid anion concentrations and increases in soil pH associated with a reduction in acid deposition are thought to have increased the solubility of potentially-dissolved organic matter (pDOM) by increasing negative charges on clay and organic matter surfaces (Tipping and Woof, 1991). There is also evidence that regional warming (e.g. Freeman et al., 2001, Pastor et al., 2003) and changes in precipitation patterns (e.g. Keller et al., 2008, Pumpanen et al., 2014) can affect DOC concentrations by influencing decomposition rates, vegetation type or export paths. A further suggested mechanism is the effect of changed flow paths due to changing precipitation patterns (e.g. Hongve et al., 2004, Erlandsson et al., 2008, Couture et al., 2012). The relative degree to which these factors have contributed to DOC trends has been debated extensively (e.g. Evans et al., 2006, Eimers et al., 2008, Futter and De Wit, 2008, Clark et al., 2010).
Several studies suggest that there is also a link between N deposition and DOC leaching (e.g. Pregitzer et al., 2004, Findlay, 2005, Bragazza et al., 2006). Nitrogen typically limits productivity in terrestrial ecosystems (Vitousek and Howarth, 1991), so increased net ecosystem productivity due to N deposition might be expected to increase the pool of ecosystem C available for DOC production. This would, however, depend on prevailing levels of ecosystem N saturation. In N-limited ecosystems addition of reactive N would be expected to exert a fertilising effect (LeBauer and Treseder, 2008). Conversely in N-saturated environments additional N would be expected to contributes to acidification (Emmett et al., 1998), that in turn could reduce decomposition (Janssens et al., 2010), and consequently a reduction in DOC production and solubility (Evans et al., 2008). To predict how DOC levels are likely to change in the future it is therefore necessary to consider the integrated effects of acidifying and eutrophying effects of air pollution and climate change on productivity, decomposition and organic matter dissolution.
One of the criticisms levelled at investigations into the drivers of DOC increases in soils or waters is that studies founded on correlation (e.g. Skjelkvale et al., 2001, Vuorenmaa et al., 2006, Monteith et al., 2007, Oulehle and Hruška, 2009, Sarkkola et al., 2009, Zhang et al., 2010, Borken et al., 2011) do not in themselves provide proof of causation (Roulet and Moore, 2006). In addition, study sites tend to be concentrated within geographically limited areas and findings may, therefore, not necessarily be universally applicable. Furthermore, although soils (particularly upper organic horizons) are recognised to often be the source of most freshwater DOC (e.g. Brooks et al., 1999, Billett et al., 2006, Evans et al., 2007a, Winterdahl et al., 2011), soil water monitoring data are scarce, and typically of shorter duration than surface water data. There is increasing evidence that shallow soil water makes a major contribution to trends in DOC in surface water (Hruška et al., 2014, Sawicka et al., 2016) although the relationship between soil and surface water concentrations is complicated by riparian and subsoil processes (Lofgren et al., 2010, Löfgren and Zetterberg, 2011). Despite their limitations, however, long-term soil water monitoring data provide the most effective resource for testing whether mechanisms that have been shown to operate in experiments also operate at larger spatial and temporal scales. Therefore, we brought together the United Kingdom's best long term soil solution records in order to provide a foundation for testing our current process understanding and consider how anticipated change in climate and deposition are likely to influence future behaviour of DOC.
To date, the majority of DOC process-based modelling studies have concentrated on model developments and potential applications, or on simulating time series for direct comparison with measurements (e.g. Futter et al., 2007, Futter et al., 2011, Jutras et al., 2011, Xu et al., 2012, Zhang et al., 2013, Dick et al., 2014). Relatively few, in contrast have gone on to consider the longer-term implications of model parameterisation, such as the most likely pre-industrial “baseline” DOC levels that can help to inform catchment restoration and management strategies. Exceptions include, Hruška et al. (2014), who linked a simple empirical DOC function to the MAGIC acidification model to recreate DOC trends in an acid-sensitive podzol site in the Czech Republic. This study was however based on modelling DOC at organo-mineral sites only. Valinia et al. (2015), in turn, reconstructed reference conditions of total organic carbon and long-term monitoring data to predict recent DOC changes in Swedish lakes. Historic reconstructions like these provide a framework with which to consider the likely relative importance of various potential anthropogenic pressures.
In the current study, DOC trends were simulated at long-term monitoring sites using an annual time-step model, with the aim of exploring the likely relative importance of different drivers and considering how DOC concentrations in soil water might be expected to change in the future. Here we use the MADOC model (Rowe et al., 2014) which simulates the long-term controls on DOC from terrestrial sites to streams, is responsive to a number of drivers, and can be applied to catchments at any scale using a lumped-parameter approach. The model is a representation of soil and vegetation carbon dynamics, acid-base dynamics and organic matter dynamics. It has been shown to reproduce the effects of the key drivers of DOC in terrestrial experimental sites and long-term surface water monitoring sites (Rowe et al., 2014). We set out to first test the model directly against soil water monitoring data, and then consider the likely relative effects of key contributory drivers in the model in influencing soil water DOC at a range of sites with different characteristics over the longer term. We therefore applied MADOC to six terrestrial long-term monitoring sites characterised by different vegetation, soil type and acid deposition loading and considered: (1) the extent of discrepancies between modelled trends, based on the hypothesised drivers (anthropogenic sulphate, chloride, N deposition, temperature change), and measured trends and (2) the changes that would have occurred with and without individual drivers to assess the magnitude of impact of each on different ecosystems and on future DOC dynamics.
Section snippets
Field sites, measurements and chemical analyses
Data from three United Kingdom Forest Level II (FLII) and three terrestrial Environmental Change Network (ECN) sites were used for this study (Fig. 1). FLII sites were established in 1995 (Vanguelova et al., 2007) and form part of the European forest monitoring network (ICP Forests) that aims to improve understanding of the effects of air pollution and other environmental factors on forest ecosystem structure, function and health. The monitoring at ECN sites started in 1993 with the objectives
Major ions in soil solution
Overall, MADOC predictions corresponded well with observed concentrations of major ions (Table 6). Observed declines in soil water SO4 were reproduced by MADOC (Fig. 3a), although at the beginning of the monitoring period SO4 concentrations were underestimated relative to observations at most sites, suggesting that anthropogenic S deposition was higher at this time than the extrapolated Eskdalemuir bulk deposition sequence would indicate. The model also reproduced the downward trends in Cl
Efficacy of the model in predicting concentrations of soil water indicators
The MADOC model reproduced observations during the monitoring period of major ion concentrations, pH and DOC well at most sites. The reproduction of SO4 and Cl trends indicates that the model captures adequately the processes governing longer term behaviour of these ions and their annual budgets. The soil water concentrations of sulphate reflected a marked and general decrease in S deposition, a pattern which was also shown in bulk deposition and soil water monitoring data from UK ECN and FLII
Conclusions
The MADOC model was able to reproduce changes in soil water DOC concentrations observed for a range of upland organic soil types, although performance was strongly dependent on deposition sequences, implying that good deposition estimates are essential for site-scale modelling. The application of MADOC to terrestrial monitoring data provides insight into the extent to which drivers other than sulphur deposition might contribute to DOC trends. According to the process understanding and
Acknowledgements
This research was funded by University of Reading, Centre for Ecology and Hydrology and Forest Research. Data were provided by Environmental Change Network and Forest Research. We would like to thank Lorna Sherrin, Sue Benham and Francois Bochereau and all sites managers for helping collating information on ECN and FLII data and sites.
References (108)
- et al.
Multi-scale relationship between peatland vegetation type and dissolved organic carbon concentration
Ecol. Eng.
(2012) - et al.
The importance of the relationship between scale and process in understanding long-term DOC dynamics
Sci. Total Environ.
(2010) - et al.
Short and long-term impacts of nitrogen deposition on carbon sequestration by forest ecosystems
Curr. Opin. Environ. Sustain.
(2014) - et al.
The impact of nitrogen deposition on carbon sequestration by European forests and heathlands
For. Ecol. Manag.
(2009) - et al.
The consequences of chronic nitrogen additions on n cycling and soilwater chemistry in a Sitka spruce stand, North Wales
For. Ecol. Manag.
(1998) - et al.
Long-term increases in surface water dissolved organic carbon: observations, possible causes and environmental impacts
Environ. Pollut.
(2005) - et al.
Long-term variability in the deposition of marine ions at west coast sites in the UK Acid Waters Monitoring Network: impacts on surface water chemistry and significance for trend determination
Sci. Total Environ.
(2001) - et al.
Nitrogen deposition in Swiss forests and its possible relevance for leaf nutrient status, parasite attacks and soil acidification
Environ. Pollut.
(1998) - et al.
Changes in the atmospheric deposition of acidifying compounds in the UK between 1986 and 2001
Environ. Pollut.
(2005) - et al.
Testing seasonal and long-term controls of streamwater DOC using empirical and process-based models
Sci. Total Environ.
(2008)