Reduced throughfall decreases autotrophic respiration, but not heterotrophic respiration in a dry temperate broadleaved evergreen forest
Introduction
Forests currently are a net carbon sink and the majority of global forest carbon is stored in soils (Ballantyne et al., 2012, Dixon et al., 1994, Pan et al., 2011). Soil respiration (RS) is the largest carbon flux from terrestrial ecosystems back to the atmosphere (Janssens et al., 2001) and consequently exerts a strong influence on the net global and net ecosystem carbon exchange (Valentini, 2000). Soil temperature and soil moisture exert a strong influence on RS (Davidson et al., 1998, Lloyd and Taylor, 1994, van’t Hoff, 1898). Hence, changes in RS in response to predicted climate change scenarios (IPCC, 2013) can have a huge impact on the strength and future direction of this forest soil carbon sink. RS increases with increasing temperature potentially leading to a positive carbon cycle-climate feedback under global warming (Bond-Lamberty and Thomson, 2010, Davidson and Janssens, 2006). However, Mediterranean-type forest ecosystems, such as the dry temperate broadleaved evergreen forests in south-eastern Australia, experience regular periods of drought and are strongly controlled by water availability. As such, soil moisture limitation of RS can exceed temperature sensitivity in these forest types (Reichstein et al., 2003, Rey et al., 2002). Moreover, in these ecosystems an increase in temperature and overall decrease in rainfall is predicted together with changes in rainfall frequency and intensity, as well prolonged drought periods (Christensen et al., 2007). Limited information exists on how changes in rainfall pattern will influence RS and the terrestrial carbon cycle in the long-term, although drought will likely lead to a decrease in RS (Rey et al., 2002) with pulses of summer rain leading to rapid and large emissions of CO2 from forest soils (Jarvis et al., 2007, Lee et al., 2004). Predicting the response of RS to changing environmental conditions is inherently difficult, especially as RS is the combined flux of heterotrophic respiration (RH) and belowground autotrophic respiration (RA).
While RH – decomposition of organic material by microorganisms – relies on the carbon input from fresh litterfall, root turnover and existing soil organic matter (SOM), RA – here referred to as respiration of living roots and heterotrophic respiration from closely associated microbes and mycorrhiza in the rhizosphere depending on rhizodeposits (Kuzyakov, 2006) – strongly depends on the allocation of recently assimilated photosynthates belowground (Ekblad and Högberg, 2001, Högberg et al., 2001, Ryan and Law, 2005). In forest ecosystems, RA contributes on average 50–60% (Subke et al., 2006) to RS but, depending on the season, can vary from 10% to 90% (Hanson et al., 2000). It has been shown that these processes can respond differently and inconsistently to changes in key environmental drivers (Trumbore, 2006), such as to soil temperature (Boone et al., 1998, Lavigne et al., 2003, Rey et al., 2002) and soil moisture (Borken et al., 2006, Burton et al., 1998, Scott-Denton et al., 2006). Most ecosystem-scale studies of soil respiration processes in response to changing soil temperature and/or soil moisture have been observational, i.e. environmental factors changing with the seasons (e.g. Carbone et al., 2011, Rey et al., 2002, Rühr and Buchmann, 2010). Interpreting responses based on seasonal changes alone is often difficult as soil temperature and soil moisture often covary (Davidson et al., 1998). Moreover, other factors such as litter quality, soil microbial biomass, plant physiological activity and plant internal carbon allocation (Brüggemann et al., 2011, Högberg and Read, 2006, Hopkins et al., 2013) can influence soil respiration processes and seasonal variation. This complexity highlights the need to better understand what drives the processes underlying forest RS. This also shows how difficult it is to determine fundamental relationships to reliably model these ecosystem processes and to predict forest soil responses to future climate scenarios.
In the Northern Hemisphere, summer droughts are predicted to increase in Mediterranean and temperate ecosystems. In contrast, in similar ecosystems of the Southern Hemisphere, such as in south-eastern Australia, rainfall is predicted to decrease mostly during autumn and winter months (Christensen et al., 2007, CSIRO, 2012). Although the future climate is expected to be warmer and drier, there is considerable uncertainty about the extent and timing of predicted changes in rainfall due to the natural climate variability in this region (CSIRO, 2012). Moreover, little is known about soil respiration dynamics in dry temperate broadleaved evergreen eucalypt forests. These forests lack a dormant season and exhibit opportunistic growth (Jacobs, 1955, Keith, 1997).
Ecosystem-scale studies that manipulate one or multiple environmental factors have the potential to disentangle and directly quantify the effect that different environmental factors have upon forest soil respiration processes, i.e. changes in throughfall pattern (Cleveland et al., 2010, Davidson et al., 2008, Matias et al., 2012, Schindlbacher et al., 2012, Sotta et al., 2007, Talmon et al., 2011). Only a few forest studies have combined manipulations of throughfall with partitioning RS into its component fluxes, but primarily for short time periods corresponding to particular growing seasons or to simulate total summer drought (Borken et al., 2006, Lavigne et al., 2004, Muhr and Borken, 2009, Schindlbacher et al., 2012). The total exclusion of throughfall does not allow pulses of increased soil moisture to occur which can have a substantial influence on soil respiration processes, particularly during dry periods (Jarvis et al., 2007). To gain a better understanding about the effect that a reduction of throughfall will have upon soil respiration dynamics, it is important to not only partition soil respiration processes but also to undertake continuous throughfall manipulation experiments, comprising entire growth cycles over all seasons to simulate a more realistic future climate scenario. To our knowledge, the only other studies quantifying soil respiration component fluxes to RS in forest soils under extended throughfall reduction have been in a tropical forest (da Costa et al., 2013, Metcalfe et al., 2010, Metcalfe et al., 2007).
In this study we aimed to: (1) partition and quantify the relative contribution of RH and RA to RS, (2) investigate the seasonal variation of soil respiration component fluxes, (3) determine the dependence of RS, RH and RA to soil temperature and soil moisture, and (4) quantify the impact of continuous throughfall reduction on RS, RH and RA in a dry temperate broadleaved evergreen forest.
Section snippets
Study site
The study site is located in the Wombat State Forest, Victoria, about 120 km west of Melbourne, Australia (37°25′20.5″S, 144°05′39.1″E) at a mean altitude of 706 m a.s.l. The dry sclerophyll eucalypt forest is dominated by three broadleaved evergreen tree species: Eucalyptus obliqua (L’Hérit.), Eucalyptus rubida (Deane & Maiden) and Eucalyptus radiata (Sieber ex DC). The study site is a ∼25 years old secondary regrowth forest with a tree height of 21–25 m (Table 1). Forest management includes
Seasonal variation in soil temperature, soil moisture and throughfall
Both surface soil temperature and soil moisture strongly varied with the seasons but showed an inverse relationship (Fig. 1b, c, f and g). Soil temperature was not different for RS and RH chambers (P > 0.7) in either the Control or TFR treatment, hence soil temperature values of RS and RH were combined into monthly means for each Control and TFR plot. Minimum soil temperature was 5.9 °C in August 2010, increasing to a maximum of 20.0 °C in February 2011 and then decreasing again to 6.5 °C in July
Partitioning of soil respiration
The limitations associated with the root exclusion method are primarily due to changes in soil properties following disturbance. Therefore, calculated component fluxes and their relative contributions to RS are estimates of natural conditions. Although other methods, i.e. isotope tracer/labelling techniques or stem phloem-girdling, disturb the soil and root system less, they can be complex or problematic in certain ecosystems (Binkley, 2006, Hanson et al., 2000, Kuzyakov, 2006). Moreover,
Conclusion
The partitioning of RS into its component fluxes RH and RA showed that RA was the main contributor to RS in the Control. However, seasonal changes were the result of very different and distinct seasonal patterns in RH and RA, and were mainly determined by indicated changes in RA. Soil temperature and soil moisture were insufficient to explain the seasonal pattern in RS, due to the contrasting dependencies of RH and RA on temperature and moisture. While soil temperature was a good predictor for
Acknowledgements
The study was supported by funding from the Terrestrial Ecosystem Research Network (TERN) Australian Supersite Network, the TERN OzFlux Network, the Australian Research Council (ARC) grants LE0882936 and DP120101735 and the Victorian Department of Environment and Primary Industries Integrated Forest Ecosystem Research program. We would like to thank Dr. Christina Schädel for her valuable comments to the manuscript. We also would like to thank the many internship students, especially from the
References (88)
- et al.
Temperature dependence of soil CO2 efflux is strongly modulated by seasonal patterns of moisture availability in a Mediterranean ecosystem
Soil Biol. Biochem.
(2009) - et al.
Microbial respiration and substrate weight loss. 1. General model of influences of abiotic variables
Soil Biol. Biochem.
(1977) - et al.
Soil CO2 efflux in a temperate deciduous forest: environmental drivers and component contributions
Soil Biol. Biochem.
(2010) - et al.
Towards a more plant physiological perspective on soil ecology
Trends Ecol. Evol.
(2006) - et al.
Fine root dynamics, soil moisture and soil carbon content in a Eucalyptus globulus plantation under different irrigation and fertilization regimes
For. Ecol. Manage.
(1995) Multiple measurements constrain estimates of net carbon exchange by a Eucalyptus forest
Agric. For. Meteorol.
(2009)- et al.
Dynamics of carbon exchange in a Eucalyptus forest in response to interacting disturbance factors
Agric. For. Meteorol.
(2012) Sources of CO2 efflux from soil and review of partitioning methods
Soil Biol. Biochem.
(2006)- et al.
Carbon cycling in a mountain ash forest: analysis of below ground respiration
Agric. For. Meteorol.
(2007) - et al.
Soil respiration fluxes in relation to photosynthetic activity in broad-leaf and needle-leaf forest stands
Agric. For. Meteorol.
(2008)