Will heterotrophic soil respiration be more sensitive to

0 downloads 0 Views 1MB Size Report
E-mail: yiqing@hawaii.edu, [email protected]. Running Title: Warming .... C.J.Chen, Mallotus lianus Croiz, Rhus chinensis Mill., Rubus corchorifolius Linn. f.,.
Will heterotrophic soil respiration be more sensitive to warming than autotrophic respiration in subtropical forests? X. F. LIUa,b, S. D. CHENa,b, Z. J. YANGa,b, C. F. LINa,b, D. C. XIONGa,b, W. S. LINa,b, C. XUa,b, G. S. CHENa,b, J. S. XIEa,b, Y. Q. LIb,c & Y. S. YANGa,b

a

Accepted Article

Key Laboratory for Subtropical Mountain Ecology, Fujian Normal University,

Fuzhou 350007, China, bCollege of Geographical Sciences, Fujian Normal University, Fuzhou 350007, China, and cCollege of Agriculture, Forestry and Natural Resource Management, University of Hawaii, Hilo, HI 96720, USA.

Correspondence: Y. Li, Y. Yang. E-mail: [email protected], [email protected]

Running Title: Warming effect on soil respiration

Summary Understanding the responses of heterotrophic (Rh) and autotrophic (Ra) components of soil respiration (Rs) to warming is important in evaluating and modelling the effects of changes in climate on soil carbon (C) cycling in terrestrial ecosystems. We used a mesocosm system with buried heating cables (5 °C warming) to investigate the responses of Rs, Rh and Ra to warming in a subtropical forest in southern China. Soil CO2 effluxes were measured with a portable automatic soil CO2 flux system from March 2014 to July 2015. We found that warming increased mean Rs and Rh from 788 to 1036 g C m−2 year-1 (+31%) and from 512 to 707 g C m−2 year-1 (+38%), respectively. There was no difference in Ra between the warming treatment and the control. The lack of response of Ra to warming was probably because the fine root biomass did not change with warming treatment. Soil warming also increased available dissolved organic carbon, microbial biomass carbon, actinomycetal biomass and arbuscular mycorrhizal biomass. Our results suggest that Rh might be more This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/ejss.12758

This article is protected by copyright. All rights reserved.

sensitive to climate warming than Ra, and future climate warming could increase soil C loss from increased Rh in subtropical forest ecosystems.

Accepted Article

Key words: Chinese-fir, climate warming, plantation, Q10, subtropical forest, soil respiration

Highlights 

A field warming experiment with partitioning of soil respiration in a humid subtropical forest.



Warming increased Rs and Rh without significantly altering soil microbial substrate availability.



Heterotrophic respiration appeared more sensitive to warming than autotrophic respiration.



Warming increased Actinomycetes bacteria and Arbuscular mycorrhizal fungi.

This article is protected by copyright. All rights reserved.

Introduction According to the Representative Concentration Pathways (RCPs) 8.5 scenario, global

Accepted Article

terrestrial mean surface temperature will probably increase by 2.6–4.8°C from the year 2000 to the end of this century (Field et al., 2014). Temperature increase could affect carbon (C) cycling in terrestrial ecosystems, and possibly cause a positive or negative feedback to future climate change (Davidson & Janssens, 2006; Carey et al., 2016). Such feedback is crucial to tropical and subtropical forests which contain approximately half of the total global C sink in established forests and exchange a considerable amount of CO2 with the atmosphere (Cavaleri et al., 2015). One of the most important mechanisms that affects such feedback is soil respiration. Soil respiration (Rs) is the second largest source of CO2 emission from terrestrial ecosystems into the atmosphere (Kuzyakov, 2006). It comprises heterotrophic respiration by soil microorganisms and fauna (Rh) and autotrophic respiration by plant roots (Ra) (Kuzyakov, 2006). Numerous studies have indicated that warming alters Rs considerably, but its responses to warming in different forest ecosystems have been contradictory (Bronson et al., 2008; Melillo et al., 2011; Liu et al., 2016). For example, warming increased Rs in a temperate mixed hardwood forest ecosystem in the United States of America (Melillo et al., 2011), but reduced it in a boreal black spruce forest in Canada (Bronson et al., 2008). Such an inconsistency might partly be due to different responses of Ra and Rh to warming of forest ecosystems at different latitudes (Wang et al., 2014). Therefore, understanding the responses of Ra and Rh from forest soils under different climate zones to warming is important for predicting

This article is protected by copyright. All rights reserved.

the global C cycle accurately under future potentially warmer climate conditions. During the past two decades, a large body of warming experiments has been

Accepted Article

carried out on Rs in various terrestrial ecosystems (Luo et al., 2001; Melillo et al., 2011; Liu et al., 2016), but few studies have focused on the individual response of Rh or Ra to warming (Schindlbacher et al., 2009; Noh et al., 2016). Within the few studies available, however, the responses of Rh and Ra to warming were markedly inconsistent. For example, experimental warming increased both Rh and Ra in a cool-temperate deciduous forest ecosystem in northern Japan, which was attributed to the increase in root litter inputs (Noh et al., 2016). In contrast, experimental warming increased Rh but reduced Ra in a deciduous forest in New England (USA), probably because of a decrease in fine root biomass (Melillo et al., 2011). A previous study with deconvolution analysis indicated that warming can decrease both Rh and Ra, largely as a result of reduced belowground plant biomass (Zhou et al., 2010). At present, there remains a lack of information on underlying mechanisms responsible for the warming-induced changes in the components of Rs and thus it limits our understanding on how will Rs respond to future climate warming. Several previous studies found that warming could directly increase both aboveand below-ground plant production, particularly in low-temperature areas (Rustad et al., 2001; Melillo et al., 2011). Warming might also increase soil organic C and the rate of litter decomposition, resulting in an increase in Rh (Schindlbacher et al., 2009). Warming could markedly change microbial community structures in soil (DeAngelis et al., 2015) and trigger a shift in the use of microbial substrate (Streit et al., 2014). In

This article is protected by copyright. All rights reserved.

addition, warming could have profound effects on Rh and Ra by changing soil moisture or microbial biomass carbon (MBC) (Heinemeyer et al., 2012;

Accepted Article

Romero-Olivares et al., 2017). Although previous studies in temperate or boreal regions provided valuable information for understanding the mechanisms of Rh and Ra responses to temperature increase, there remains uncertainty about how these research results could be extrapolated to tropical and subtropical forest ecosystems. In the absence of field warming experiments in the tropics and subtropics, laboratory incubation experiments have suggested that soil microbes had higher thermal optima in the tropics and subtropics than in temperate or boreal regions (Balser & Wixon, 2009). Temperatures in the tropics and subtropics have been reported as being close to the high-temperature threshold for plant growth (Doughty & Goulden, 2008). Therefore, climate warming in these regions could reduce plant photosynthesis or even cause plant death (Doughty & Goulden, 2008; Cheesman & Winter, 2013). These studies indicated that the responses of Rh and Ra to climate warming in tropical and subtropical forests could be very different from those in temperate or boreal forests. The lack of information on the responses of Rh and Ra to warming in tropical and subtropical forests limits our ability to predict soil C dynamics under future climate scenarios at a global scale. Therefore, there is an urgent need for field warming experiments in tropical and subtropical forests (Cavaleri et al., 2015). In this study, we did a field soil warming (+5°C) experiment with the heating cable technique. From previous observations and findings, we addressed the following

This article is protected by copyright. All rights reserved.

questions: (i) how will Rs and its components (Rh and Ra) respond to warming in a humid subtropical forest? and (ii) will Rh or Ra dominate CO2 emissions from

Accepted Article

subtropical forest soils under warming?

Materials and methods Study site

The experiment was carried out at the Chenda Observation site (26°20′3″ N, 117° 27′22″ E, 300 m a.s.l.), Sanming Research Station of Forest Ecosystem and Global Change, Fujian province, China. The mean annual precipitation is approximately 1656 mm, >70% of the precipitation occurs from March to August (1959–2015). The mean annual air temperature is 19.1°C and the relative humidity averages 80%. The study site is under a typical subtropical monsoon climate and its soil is classified as red soil based on the China’s soil classification systems, equivalent to Oxisol in the USDA Soil Taxonomy (State Soil Survey Service of China, 1998; Soil Survey Staff of USDA, 2014). The basic physicochemical characteristics of the soils at the 0–10-cm depth at the study site are given in Table 1. The dominant plant species in the research area were Trema cannabina Lour. var. dielsiana (Hand.-Mazz.) C.J.Chen, Mallotus lianus Croiz, Rhus chinensis Mill., Rubus corchorifolius Linn. f., Litsea cubeba (Lour.) Pers. and Aralia chinensis Linn.

Experimental design

This article is protected by copyright. All rights reserved.

The experiment included three warming and three control plots. Each plot had a size

Accepted Article

of 2 m × 2 m (Figure 1). They were established in August of 2013. Heating cables (Nexans, TXLP/1, Oslo, Norway) were buried in both control and warmed soils at a depth of 10 cm with a horizontal interval of 20 cm; the cables buried in the control plots were not heated. Each plot was surrounded by a 2-m wide buffer strip with PVC boards (2 m × 0.7 m × 0.7 mm) to prevent interference from adjacent plots. Temperature sensors (T109, Campbell Scientific Inc., Logan, UT, USA) were placed between two cables at a depth of 10 cm, three in each warming plot and two in each control plot. The soil temperature in the warming plots was maintained continuously at 5°C above the temperature in the control plots using thermocouples and a data-logger (CR1000; Campbell Scientific Inc., North Logan, UT, USA). Soil moisture in each plot was measured with two ECH2O-5 soil moisture probes (Decagon, Pullman, WA, USA) placed between cable lines in the soil at a depth of 10 cm.

Soil respiration measurements

Heterotrophic respiration was measured by the trenching method before heating cables were buried. One subplot (40 cm × 40 cm) was established at each plot (Figure 1). The subplots were prepared by trenching along the boundaries to a depth of 60 cm. Nylon mesh sheets (100 mesh, Sefar, Heiden, Switzerland) were inserted into the

This article is protected by copyright. All rights reserved.

vertical profiles to prevent root growth into the subplots, but they allowed air and water to pass through. To minimize the transient response caused by the

Accepted Article

decomposition of dead roots, soil respiration measurements were not made until 6 months after the trenching work. In November 2013, PVC collars (20 cm in diameter) were inserted into the soil at a depth of 4 cm for Rs and Rh measurements. Measurements of soil respiration were carried out over a 17-month period from March 2014 to July 2015. Soil respiration (Rs) and Rh in each collar was measured with a portable LI-8100 automatic soil CO2 flux system (LiCOR Inc., Lincoln, NE, USA) at a biweekly interval. At the same time, soil temperature at the 5-cm depth was measured with a long-handled thermometer (Model SK-250WP/SWP-04, Sato Keiryoki MFG, Tokyo, Japan) and soil moisture (0–12 cm) was monitored by time-domain reflectometry (Model TDR300, Spectrum Technologies Inc., Plainfield, IL, USA).

Soil sampling and analysis

Soil was sampled for chemical and physical property analyses in April of 2015, one year after the experiment was established. Five soil cores (2 cm in diameter) were taken at a depth of 0–10 cm from each plot. Soil samples were stored immediately in an icebox and then transferred to the laboratory within 2 hours. Each sample was separated into two subsamples: one subsample was stored at 4ºC in a refrigerator (< 1 week) for analyses of MBC, dissolved organic carbon (DOC), NH4+ and NO3-. The

This article is protected by copyright. All rights reserved.

other subsample was stored at –20°C in a fridge for analysis of soil microbial community structures. The soil sample was passed through a 2-mm mesh sieve before

Accepted Article

chemical analyses. Soil moisture was determined by oven-drying at 105°C for 24 hours and then measured gravimetrically. DOC was extracted in deionized water at 20°C by the following steps (Jones & Willett, 2006). A 10-g field fresh soil was placed in 40 ml distilled water and then shaken for 1 hour with an end-to-end shaker. The mixture was then centrifuged at 1369 g for 20 minutes, before being filtered through a 0.45-μm filter membrane. The organic C concentrations in the water extracts were measured using a total organic carbon analyser (TOC-VCPH/CPN, Shimadzu Corporation, Kyoto, Japan). Soil microbial biomass was measured by the fumigation extraction method (Vance et al., 1987). A 25-g fresh soil sample was fumigated with chloroform for 24 hours and then extracted with 100 ml 0.5 M K2SO4. The samples were shaken for 30 minutes and then filtered through a 0.45-μm membrane filter. The non-fumigated soil samples were also extracted following the same procedure. These filtrates were analysed for organic C using a TOC auto-analyser (TOC-VCPH/CPN, Shimadzu Corporation, Kyoto, Japan). Microbial biomass C was determined from the difference between the C extracted before and after fumigation using a conversion factor of 0.45 (Jenkinson et al., 2004). For the nitrate and ammonium analyses, five grams of fresh soil were extracted with a 2 mol l-1 KCl solution (Carter & Gregorich, 2008). The solutions were shaken for 40 minutes and then filtered for nitrate and ammonium determination using a Continuous Flow Analyzer (Skalar San ++, Breda,

This article is protected by copyright. All rights reserved.

Netherlands). For fine root sampling, ten soil cores were taken randomly in each subplot with a

Accepted Article

soil corer (3.5 cm in diameter) at 0–10-cm depth in April 2015. The soil samples were washed with deionized water to remove soil and organic debris that adhered, and then wet-sieved with a mesh size of 0.5 mm. The sieved samples were placed in deionized water at 1ºC and stirred repeatedly to float the fine root (