- Email: [email protected]
A new incubation and measurement approach to estimate the temperature response of soil organic matter decomposition
Soil Biology and Biochemistry 138 (2019) 107596
Contents lists available at ScienceDirect
Soil Biology and Biochemistry journal homepage: www.elsevier.com/locate/soilbio
A new incubation and measurement approach to estimate the temperature response of soil organic matter decomposition
T
Yuan Liua,b, Nianpeng Hea,b,c,∗, Li Xua, Jing Tiana, Yang Gaoa,b, Shuai Zhengd, Qing Wanga, Xuefa Wena,b, Xingliang Xua, Kuzyakov Yakove,f,g a Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China b College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China c Institute of Grassland Science, Northeast Normal University, and Key Laboratory of Vegetation Ecology, Ministry of Education, Changchun, 130024, China d State Environmental Protection Key Laboratory of Wetland Ecology and Vegetation Restoration, Northeast Normal University, Changchun, 130117, China e Department of Soil Science of Temperate Ecosystems, Department of Agricultural Soil Science, University of Goettingen, 37077, Göttingen, Germany f Agro-Technology Institute, RUDN University, Moscow, Russia g Institute of Environmental Sciences, Kazan Federal University, 420049, Kazan, Russia
A R T I C LE I N FO
A B S T R A C T
Keywords: Microbial adaption Substrates depletion Soil organic matter decomposition Temperature sensitivity
A reliable and precise estimate of the temperature sensitivity (Q10) of soil organic matter (SOM) decomposition is critical to predict feedbacks between the global carbon (C) cycle and climate change. In this study, we first summarize two commonly used approaches for estimating Q10 (Approach A: constant temperature incubation and discontinuous measurements, CDM model; Approach B: varying temperature incubation and discontinuous measurements, VDM model). We then introduced a newly developed approach (Approach C, VCM model) that combines rapidly varying temperature incubations and continuous measurements of SOM decomposition rates (Rs) that may be more realistic and suitable for Q10 estimation, especially for large scale estimation. Then, we conducted a 26-day incubation experiment using three different soils to compare the performance of these three approaches for estimating Q10 using R2 and P-values as indicators. Our results demonstrate that the fitting goodness of the exponential model was consistently higher for Approach C, with higher R2 values, lower confidence intervals, and lower P-values in almost all cases compared with Approaches A and B. Furthermore, results showed that Approaches A and B underestimated the Q10 value by 9.5–13% and 2.9–5.7%, respectively, in three different soils throughout the entire incubation period. Compared with traditional commonly used methods, the newly developed Approach C (VCM model) provides a more accurate and rapid estimation of the temperature response of SOM decomposition and can be used for large-scale estimation of Q10.
1. Introduction The decomposition rate (Rs) of soil organic matter (SOM) is highly sensitive to changes in temperature (Fang and Moncrieff, 2001). Therefore, the temperature response of the Rs under various global warming scenarios has received considerable attention in recent decades (Cox et al., 2000; Knorr et al., 2005; Liu et al., 2017). Given that soil stores enormous global organic carbon (C), minor changes in Rs will influence atmospheric carbon dioxide (CO2) concentrations and result in strong feedback between climate change and terrestrial C pools (Conant et al., 2011). The response of Rs to changing temperatures is referred to as temperature sensitivity (Q10), which is defined as the
factor by which Rs increases with each 10 °C increase in temperature and is usually calculated by the assumed exponential relationship between Rs and temperature (Kirschbaum, 1995; Davidson and Janssens, 2006). The Q10 parameter is important and is widely used in ecological models to predict changes in SOM decomposition under various temperature scenarios (Foereid et al., 2014). Therefore, a reliable and accurate estimate of Q10 is critical to predict the amplitude and direction of feedbacks between the global C cycle and climate change (Cox et al., 2000; Friedlingstein et al., 2006). Over the last few decades, researchers have conducted numerous laboratory experiments to estimate the Q10 of SOM decomposition using different incubation and measurement approaches (Constant incubation
∗ Corresponding author. Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China. E-mail address: [email protected] (N. He).
https://doi.org/10.1016/j.soilbio.2019.107596 Received 7 May 2019; Received in revised form 5 September 2019; Accepted 7 September 2019 Available online 07 September 2019 0038-0717/ © 2019 Elsevier Ltd. All rights reserved.
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
are incubated at several discrete constant temperatures (e.g., 5, 10, 15, 20, and 25 °C), and Rs is measured in intervals of days, weeks, or months throughout the experimental period (Knorr et al., 2005; Conant et al., 2011). The SOM decomposition rate is measured by multiple methods, including alkali absorption, gas chromatography, or CO2 infrared spectroscopy. Subsequently, Q10 is commonly calculated based on the assumed exponential relationship between Rs and soil temperature. The advantage of Approach A is that it is simple (just need alkali absorption) and there is no need for complex equipment (such as continuous temperature regulating system and CO2 monitoring system), making it easy for most laboratories to implement. This approach has helped us have a better understanding of the temperature response of SOM decomposition (Bradford et al., 2008; Conant et al., 2008) and the feedback between climate change and terrestrial C pools. Despite the obvious advantages, this approach has several inevitable shortcomings (Table 1). The first one was the unevenly substrate depletion among different temperatures. Generally, soils that was incubated at higher temperatures might consume more substrate than those incubated at lower temperatures, which will result in relative lower Rs at higher temperatures. These issues strongly affect the accuracy of Q10 estimation, especially in long-term incubation experiments (Conant et al., 2008; Fissore et al., 2009). Second, long-term incubation at different but constant temperatures commonly leads to a certain degree of microbial adaption (Bradford et al., 2008; Bradford, 2013), which might influence the accuracy of estimated Q10 values. This may be realistic for deeper soils because deeper soils experience much less temperature variation than surface soils due to heat transfer. However, surface soil usually experiences large temperature variations due to direct solar radiation. Third, the temperature response of Rs was commonly determined using 3–5 temperatures and these lower number of separated temperature points is not statistically adequate to accurately estimate the relationship between Rs and temperature. The classically adopted method using the exponential equation requires a high frequency of temperature-Rs measurements to increase the accuracy of Q10 estimation (Conant et al., 2011). To overcome the issue of low frequency measurement, a group of New Zealand scientists recently developed a new incubation system that can incubate soils along a temperature gradient (2–50 °C) with 44 discrete temperatures and simultaneously measure the soil respiration rate (Robinson et al., 2017). In any case, improving the fitting accuracy with more temperature measurements is only useful for short-time incubations and still cannot overcome the issue of uneven substrate depletion under different temperature conditions in long-time incubations. Finally, it is difficult to use Approach A in experiments with large scale soil samples because this approach is time consuming and labor intensive and consequently hinders the ability to explore the spatial variation in Q10 using a unified methodology.
or sequential incubation method) (Fang and Moncrieff, 2001; Conant et al., 2011; Ding et al., 2016) and helped us have a deeper understanding of the temperature response of SOM decomposition. However, diverse approaches make it difficult to interpret results from different studies and might result in inaccuracy and uncertainty in Q10 estimation (Liang et al., 2015). The selection between different approaches depends on the research question (Conant et al., 2011; Meyer et al., 2018). In the laboratory, large Q10 variation is mainly derived from changes in substrate quality and quantity during incubation at different temperatures. Commonly, soils (whether surface or deep soil) tend to be incubated at several discrete but constant temperatures (often 3–5 temperatures; Conant et al., 2011). Under this condition, substrates are always consumed faster and more intensely at higher temperatures during long-term incubation, which may lead to an underestimation of the Q10 value in the final stage of incubation (Liang et al., 2015). Commonly, the temperature response of SOM decomposition is estimated using either the Arrhenius equation (Arrhenius, 1889; Davidson and Janssens, 2006) or the exponential equation (Gershenson et al., 2009; Bracho et al., 2016). Below optimum temperature, Q10 is commonly calculated based on the assumed exponential relationship between Rs and temperature (Liu et al., 2018a). Limited data points derived from those discrete temperatures (3–5 temperature points) may decrease the fitting goodness and increase the fitting errors of the temperature responses of SOM decomposition, which may affect the accuracy of the estimated Q10 value (Chen et al., 2010; Robinson et al., 2017). Robinson et al. (2017) suggested that confidence in parameter fits requires about 20 evenly distributed incubation temperatures. Furthermore, fitting the temperature response curve of soil respiration with several more temperature measurements provides a better estimation of Q10 than estimating Q10 with data at two or three temperatures (Chen et al., 2010). Laboratory incubation experiments have improved our understanding of the temperature response of SOM decomposition in different sites and ecosystems (Fang and Moncrieff, 2001; Gershenson et al., 2009; Liang et al., 2015; Ding et al., 2016; Liu et al., 2017). However, a systematic comparison of the effect of different incubation and measurement approaches on estimated Q10 values is still lacking, making it difficult to compare results from different studies using different methods, which consequently impedes our ability to accurately predict large-scale temperature responses of SOM decomposition (Meyer et al., 2018). In this study, we first summarize the advantages and disadvantages of two commonly used incubation and measurement approaches to estimated Q10 values. Then, we suggest a newly developed approach to conveniently estimate Q10 with greater accuracy. Finally, we conducted a 26-day laboratory incubation experiment using three different soils to compare the performance of these three approaches for estimating Q10 using R2 and P values as indicators. 2. Current approaches used to estimate and calculate Q10
2.2. Approach B: varying temperature incubation with discontinuous measurements (VDM model)
Feasible incubation and measurement approaches are essential for obtaining an accurate estimate of Q10. Thus, it is important to consider the factors that could potentially affect Q10 estimation (e.g., SOC component, microbial biomass and species composition, incubation time) and the operability of the measuring system. We first summarize two commonly used approaches from previous studies for estimating Q10.
With advance in science, Approach B was proposed to overcome the shortcomings of substrate depletion and microbial adaption associated with Approach A. Approach B involves varying temperature incubation and discontinuous measurements of Rs (Fig. 1 B). In brief, 5 or 6 discrete constant temperatures (e.g., 5, 10, 15, 20, and 25 °C) are gradually increased and then decreased in a stepwise manner (e.g., 5–10–15–20–25–20–15–10–5 °C). Subsequently, the corresponding Rs at a specific and constant temperature is measured after stabilizing the temperature for 30 min to several hours. After that, Q10 is estimated based on the corresponding exponential relationship between Rs and temperature (Fang et al., 2005; Ding et al., 2016). In practice, all samples are pre-incubated at a constant temperature (e.g., 20 °C) for three to seven days to avoid the pulse effect of disturbance and rewetting.
2.1. Approach A: constant temperature incubation with discontinuous measurements (CDM model) Approach A is a commonly used method in recent decades that combines constant incubation temperatures and discontinuous Rs measurement to estimate Q10 (Fig. 1). In practice, all samples are preincubated at a constant temperature (e.g., 20 °C) for three to seven days to avoid the pulse effect of disturbance and wetting. Then, soil samples 2
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 1. Three incubation and measurement approaches used to investigate how the soil organic matter (SOM) decomposition responds to changing temperature. Approach A: constant temperature incubation and discontinuous measurements. Approach B: varying temperature incubation and discontinuous measurements. Approach C: a newly developed varying temperature incubation and continuous measurements with higher frequency measurement (36 times measurement during 12 h) of SOM decomposition that may overcome the shortcomings of the limited data points associated with Approaches A and B. Table 1 Comparison of the three approaches (using different incubation and measurement method) to estimate the temperature sensitivity (Q10) of SOM decomposition. Brief description of each approach
Advantagesa
Disadvantages
Approach A (CDM model)
Constant temperature incubation and discontinuous measurement
a) Simple, economic, and operable
Approach B (VDM model)
Step-by-step varying temperature incubation and discontinuous measurement
Approach C (VCM model)
Varying temperature incubation and continuous measurement
a) Operable b) Overcome microbial adaption to constant temperature c) Overcome different substrate depletion at different temperature a) Overcome microbial adaption to constant temperature b) Overcome different substrate depletion c) Higher measurement frequency (at minutes intervals with changing temperature) d) Automatic, rapid and continuous, time-saving
a) Microbial adaption to constant temperature b) Different substrate depletion under different temperature c) Lower measurement frequency (commonly 3–5 temperature gradients) d) Time-consuming and laborious a) Higher initial investment b) Lower measurement frequency (commonly 3–5 temperature gradients) c) Complex operation procedure d) Time-consuming and laborious a) Higher initial investment b) No commercial equipment in market
a Advantages and disadvantages of each approach are summarized in view of the factors that influence the temperature response of soil organic matter decomposition (microbial adaption, substrate depletion) during incubation, and the operability of each approach.
Approach B partially overcomes the issues of substrate depletion and microbial adaption that occur using Approach A (Table 1). In Approach B, soil samples with the same incubation temperatures are subject to similar levels of substrate depletion with a short-term stabilization processes (30 min to several hours) to prevent microbial adaption. However, Approach B appears to be suitable for short-term incubation experiments, or soils with abundant substrate, but is not appropriate for long-term incubation experiments or soils with poor
substrate. Furthermore, CO2 concentrations in incubated bottles are commonly measured using alkali absorption, gas chromatography, or CO2 infrared spectroscopy (Fang et al., 2005; Xu et al., 2012; Ding et al., 2016), which limit the experimental scale and the frequency at which Rs is measured. Like Approach A, Approach B is still not statistically adequate to simulate the relationship between Rs and temperature, making it difficult to obtain accurate Q10 values (Robinson et al., 2017).
3
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
together to prevent soil hardening, adjusted to 60% water holding capacity (WHC), and placed in a 150-mL polyethylene plastic bottle. The soil samples were pre-incubated at 25 °C for 7 days, to avoid any pulse effect on microbial activities (Liu et al., 2017). All sample bottles were sealed with preservative films to reduce water loss, and several small holes were punched for ventilation. After 1, 5, 8, 15, 22, and 26 days incubation, SOM decomposition rates (Rs) in each bottle were measured using a new PRI-8800 Automatic Temperature Control Soil Flux System (PRI-8800; Pre-Eco, Beijing, China) referred to in previous studies (He et al., 2013; Liu et al., 2017). This system is an open system with nitrogen gas continuously passing the incubation bottles to minimize the inhibitory effects of high CO2 concentrations on SOM decomposition rates. Furthermore, this system enabled us to continuously vary the incubation temperature and measure Rs at high frequency (Rs was measured based on the linear relationship with time every 75 s). For Approach A and Approach B, each soil sample was first stabilized for 30 min at each temperature and then measured for three rounds, and average values of the three rounds were used for each replicate. For Approach C, each soil sample was stabilized for 30 min and each soil sample was then measured 36 times over a 12 h measuring period. In brief, an electric water bath controlled by an automatic temperature regulator (Julabo, Seelbach, Ortenau, Germany) was connected to a LiCOR CO2 analyzer (Li-7100, LI-COR, Lincoln, Nebraska, USA), which recorded the CO2 concentration every second. Soil temperatures in each bottle were simultaneously measured using a button thermometer (iButton® DS, 1922L; Maxim Integrated, Dallas, USA), which recorded the soil temperature every minute to provide accurate paired data of Rs and soil temperature. Soil moisture was monitored by weighing and adjusted by adding deionized water. Finally, we calculated the Q10 value with the classic exponential equation, the R2 and P values for each approach, and compared the performance of these three different approaches among three different soils (He et al., 2013; He and Yu, 2016; Wang et al., 2016b; Liu et al., 2017). The temperature response curve of SOM decomposition measured with these three approaches at different incubation times for three different soils were shown in Figs. 2–4. Using a repeated measures ANOVA, our results showed that measurement approach, incubation time, and soil types significantly influenced the estimation of Q10 (Table 2). The average Q10 in Approach A, B, and C was 1.43, 1.60, and 1.65, respectively in the LS site; 1.48, 1.55, and 1.62, respectively, in the DL site; and 1.29, 1.34, and 1.42, respectively, in the DH site. Overall, Q10 in Approach A was significantly lower than that in Approach C for all three soils (Fig. 5). Q10 in Approach B was significantly lower than that Approach C in the DL and DH sites, but the difference was not significant in the LS site (Fig. 5). Compared with Approach C, Approaches A and B underestimated Q10 by 13% and 2.9% in LS site, 8.8% and 4.2% in DL site, and 9.5% and 5.7% in DH site, respectively. Furthermore, Rs and Q10 gradually decreased with prolonged incubation times (Figs. 2–5) as did R2 in the exponential equations. The goodness of fit of the exponential model of Approach C was consistently higher than that of Approaches A and B, with a smaller confidence interval, higher R2, and lower P-values regardless of short-term or longterm incubation (Figs. 2–5). SOM decomposition rate decreased greatly for higher temperatures (Figs. 2–4) over 26-day incubation. Long-term incubation might cause a higher substrate depletion at high temperature relative to low temperature for Approach A, resulting in a weak response of Rs to increasing temperature and leading to lower Q10 estimations. By overcoming the shortcomings of substrate depletion and microbial adaption in Approach A, Approach B performed better than Approach A (Approach B had a higher R2 value, a lower confidence interval, lower P-values, and higher Q10 values than Approach A, Figs. 2–6). However, the low frequency of measurement in Approach B significantly increased the P-value of the model, especially in soils with a poor substrate (e.g. DH site; Fig. 4). As expected, Approach C performed better over the whole 26-day incubation period and suggests that traditional and commonly used Approaches A and B underestimate
2.3. Approach C: a newly developed approach combines varying temperature incubation with continuous measurement (VCM model) To overcome the shortcomings of the low frequency measurements of Approaches A and B, incubation and measurement approaches have recently been improved (Fig. 1 C). Based on the work by Cheng and Virginia (1993) with modification of temperature regulating system, scientists have recently developed a system to incubate soils under a varying temperature system and automatically and rapidly measure soil respiration rates with high frequency (up to 36 discrete temperature measurements during 12 h at an empirical setting) to accurately estimate Q10 (Approach C, Fig. 1; He et al., 2013). Recently, many studies have applied this approach to accurately estimate the temperature response of soil respiration (Wang et al., 2016a, 2016b, 2018b; Li et al., 2017; Liu et al., 2017, 2018a, 2018b). Approach C fully incorporates the basic theories and calculation methods of Q10 (Table 1). First, it continually varies the incubation temperature for all samples during the entire incubation period. In this way, Approach C overcomes the main shortcomings of Approach A, such as microbial adaptation to a specific constant temperature as well as different substrate depletion at different constant temperatures (Wang et al., 2016b; Liu et al., 2017). In practice, Approach C uses the same soil sample under different incubation temperatures. Furthermore, it can more precisely simulate the daily temperature fluctuations observed in the field, which can overcome microbial adaption to some extent. In addition, substrate depletion may be less significant during short-term periods (for several hours or a day; Liang et al., 2015). Second, this approach has the obvious advantage of rapidly and automatically measuring CO2 concentrations for every second and calculating Rs on a minute scale. For 12 h measurements, this approach can achieve 36 measurements of SOM decomposition at continually different temperatures. In this way, this approach can overcome shortcomings associated with a low frequency of measurement in Approach A and Approach B. Consequently, the estimation accuracy of Q10 is enhanced by increasing the frequency of measurement (Li et al., 2017). The drawback of Approach C is the requirement for professional experimental equipment, which might limit its application (Table 1). Nonetheless, many studies have employed this approach to estimate the temperature response of SOM decomposition in different sites and ecosystems (Wang et al., 2016b; Li et al., 2017; Liu et al., 2017). With the continuous development of the equipment, Approach C may be well-placed to estimate the temperature response of SOM decomposition in various sites and ecosystems and to predict feedback between the global C cycle and climate change (Fierer et al., 2006; Craine et al., 2010; Meyer et al., 2018). 3. Comparison of these three approaches in Q10 estimation by laboratory experiments To compare the performance of these three approaches, we conducted a laboratory incubation experiment using three different soils along a climate gradient from southern to northern China, including a subtropical evergreen broad-leaved forest soil on Dinghu Mountain (DH), a warm temperate coniferous broad-leaved forest soil on Dongling Mountain (DH), and a cold temperate coniferous forest soil in Liangshui (LS; Table S2). Detailed information regarding soil and vegetation have been described in Wang et al. (2016b). In brief, each forest soil sample was divided into three parts: the first part was subjected to Approach A (10, 15, 20, 25, and 30 °C constant temperature incubation and discontinuous measurement), the second part was subjected to Approach B (10–30 °C varying temperature incubation and 10, 15, 20, 25, and 30 °C constant temperature measurement), and the third part was subjected to Approach C (10–30 °C varying temperature incubation and continuous measurement; see Table S1 for details of the experimental design). In practice, each approach had five repetitions. In each replicate, 40 g of fresh soil and 10 g quartz sand were mixed 4
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 2. Comparison of three approaches for measuring the temperature response of SOM decomposition in cold temperate forest soil in Liangshui (LS). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5).
4. Application and prospects of the newly developed Approach C (VCM model)
Q10 values. Since Q10 is an important parameter for the prediction of the response of SOM decomposition to climate warming, a slight change in Q10 will result in large uncertainty when estimating the amplitude and direction of the feedback between the global C cycle and climate change. Therefore, future studies should be cautious using Approaches A and B to estimate Q10.
Approach C overcomes the shortcomings of traditional incubation experiments for estimating Q10 and has obvious advantages over previous approaches in terms of adjustable incubation temperatures and rapid automatic measurements (Table 1). Furthermore, this approach can also be applied in innovative studies of Q10. For example, this approach allows us to simulate field temperature change more precisely with continuous measurements for soil respiration and allows us to explore
Fig. 3. Comparison of the three approaches for measuring the temperature response of SOM decomposition in warm temperate forest soil in Dongling Mountain (DL). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5). 5
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 4. Comparison of the three approaches for measuring the temperature response of SOM decomposition in subtropical forest soil in Dinghu Mountain (DH). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5).
on soil microbial respiration occur rapidly (at minute scales), and it has been proved difficult to capture the quick response curves after a rainfall pulse using traditional approaches (at hours scale; Wang et al., 2016a; Song et al., 2017). For example, precipitation pulses commonly occur in arid or semi-arid regions; however, variation in soil microbial activity (e.g., soil C mineralization rate) in response to soil water availability is poorly understood (Wang et al., 2016a). Different precipitation levels and soil drying intensities might generate varied response times and intensities. Once a response appears in minutes, it is difficult to observe the actual response and conduct a series of experiments at the same time using traditional equipment. Similar limitations occur in other pulse experiments. Theoretically, differences in estimated Q10 values are inevitable among the different approaches. We believe that as the required technology and methodologies improve, the improved approach (Approach C, VCM model) will have the potential to overcome the shortcomings of previous approaches. However, it is necessary to further identify differences in Q10 estimates among the different approaches using a series of different soils, although Approach C has clearly outperformed the other two approaches in this study. Ideally, future studies should integrate the Q10 values documented by the different approaches to quantify the transformation factors among different approaches through a systematic comparison experiment along a soil substrate gradient. Furthermore, the choice between different approaches depends on the research question. Deeper soils experience much less temperature variation than surface soils due to heat transfer. Therefore, the newly developed method (VCM model) may indeed be better for shallow soils, but the constant-temperature method may be more realistic for deeper soils.
Table 2 Repeated measures analysis of variance of Q10 and R2 for the effect of Approaches (three approaches), Incubation times (six incubation time), Soils (three soils), and their interactions. df
Approach Time Soil Approach × Time Approach × Soil Soil × Time Approach × Time × Soil
2 5 2 10 4 10 20
R2
Q10 F
P
F
P
135.83 82.29 278.01 9.45 6.29 4 2.38
< 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01
50.06 10.4 70.25 4.92 2.204 2.36 1.81
< 0.01 < 0.01 < 0.01 < 0.05 0.07 < 0.01 < 0.01
df, degrees of freedom.
the underlying mechanisms controlling Q10. Such insights can improve the accuracy of extrapolating the results of laboratory incubations to the field. Owing to the limitations of traditional approaches, most existing studies have been conducted at hourly, daily, or weekly scales. Yet, such scales introduce inaccuracies when simulating diel temperature dynamics. Using Approach C, we explored the different responses of Rs to increasing and decreasing diel temperature dynamics in grassland (Li et al., 2017) and forest soils (Liu et al., 2018b; Wang et al., 2018a) and found significant differences in Q10 values between the increasing and decreasing phases. The advantages of Approach C (i.e., automatic, continuous, and rapid) could be further utilized to investigate the spatial variability in Q10 across sites or regions through a consistent network (He et al., 2013; Liu et al., 2017). Using Approach C, previous results showed that Q10 varies greatly among different ecosystems, with different factors dominating Q10 in the different ecosystems (Liu et al., 2017). In practice, it is difficult to integrate Q10 measurements from multiple studies due to considerable differences in the objectives, incubation processes, and measurement methods of the different studies. Approach C also provides the opportunity to explore and capture the pulse response of soil microbes to resources or water availability. Generally, the pulse effect of rainfall or resources (e.g. root exudates)
5. Conclusion In summary, we compared two commonly used approaches for estimating Q10 and then developed a convenient and accurate approach (Approach C, VCM model) that can be used as a reference method to estimate Q10 for Rs or soil organic matter decomposition studies. This study is the first to compare the performance of different approaches in estimating Q10 of SOM decomposition using a laboratory incubation 6
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 5. Changes in temperature sensitivity (Q10) with incubation time for the three approaches. Lowercase letters in each column indicate a significant difference between two values when P < 0.05. Values are mean ± SE (n = 5). Ave represent the average Q10 value over 26-day incubation. Liangshui (LS), Dongling (DL), Dinghu (DH).
Fig. 6. Changes in the R-square (R2) value of the fitted equations with incubation time for the three approaches. Lowercase letters in each column indicate a significant difference between two values when P < 0.05. Values are mean ± SE (n = 5). Ave represent the average R2 value over the 26-day incubation. Liangshui (LS), Dongling (DL), Dinghu (DH). 7
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
method with different soils. We found that the fitting goodness of the exponential model was consistently higher for Approach C, with higher R2 values, lower confidence intervals, and lower P-values compared with Approaches A and B, regardless of incubation time. For these three approaches, Q10 in three soils all gradually decreased with prolonged incubation times. In addition, Approaches A and B underestimated the Q10 value by 9.5–13% and 2.9–5.7%, respectively, in three different soils throughout the entire incubation period. Additionally, Approach C (VCM model) has the obvious advantages in terms of substrate depletion, microbial adaption, and operability (Table 1). Therefore, this new developed VCM model is a powerful method for rapidly monitoring soil microbial responses to changes in temperature, which consequently improving our ability to explore the mechanisms underlying SOM decomposition.
decomposition and feedbacks to climate change. Nature 440, 165–173. Ding, J.Z., Chen, L.Y., Zhang, B.B., Liu, L., Yang, G.B., Fang, K., Chen, Y.L., Li, F., Kou, D., Ji, C.J., Luo, Y.Q., Yang, Y.H., 2016. Linking temperature sensitivity of soil CO2 release to substrate, environmental and microbial properties across alpine ecosystems. Global Biogeochemical Cycles 30, 1310–1323. Fang, C., Moncrieff, J.B., 2001. The dependence of soil CO2 efflux on temperature Soil Biology and Biochemistry 33, 155–165. Fang, C.M., Smith, P., Moncrieff, J.B., Smith, J.U., 2005. Similar response of labile and resistant soil organic matter pools to changes in temperature. Nature 433, 57–59. Fierer, N., Colman, B.P., Schimel, J.P., Jackson, R.B., 2006. Predicting the temperature dependence of microbial respiration in soil: a continental-scale analysis. Global Biogeochemical Cycles 20, 1–10. Fissore, C., Giardina, C.P., Kolka, R.K., Trettin, C.C., 2009. Soil organic carbon quality in forested mineral wetlands at different mean annual temperature. Soil Biology and Biochemistry 41, 458–466. Foereid, B., Ward, D.S., Mahowald, N., Paterson, E., Lehmann, J., 2014. The sensitivity of carbon turnover in the Community Land Model to modified assumptions about soil processes. Earth System Dynamics 5, 211–221. Friedlingstein, P., Cox, P., Betts, R., Bopp, L., Von Bloh, W., Brovkin, V., Cadule, P., Doney, S., Eby, M., Fung, I., 2006. Climate-carbon cycle feedback analysis: results from the C4MIP model intercomparison. Journal of Climate 19, 3337–3353. Gershenson, A., Bader, N.E., Cheng, W.X., 2009. Effects of substrate availability on the temperature sensitivity of soil organic matter decomposition. Global Change Biology 15, 176–183. He, N., Yu, G., 2016. Stoichiometrical regulation of soil organic matter decomposition and its temperature sensitivity. Ecology and Evolution 6, 620–627. He, N.P., Wang, R.M., Gao, Y., Dai, J.Z., Wen, X.F., Yu, G.R., 2013. Changes in the temperature sensitivity of SOM decomposition with grassland succession: implications for soil C sequestration. Ecology and Evolution 3, 5045–5054. Kirschbaum, M.U.F., 1995. The temperature-dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry 27, 753–760. Knorr, W., Prentice, I., House, J., Holland, E., 2005. Long-term sensitivity of soil carbon turnover to warming. Nature 433, 298–301. Li, J., He, N., Xu, L., Chai, H., Liu, Y., Wang, D., Wang, L., Wei, X., Xue, J., Wen, X., Sun, X., 2017. Asymmetric responses of soil heterotrophic respiration to rising and decreasing temperatures. Soil Biology and Biochemistry 106, 18–27. Liang, J., Li, D., Shi, Z., Tiedje, J.M., Zhou, J., Schuur, E.A., Konstantinidis, K.T., Luo, Y., 2015. Methods for estimating temperature sensitivity of soil organic matter based on incubation data: a comparative evaluation. Soil Biology and Biochemistry 80, 127–135. Liu, Y., He, N., Wen, X., Xu, L., Sun, X., Yu, G., Liang, L.L., Schipper, L.A., 2018a. The optimum temperature of soil microbial respiration: patterns and controls. Soil Biology and Biochemistry 121, 35–42. Liu, Y., He, N., Zhu, J., Xu, L., Yu, G., Niu, S., Sun, X., Wen, X., 2017. Regional variation in the temperature sensitivity of soil organic matter decomposition in China's forests and grasslands. Global Change Biology 23, 3393–3402. Liu, Y., Wen, X., Zhang, Y., Tian, J., Gao, Y., Ostle, N.J., Niu, S., Chen, S., Sun, X., He, N., 2018b. Widespread asymmetric response of soil heterotrophic respiration to warming and cooling. The Science of the Total Environment 635, 423–431. Meyer, N., Welp, G., Amelung, W., 2018. The temperature sensitivity (Q10) of soil respiration: controlling factors and spatial prediction at regional scale based on environmental soil classes. Global Biogeochemical Cycles 32, 306–323. Robinson, J.M., O'Neill, T.A., Ryburn, J., Liang, L.L., Arcus, V.L., Schipper, L.A., 2017. Rapid laboratory measurement of the temperature dependence of soil respiration and application to changes in three diverse soils through the year. Biogeochemistry 133, 101–112. Song, X., Zhu, J., He, N., Huang, J., Tian, J., Zhao, X., Liu, Y., Wang, C., 2017. Asynchronous pulse responses of soil carbon and nitrogen mineralization to rewetting events at a short-term: regulation by microbes. Scientific Reports 7 7492. Wang, Q., He, N., Liu, Y., Li, M., Xu, L., 2016a. Strong pulse effects of precipitation events on soil microbial respiration in temperate forests. Geoderma 275, 67–73. Wang, Q., He, N., Liu, Y., Li, M., Xu, L., Zhou, X., 2018a. Important interaction of chemicals, microbial biomass and dissolved substrates in the diel hysteresis loop of soil heterotrophic respiration. Plant and Soil 428, 279–290. Wang, Q., He, N., Xu, L., Zhou, X., 2018b. Microbial properties regulate spatial variation in the differences in heterotrophic respiration and its temperature sensitivity between primary and secondary forests from tropical to cold-temperate zones. Agricultural and Forest Meteorology 262, 81–88. Wang, Q., He, N.P., Y, G.R., Gao, Y., Wen, X.F., Wang, R.F., Koerner, S.E., Yu, Q., 2016b. Soil microbial respiration rate and temperature sensitivity along a North‐South forest transect in eastern China: patterns and influencing factors. Journal of Geophysical Research: Biogeosciences 121, 399–410. Xu, X., Luo, Y.Q., Zhou, J.Z., 2012. Carbon quality and the temperature sensitivity of soil organic carbon decomposition in a tallgrass prairie. Soil Biology and Biochemistry 50, 142–148.
Conflicts of interest There are no conflicts of interest to declare. Acknowledgements We thank for the useful comments from Professor Yiqi Luo. Funding for this work came from the Natural Science Foundation of China (31770655, 41571130043), the National Key R&D Program of China (2016YFC0500102), and the Program of Youth Innovation Research Team Project (LENOM2016Q0005). There are no conflicts of interest to declare. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.soilbio.2019.107596. References Arrhenius, S., 1889. On the rate of reaction of the inversion of sucrose by acids. Journal of Physical Chemistry 4, 226–248. Bracho, R., Natali, S., Pegoraro, E., Crummer, K.G., Schädel, C., Celis, G., Hale, L., Wu, L., Yin, H., Tiedje, J.M., 2016. Temperature sensitivity of organic matter decomposition of permafrost-region soils during laboratory incubations. Soil Biology and Biochemistry 97, 1–14. Bradford, M.A., 2013. Thermal adaptation of decomposer communities in warming soils. Frontiers in Microbiology 4, 1–16. Bradford, M.A., Davies, C.A., Frey, S.D., Maddox, T.R., Melillo, J.M., Mohan, J.E., Reynolds, J.F., Treseder, K.K., Wallenstein, M.D., 2008. Thermal adaptation of soil microbial respiration to elevated temperature. Ecology Letters 11, 1316–1327. Chen, X.P., Tang, J., Jiang, L.F., Li, B., Chen, J.K., Fang, C.M., 2010. Evaluating the impacts of incubation procedures on estimated Q10 values of soil respiration. Soil Biology and Biochemistry 42, 2282–2288. Cheng, W., Virginia, R.A., 1993. Measurement of microbial biomass in arctic tundra soils using fumigation-extraction and substrate-induced repiration procedures. Soil Biology and Biochemistry 25, 135–141. Conant, R.T., Ryan, M.G., Agren, G.I., Birge, H.E., Davidson, E.A., Eliasson, P.E., Evans, S.E., Frey, S.D., Giardina, C.P., Hopkins, F.M., Hyvonen, R., Kirschbaum, M.U.F., Lavallee, J.M., Leifeld, J., Parton, W.J., Steinweg, J.M., Wallenstein, M.D., Wetterstedt, J.A.M., Bradford, M.A., 2011. Temperature and soil organic matter decomposition rates - synthesis of current knowledge and a way forward. Global Change Biology 17, 3392–3404. Conant, R.T., Steinweg, J.M., Haddix, M.L., Paul, E.A., Plante, A.F., Six, J., 2008. Experimental warming shows that decomposition temperature sensitivity increases with soil organic matter recalcitrance. Ecology 89, 2384–2391. Cox, P.M., Betts, R.A., Jones, C.D., Spall, S.A., Totterdell, I.J., 2000. Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model. Nature 408, 184–187. Craine, J.M., Fierer, N., McLauchlan, K.K., 2010. Widespread coupling between the rate and temperature sensitivity of organic matter decay. Nature Geoscience 3, 854–857. Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbon
8
Contents lists available at ScienceDirect
Soil Biology and Biochemistry journal homepage: www.elsevier.com/locate/soilbio
A new incubation and measurement approach to estimate the temperature response of soil organic matter decomposition
T
Yuan Liua,b, Nianpeng Hea,b,c,∗, Li Xua, Jing Tiana, Yang Gaoa,b, Shuai Zhengd, Qing Wanga, Xuefa Wena,b, Xingliang Xua, Kuzyakov Yakove,f,g a Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China b College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China c Institute of Grassland Science, Northeast Normal University, and Key Laboratory of Vegetation Ecology, Ministry of Education, Changchun, 130024, China d State Environmental Protection Key Laboratory of Wetland Ecology and Vegetation Restoration, Northeast Normal University, Changchun, 130117, China e Department of Soil Science of Temperate Ecosystems, Department of Agricultural Soil Science, University of Goettingen, 37077, Göttingen, Germany f Agro-Technology Institute, RUDN University, Moscow, Russia g Institute of Environmental Sciences, Kazan Federal University, 420049, Kazan, Russia
A R T I C LE I N FO
A B S T R A C T
Keywords: Microbial adaption Substrates depletion Soil organic matter decomposition Temperature sensitivity
A reliable and precise estimate of the temperature sensitivity (Q10) of soil organic matter (SOM) decomposition is critical to predict feedbacks between the global carbon (C) cycle and climate change. In this study, we first summarize two commonly used approaches for estimating Q10 (Approach A: constant temperature incubation and discontinuous measurements, CDM model; Approach B: varying temperature incubation and discontinuous measurements, VDM model). We then introduced a newly developed approach (Approach C, VCM model) that combines rapidly varying temperature incubations and continuous measurements of SOM decomposition rates (Rs) that may be more realistic and suitable for Q10 estimation, especially for large scale estimation. Then, we conducted a 26-day incubation experiment using three different soils to compare the performance of these three approaches for estimating Q10 using R2 and P-values as indicators. Our results demonstrate that the fitting goodness of the exponential model was consistently higher for Approach C, with higher R2 values, lower confidence intervals, and lower P-values in almost all cases compared with Approaches A and B. Furthermore, results showed that Approaches A and B underestimated the Q10 value by 9.5–13% and 2.9–5.7%, respectively, in three different soils throughout the entire incubation period. Compared with traditional commonly used methods, the newly developed Approach C (VCM model) provides a more accurate and rapid estimation of the temperature response of SOM decomposition and can be used for large-scale estimation of Q10.
1. Introduction The decomposition rate (Rs) of soil organic matter (SOM) is highly sensitive to changes in temperature (Fang and Moncrieff, 2001). Therefore, the temperature response of the Rs under various global warming scenarios has received considerable attention in recent decades (Cox et al., 2000; Knorr et al., 2005; Liu et al., 2017). Given that soil stores enormous global organic carbon (C), minor changes in Rs will influence atmospheric carbon dioxide (CO2) concentrations and result in strong feedback between climate change and terrestrial C pools (Conant et al., 2011). The response of Rs to changing temperatures is referred to as temperature sensitivity (Q10), which is defined as the
factor by which Rs increases with each 10 °C increase in temperature and is usually calculated by the assumed exponential relationship between Rs and temperature (Kirschbaum, 1995; Davidson and Janssens, 2006). The Q10 parameter is important and is widely used in ecological models to predict changes in SOM decomposition under various temperature scenarios (Foereid et al., 2014). Therefore, a reliable and accurate estimate of Q10 is critical to predict the amplitude and direction of feedbacks between the global C cycle and climate change (Cox et al., 2000; Friedlingstein et al., 2006). Over the last few decades, researchers have conducted numerous laboratory experiments to estimate the Q10 of SOM decomposition using different incubation and measurement approaches (Constant incubation
∗ Corresponding author. Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China. E-mail address: [email protected] (N. He).
https://doi.org/10.1016/j.soilbio.2019.107596 Received 7 May 2019; Received in revised form 5 September 2019; Accepted 7 September 2019 Available online 07 September 2019 0038-0717/ © 2019 Elsevier Ltd. All rights reserved.
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
are incubated at several discrete constant temperatures (e.g., 5, 10, 15, 20, and 25 °C), and Rs is measured in intervals of days, weeks, or months throughout the experimental period (Knorr et al., 2005; Conant et al., 2011). The SOM decomposition rate is measured by multiple methods, including alkali absorption, gas chromatography, or CO2 infrared spectroscopy. Subsequently, Q10 is commonly calculated based on the assumed exponential relationship between Rs and soil temperature. The advantage of Approach A is that it is simple (just need alkali absorption) and there is no need for complex equipment (such as continuous temperature regulating system and CO2 monitoring system), making it easy for most laboratories to implement. This approach has helped us have a better understanding of the temperature response of SOM decomposition (Bradford et al., 2008; Conant et al., 2008) and the feedback between climate change and terrestrial C pools. Despite the obvious advantages, this approach has several inevitable shortcomings (Table 1). The first one was the unevenly substrate depletion among different temperatures. Generally, soils that was incubated at higher temperatures might consume more substrate than those incubated at lower temperatures, which will result in relative lower Rs at higher temperatures. These issues strongly affect the accuracy of Q10 estimation, especially in long-term incubation experiments (Conant et al., 2008; Fissore et al., 2009). Second, long-term incubation at different but constant temperatures commonly leads to a certain degree of microbial adaption (Bradford et al., 2008; Bradford, 2013), which might influence the accuracy of estimated Q10 values. This may be realistic for deeper soils because deeper soils experience much less temperature variation than surface soils due to heat transfer. However, surface soil usually experiences large temperature variations due to direct solar radiation. Third, the temperature response of Rs was commonly determined using 3–5 temperatures and these lower number of separated temperature points is not statistically adequate to accurately estimate the relationship between Rs and temperature. The classically adopted method using the exponential equation requires a high frequency of temperature-Rs measurements to increase the accuracy of Q10 estimation (Conant et al., 2011). To overcome the issue of low frequency measurement, a group of New Zealand scientists recently developed a new incubation system that can incubate soils along a temperature gradient (2–50 °C) with 44 discrete temperatures and simultaneously measure the soil respiration rate (Robinson et al., 2017). In any case, improving the fitting accuracy with more temperature measurements is only useful for short-time incubations and still cannot overcome the issue of uneven substrate depletion under different temperature conditions in long-time incubations. Finally, it is difficult to use Approach A in experiments with large scale soil samples because this approach is time consuming and labor intensive and consequently hinders the ability to explore the spatial variation in Q10 using a unified methodology.
or sequential incubation method) (Fang and Moncrieff, 2001; Conant et al., 2011; Ding et al., 2016) and helped us have a deeper understanding of the temperature response of SOM decomposition. However, diverse approaches make it difficult to interpret results from different studies and might result in inaccuracy and uncertainty in Q10 estimation (Liang et al., 2015). The selection between different approaches depends on the research question (Conant et al., 2011; Meyer et al., 2018). In the laboratory, large Q10 variation is mainly derived from changes in substrate quality and quantity during incubation at different temperatures. Commonly, soils (whether surface or deep soil) tend to be incubated at several discrete but constant temperatures (often 3–5 temperatures; Conant et al., 2011). Under this condition, substrates are always consumed faster and more intensely at higher temperatures during long-term incubation, which may lead to an underestimation of the Q10 value in the final stage of incubation (Liang et al., 2015). Commonly, the temperature response of SOM decomposition is estimated using either the Arrhenius equation (Arrhenius, 1889; Davidson and Janssens, 2006) or the exponential equation (Gershenson et al., 2009; Bracho et al., 2016). Below optimum temperature, Q10 is commonly calculated based on the assumed exponential relationship between Rs and temperature (Liu et al., 2018a). Limited data points derived from those discrete temperatures (3–5 temperature points) may decrease the fitting goodness and increase the fitting errors of the temperature responses of SOM decomposition, which may affect the accuracy of the estimated Q10 value (Chen et al., 2010; Robinson et al., 2017). Robinson et al. (2017) suggested that confidence in parameter fits requires about 20 evenly distributed incubation temperatures. Furthermore, fitting the temperature response curve of soil respiration with several more temperature measurements provides a better estimation of Q10 than estimating Q10 with data at two or three temperatures (Chen et al., 2010). Laboratory incubation experiments have improved our understanding of the temperature response of SOM decomposition in different sites and ecosystems (Fang and Moncrieff, 2001; Gershenson et al., 2009; Liang et al., 2015; Ding et al., 2016; Liu et al., 2017). However, a systematic comparison of the effect of different incubation and measurement approaches on estimated Q10 values is still lacking, making it difficult to compare results from different studies using different methods, which consequently impedes our ability to accurately predict large-scale temperature responses of SOM decomposition (Meyer et al., 2018). In this study, we first summarize the advantages and disadvantages of two commonly used incubation and measurement approaches to estimated Q10 values. Then, we suggest a newly developed approach to conveniently estimate Q10 with greater accuracy. Finally, we conducted a 26-day laboratory incubation experiment using three different soils to compare the performance of these three approaches for estimating Q10 using R2 and P values as indicators. 2. Current approaches used to estimate and calculate Q10
2.2. Approach B: varying temperature incubation with discontinuous measurements (VDM model)
Feasible incubation and measurement approaches are essential for obtaining an accurate estimate of Q10. Thus, it is important to consider the factors that could potentially affect Q10 estimation (e.g., SOC component, microbial biomass and species composition, incubation time) and the operability of the measuring system. We first summarize two commonly used approaches from previous studies for estimating Q10.
With advance in science, Approach B was proposed to overcome the shortcomings of substrate depletion and microbial adaption associated with Approach A. Approach B involves varying temperature incubation and discontinuous measurements of Rs (Fig. 1 B). In brief, 5 or 6 discrete constant temperatures (e.g., 5, 10, 15, 20, and 25 °C) are gradually increased and then decreased in a stepwise manner (e.g., 5–10–15–20–25–20–15–10–5 °C). Subsequently, the corresponding Rs at a specific and constant temperature is measured after stabilizing the temperature for 30 min to several hours. After that, Q10 is estimated based on the corresponding exponential relationship between Rs and temperature (Fang et al., 2005; Ding et al., 2016). In practice, all samples are pre-incubated at a constant temperature (e.g., 20 °C) for three to seven days to avoid the pulse effect of disturbance and rewetting.
2.1. Approach A: constant temperature incubation with discontinuous measurements (CDM model) Approach A is a commonly used method in recent decades that combines constant incubation temperatures and discontinuous Rs measurement to estimate Q10 (Fig. 1). In practice, all samples are preincubated at a constant temperature (e.g., 20 °C) for three to seven days to avoid the pulse effect of disturbance and wetting. Then, soil samples 2
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 1. Three incubation and measurement approaches used to investigate how the soil organic matter (SOM) decomposition responds to changing temperature. Approach A: constant temperature incubation and discontinuous measurements. Approach B: varying temperature incubation and discontinuous measurements. Approach C: a newly developed varying temperature incubation and continuous measurements with higher frequency measurement (36 times measurement during 12 h) of SOM decomposition that may overcome the shortcomings of the limited data points associated with Approaches A and B. Table 1 Comparison of the three approaches (using different incubation and measurement method) to estimate the temperature sensitivity (Q10) of SOM decomposition. Brief description of each approach
Advantagesa
Disadvantages
Approach A (CDM model)
Constant temperature incubation and discontinuous measurement
a) Simple, economic, and operable
Approach B (VDM model)
Step-by-step varying temperature incubation and discontinuous measurement
Approach C (VCM model)
Varying temperature incubation and continuous measurement
a) Operable b) Overcome microbial adaption to constant temperature c) Overcome different substrate depletion at different temperature a) Overcome microbial adaption to constant temperature b) Overcome different substrate depletion c) Higher measurement frequency (at minutes intervals with changing temperature) d) Automatic, rapid and continuous, time-saving
a) Microbial adaption to constant temperature b) Different substrate depletion under different temperature c) Lower measurement frequency (commonly 3–5 temperature gradients) d) Time-consuming and laborious a) Higher initial investment b) Lower measurement frequency (commonly 3–5 temperature gradients) c) Complex operation procedure d) Time-consuming and laborious a) Higher initial investment b) No commercial equipment in market
a Advantages and disadvantages of each approach are summarized in view of the factors that influence the temperature response of soil organic matter decomposition (microbial adaption, substrate depletion) during incubation, and the operability of each approach.
Approach B partially overcomes the issues of substrate depletion and microbial adaption that occur using Approach A (Table 1). In Approach B, soil samples with the same incubation temperatures are subject to similar levels of substrate depletion with a short-term stabilization processes (30 min to several hours) to prevent microbial adaption. However, Approach B appears to be suitable for short-term incubation experiments, or soils with abundant substrate, but is not appropriate for long-term incubation experiments or soils with poor
substrate. Furthermore, CO2 concentrations in incubated bottles are commonly measured using alkali absorption, gas chromatography, or CO2 infrared spectroscopy (Fang et al., 2005; Xu et al., 2012; Ding et al., 2016), which limit the experimental scale and the frequency at which Rs is measured. Like Approach A, Approach B is still not statistically adequate to simulate the relationship between Rs and temperature, making it difficult to obtain accurate Q10 values (Robinson et al., 2017).
3
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
together to prevent soil hardening, adjusted to 60% water holding capacity (WHC), and placed in a 150-mL polyethylene plastic bottle. The soil samples were pre-incubated at 25 °C for 7 days, to avoid any pulse effect on microbial activities (Liu et al., 2017). All sample bottles were sealed with preservative films to reduce water loss, and several small holes were punched for ventilation. After 1, 5, 8, 15, 22, and 26 days incubation, SOM decomposition rates (Rs) in each bottle were measured using a new PRI-8800 Automatic Temperature Control Soil Flux System (PRI-8800; Pre-Eco, Beijing, China) referred to in previous studies (He et al., 2013; Liu et al., 2017). This system is an open system with nitrogen gas continuously passing the incubation bottles to minimize the inhibitory effects of high CO2 concentrations on SOM decomposition rates. Furthermore, this system enabled us to continuously vary the incubation temperature and measure Rs at high frequency (Rs was measured based on the linear relationship with time every 75 s). For Approach A and Approach B, each soil sample was first stabilized for 30 min at each temperature and then measured for three rounds, and average values of the three rounds were used for each replicate. For Approach C, each soil sample was stabilized for 30 min and each soil sample was then measured 36 times over a 12 h measuring period. In brief, an electric water bath controlled by an automatic temperature regulator (Julabo, Seelbach, Ortenau, Germany) was connected to a LiCOR CO2 analyzer (Li-7100, LI-COR, Lincoln, Nebraska, USA), which recorded the CO2 concentration every second. Soil temperatures in each bottle were simultaneously measured using a button thermometer (iButton® DS, 1922L; Maxim Integrated, Dallas, USA), which recorded the soil temperature every minute to provide accurate paired data of Rs and soil temperature. Soil moisture was monitored by weighing and adjusted by adding deionized water. Finally, we calculated the Q10 value with the classic exponential equation, the R2 and P values for each approach, and compared the performance of these three different approaches among three different soils (He et al., 2013; He and Yu, 2016; Wang et al., 2016b; Liu et al., 2017). The temperature response curve of SOM decomposition measured with these three approaches at different incubation times for three different soils were shown in Figs. 2–4. Using a repeated measures ANOVA, our results showed that measurement approach, incubation time, and soil types significantly influenced the estimation of Q10 (Table 2). The average Q10 in Approach A, B, and C was 1.43, 1.60, and 1.65, respectively in the LS site; 1.48, 1.55, and 1.62, respectively, in the DL site; and 1.29, 1.34, and 1.42, respectively, in the DH site. Overall, Q10 in Approach A was significantly lower than that in Approach C for all three soils (Fig. 5). Q10 in Approach B was significantly lower than that Approach C in the DL and DH sites, but the difference was not significant in the LS site (Fig. 5). Compared with Approach C, Approaches A and B underestimated Q10 by 13% and 2.9% in LS site, 8.8% and 4.2% in DL site, and 9.5% and 5.7% in DH site, respectively. Furthermore, Rs and Q10 gradually decreased with prolonged incubation times (Figs. 2–5) as did R2 in the exponential equations. The goodness of fit of the exponential model of Approach C was consistently higher than that of Approaches A and B, with a smaller confidence interval, higher R2, and lower P-values regardless of short-term or longterm incubation (Figs. 2–5). SOM decomposition rate decreased greatly for higher temperatures (Figs. 2–4) over 26-day incubation. Long-term incubation might cause a higher substrate depletion at high temperature relative to low temperature for Approach A, resulting in a weak response of Rs to increasing temperature and leading to lower Q10 estimations. By overcoming the shortcomings of substrate depletion and microbial adaption in Approach A, Approach B performed better than Approach A (Approach B had a higher R2 value, a lower confidence interval, lower P-values, and higher Q10 values than Approach A, Figs. 2–6). However, the low frequency of measurement in Approach B significantly increased the P-value of the model, especially in soils with a poor substrate (e.g. DH site; Fig. 4). As expected, Approach C performed better over the whole 26-day incubation period and suggests that traditional and commonly used Approaches A and B underestimate
2.3. Approach C: a newly developed approach combines varying temperature incubation with continuous measurement (VCM model) To overcome the shortcomings of the low frequency measurements of Approaches A and B, incubation and measurement approaches have recently been improved (Fig. 1 C). Based on the work by Cheng and Virginia (1993) with modification of temperature regulating system, scientists have recently developed a system to incubate soils under a varying temperature system and automatically and rapidly measure soil respiration rates with high frequency (up to 36 discrete temperature measurements during 12 h at an empirical setting) to accurately estimate Q10 (Approach C, Fig. 1; He et al., 2013). Recently, many studies have applied this approach to accurately estimate the temperature response of soil respiration (Wang et al., 2016a, 2016b, 2018b; Li et al., 2017; Liu et al., 2017, 2018a, 2018b). Approach C fully incorporates the basic theories and calculation methods of Q10 (Table 1). First, it continually varies the incubation temperature for all samples during the entire incubation period. In this way, Approach C overcomes the main shortcomings of Approach A, such as microbial adaptation to a specific constant temperature as well as different substrate depletion at different constant temperatures (Wang et al., 2016b; Liu et al., 2017). In practice, Approach C uses the same soil sample under different incubation temperatures. Furthermore, it can more precisely simulate the daily temperature fluctuations observed in the field, which can overcome microbial adaption to some extent. In addition, substrate depletion may be less significant during short-term periods (for several hours or a day; Liang et al., 2015). Second, this approach has the obvious advantage of rapidly and automatically measuring CO2 concentrations for every second and calculating Rs on a minute scale. For 12 h measurements, this approach can achieve 36 measurements of SOM decomposition at continually different temperatures. In this way, this approach can overcome shortcomings associated with a low frequency of measurement in Approach A and Approach B. Consequently, the estimation accuracy of Q10 is enhanced by increasing the frequency of measurement (Li et al., 2017). The drawback of Approach C is the requirement for professional experimental equipment, which might limit its application (Table 1). Nonetheless, many studies have employed this approach to estimate the temperature response of SOM decomposition in different sites and ecosystems (Wang et al., 2016b; Li et al., 2017; Liu et al., 2017). With the continuous development of the equipment, Approach C may be well-placed to estimate the temperature response of SOM decomposition in various sites and ecosystems and to predict feedback between the global C cycle and climate change (Fierer et al., 2006; Craine et al., 2010; Meyer et al., 2018). 3. Comparison of these three approaches in Q10 estimation by laboratory experiments To compare the performance of these three approaches, we conducted a laboratory incubation experiment using three different soils along a climate gradient from southern to northern China, including a subtropical evergreen broad-leaved forest soil on Dinghu Mountain (DH), a warm temperate coniferous broad-leaved forest soil on Dongling Mountain (DH), and a cold temperate coniferous forest soil in Liangshui (LS; Table S2). Detailed information regarding soil and vegetation have been described in Wang et al. (2016b). In brief, each forest soil sample was divided into three parts: the first part was subjected to Approach A (10, 15, 20, 25, and 30 °C constant temperature incubation and discontinuous measurement), the second part was subjected to Approach B (10–30 °C varying temperature incubation and 10, 15, 20, 25, and 30 °C constant temperature measurement), and the third part was subjected to Approach C (10–30 °C varying temperature incubation and continuous measurement; see Table S1 for details of the experimental design). In practice, each approach had five repetitions. In each replicate, 40 g of fresh soil and 10 g quartz sand were mixed 4
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 2. Comparison of three approaches for measuring the temperature response of SOM decomposition in cold temperate forest soil in Liangshui (LS). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5).
4. Application and prospects of the newly developed Approach C (VCM model)
Q10 values. Since Q10 is an important parameter for the prediction of the response of SOM decomposition to climate warming, a slight change in Q10 will result in large uncertainty when estimating the amplitude and direction of the feedback between the global C cycle and climate change. Therefore, future studies should be cautious using Approaches A and B to estimate Q10.
Approach C overcomes the shortcomings of traditional incubation experiments for estimating Q10 and has obvious advantages over previous approaches in terms of adjustable incubation temperatures and rapid automatic measurements (Table 1). Furthermore, this approach can also be applied in innovative studies of Q10. For example, this approach allows us to simulate field temperature change more precisely with continuous measurements for soil respiration and allows us to explore
Fig. 3. Comparison of the three approaches for measuring the temperature response of SOM decomposition in warm temperate forest soil in Dongling Mountain (DL). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5). 5
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 4. Comparison of the three approaches for measuring the temperature response of SOM decomposition in subtropical forest soil in Dinghu Mountain (DH). Significance was determined by P < 0.05. Shaded areas indicate the 95% confidence interval. Values are mean ± SE (n = 5).
on soil microbial respiration occur rapidly (at minute scales), and it has been proved difficult to capture the quick response curves after a rainfall pulse using traditional approaches (at hours scale; Wang et al., 2016a; Song et al., 2017). For example, precipitation pulses commonly occur in arid or semi-arid regions; however, variation in soil microbial activity (e.g., soil C mineralization rate) in response to soil water availability is poorly understood (Wang et al., 2016a). Different precipitation levels and soil drying intensities might generate varied response times and intensities. Once a response appears in minutes, it is difficult to observe the actual response and conduct a series of experiments at the same time using traditional equipment. Similar limitations occur in other pulse experiments. Theoretically, differences in estimated Q10 values are inevitable among the different approaches. We believe that as the required technology and methodologies improve, the improved approach (Approach C, VCM model) will have the potential to overcome the shortcomings of previous approaches. However, it is necessary to further identify differences in Q10 estimates among the different approaches using a series of different soils, although Approach C has clearly outperformed the other two approaches in this study. Ideally, future studies should integrate the Q10 values documented by the different approaches to quantify the transformation factors among different approaches through a systematic comparison experiment along a soil substrate gradient. Furthermore, the choice between different approaches depends on the research question. Deeper soils experience much less temperature variation than surface soils due to heat transfer. Therefore, the newly developed method (VCM model) may indeed be better for shallow soils, but the constant-temperature method may be more realistic for deeper soils.
Table 2 Repeated measures analysis of variance of Q10 and R2 for the effect of Approaches (three approaches), Incubation times (six incubation time), Soils (three soils), and their interactions. df
Approach Time Soil Approach × Time Approach × Soil Soil × Time Approach × Time × Soil
2 5 2 10 4 10 20
R2
Q10 F
P
F
P
135.83 82.29 278.01 9.45 6.29 4 2.38
< 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01
50.06 10.4 70.25 4.92 2.204 2.36 1.81
< 0.01 < 0.01 < 0.01 < 0.05 0.07 < 0.01 < 0.01
df, degrees of freedom.
the underlying mechanisms controlling Q10. Such insights can improve the accuracy of extrapolating the results of laboratory incubations to the field. Owing to the limitations of traditional approaches, most existing studies have been conducted at hourly, daily, or weekly scales. Yet, such scales introduce inaccuracies when simulating diel temperature dynamics. Using Approach C, we explored the different responses of Rs to increasing and decreasing diel temperature dynamics in grassland (Li et al., 2017) and forest soils (Liu et al., 2018b; Wang et al., 2018a) and found significant differences in Q10 values between the increasing and decreasing phases. The advantages of Approach C (i.e., automatic, continuous, and rapid) could be further utilized to investigate the spatial variability in Q10 across sites or regions through a consistent network (He et al., 2013; Liu et al., 2017). Using Approach C, previous results showed that Q10 varies greatly among different ecosystems, with different factors dominating Q10 in the different ecosystems (Liu et al., 2017). In practice, it is difficult to integrate Q10 measurements from multiple studies due to considerable differences in the objectives, incubation processes, and measurement methods of the different studies. Approach C also provides the opportunity to explore and capture the pulse response of soil microbes to resources or water availability. Generally, the pulse effect of rainfall or resources (e.g. root exudates)
5. Conclusion In summary, we compared two commonly used approaches for estimating Q10 and then developed a convenient and accurate approach (Approach C, VCM model) that can be used as a reference method to estimate Q10 for Rs or soil organic matter decomposition studies. This study is the first to compare the performance of different approaches in estimating Q10 of SOM decomposition using a laboratory incubation 6
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
Fig. 5. Changes in temperature sensitivity (Q10) with incubation time for the three approaches. Lowercase letters in each column indicate a significant difference between two values when P < 0.05. Values are mean ± SE (n = 5). Ave represent the average Q10 value over 26-day incubation. Liangshui (LS), Dongling (DL), Dinghu (DH).
Fig. 6. Changes in the R-square (R2) value of the fitted equations with incubation time for the three approaches. Lowercase letters in each column indicate a significant difference between two values when P < 0.05. Values are mean ± SE (n = 5). Ave represent the average R2 value over the 26-day incubation. Liangshui (LS), Dongling (DL), Dinghu (DH). 7
Soil Biology and Biochemistry 138 (2019) 107596
Y. Liu, et al.
method with different soils. We found that the fitting goodness of the exponential model was consistently higher for Approach C, with higher R2 values, lower confidence intervals, and lower P-values compared with Approaches A and B, regardless of incubation time. For these three approaches, Q10 in three soils all gradually decreased with prolonged incubation times. In addition, Approaches A and B underestimated the Q10 value by 9.5–13% and 2.9–5.7%, respectively, in three different soils throughout the entire incubation period. Additionally, Approach C (VCM model) has the obvious advantages in terms of substrate depletion, microbial adaption, and operability (Table 1). Therefore, this new developed VCM model is a powerful method for rapidly monitoring soil microbial responses to changes in temperature, which consequently improving our ability to explore the mechanisms underlying SOM decomposition.
decomposition and feedbacks to climate change. Nature 440, 165–173. Ding, J.Z., Chen, L.Y., Zhang, B.B., Liu, L., Yang, G.B., Fang, K., Chen, Y.L., Li, F., Kou, D., Ji, C.J., Luo, Y.Q., Yang, Y.H., 2016. Linking temperature sensitivity of soil CO2 release to substrate, environmental and microbial properties across alpine ecosystems. Global Biogeochemical Cycles 30, 1310–1323. Fang, C., Moncrieff, J.B., 2001. The dependence of soil CO2 efflux on temperature Soil Biology and Biochemistry 33, 155–165. Fang, C.M., Smith, P., Moncrieff, J.B., Smith, J.U., 2005. Similar response of labile and resistant soil organic matter pools to changes in temperature. Nature 433, 57–59. Fierer, N., Colman, B.P., Schimel, J.P., Jackson, R.B., 2006. Predicting the temperature dependence of microbial respiration in soil: a continental-scale analysis. Global Biogeochemical Cycles 20, 1–10. Fissore, C., Giardina, C.P., Kolka, R.K., Trettin, C.C., 2009. Soil organic carbon quality in forested mineral wetlands at different mean annual temperature. Soil Biology and Biochemistry 41, 458–466. Foereid, B., Ward, D.S., Mahowald, N., Paterson, E., Lehmann, J., 2014. The sensitivity of carbon turnover in the Community Land Model to modified assumptions about soil processes. Earth System Dynamics 5, 211–221. Friedlingstein, P., Cox, P., Betts, R., Bopp, L., Von Bloh, W., Brovkin, V., Cadule, P., Doney, S., Eby, M., Fung, I., 2006. Climate-carbon cycle feedback analysis: results from the C4MIP model intercomparison. Journal of Climate 19, 3337–3353. Gershenson, A., Bader, N.E., Cheng, W.X., 2009. Effects of substrate availability on the temperature sensitivity of soil organic matter decomposition. Global Change Biology 15, 176–183. He, N., Yu, G., 2016. Stoichiometrical regulation of soil organic matter decomposition and its temperature sensitivity. Ecology and Evolution 6, 620–627. He, N.P., Wang, R.M., Gao, Y., Dai, J.Z., Wen, X.F., Yu, G.R., 2013. Changes in the temperature sensitivity of SOM decomposition with grassland succession: implications for soil C sequestration. Ecology and Evolution 3, 5045–5054. Kirschbaum, M.U.F., 1995. The temperature-dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry 27, 753–760. Knorr, W., Prentice, I., House, J., Holland, E., 2005. Long-term sensitivity of soil carbon turnover to warming. Nature 433, 298–301. Li, J., He, N., Xu, L., Chai, H., Liu, Y., Wang, D., Wang, L., Wei, X., Xue, J., Wen, X., Sun, X., 2017. Asymmetric responses of soil heterotrophic respiration to rising and decreasing temperatures. Soil Biology and Biochemistry 106, 18–27. Liang, J., Li, D., Shi, Z., Tiedje, J.M., Zhou, J., Schuur, E.A., Konstantinidis, K.T., Luo, Y., 2015. Methods for estimating temperature sensitivity of soil organic matter based on incubation data: a comparative evaluation. Soil Biology and Biochemistry 80, 127–135. Liu, Y., He, N., Wen, X., Xu, L., Sun, X., Yu, G., Liang, L.L., Schipper, L.A., 2018a. The optimum temperature of soil microbial respiration: patterns and controls. Soil Biology and Biochemistry 121, 35–42. Liu, Y., He, N., Zhu, J., Xu, L., Yu, G., Niu, S., Sun, X., Wen, X., 2017. Regional variation in the temperature sensitivity of soil organic matter decomposition in China's forests and grasslands. Global Change Biology 23, 3393–3402. Liu, Y., Wen, X., Zhang, Y., Tian, J., Gao, Y., Ostle, N.J., Niu, S., Chen, S., Sun, X., He, N., 2018b. Widespread asymmetric response of soil heterotrophic respiration to warming and cooling. The Science of the Total Environment 635, 423–431. Meyer, N., Welp, G., Amelung, W., 2018. The temperature sensitivity (Q10) of soil respiration: controlling factors and spatial prediction at regional scale based on environmental soil classes. Global Biogeochemical Cycles 32, 306–323. Robinson, J.M., O'Neill, T.A., Ryburn, J., Liang, L.L., Arcus, V.L., Schipper, L.A., 2017. Rapid laboratory measurement of the temperature dependence of soil respiration and application to changes in three diverse soils through the year. Biogeochemistry 133, 101–112. Song, X., Zhu, J., He, N., Huang, J., Tian, J., Zhao, X., Liu, Y., Wang, C., 2017. Asynchronous pulse responses of soil carbon and nitrogen mineralization to rewetting events at a short-term: regulation by microbes. Scientific Reports 7 7492. Wang, Q., He, N., Liu, Y., Li, M., Xu, L., 2016a. Strong pulse effects of precipitation events on soil microbial respiration in temperate forests. Geoderma 275, 67–73. Wang, Q., He, N., Liu, Y., Li, M., Xu, L., Zhou, X., 2018a. Important interaction of chemicals, microbial biomass and dissolved substrates in the diel hysteresis loop of soil heterotrophic respiration. Plant and Soil 428, 279–290. Wang, Q., He, N., Xu, L., Zhou, X., 2018b. Microbial properties regulate spatial variation in the differences in heterotrophic respiration and its temperature sensitivity between primary and secondary forests from tropical to cold-temperate zones. Agricultural and Forest Meteorology 262, 81–88. Wang, Q., He, N.P., Y, G.R., Gao, Y., Wen, X.F., Wang, R.F., Koerner, S.E., Yu, Q., 2016b. Soil microbial respiration rate and temperature sensitivity along a North‐South forest transect in eastern China: patterns and influencing factors. Journal of Geophysical Research: Biogeosciences 121, 399–410. Xu, X., Luo, Y.Q., Zhou, J.Z., 2012. Carbon quality and the temperature sensitivity of soil organic carbon decomposition in a tallgrass prairie. Soil Biology and Biochemistry 50, 142–148.
Conflicts of interest There are no conflicts of interest to declare. Acknowledgements We thank for the useful comments from Professor Yiqi Luo. Funding for this work came from the Natural Science Foundation of China (31770655, 41571130043), the National Key R&D Program of China (2016YFC0500102), and the Program of Youth Innovation Research Team Project (LENOM2016Q0005). There are no conflicts of interest to declare. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.soilbio.2019.107596. References Arrhenius, S., 1889. On the rate of reaction of the inversion of sucrose by acids. Journal of Physical Chemistry 4, 226–248. Bracho, R., Natali, S., Pegoraro, E., Crummer, K.G., Schädel, C., Celis, G., Hale, L., Wu, L., Yin, H., Tiedje, J.M., 2016. Temperature sensitivity of organic matter decomposition of permafrost-region soils during laboratory incubations. Soil Biology and Biochemistry 97, 1–14. Bradford, M.A., 2013. Thermal adaptation of decomposer communities in warming soils. Frontiers in Microbiology 4, 1–16. Bradford, M.A., Davies, C.A., Frey, S.D., Maddox, T.R., Melillo, J.M., Mohan, J.E., Reynolds, J.F., Treseder, K.K., Wallenstein, M.D., 2008. Thermal adaptation of soil microbial respiration to elevated temperature. Ecology Letters 11, 1316–1327. Chen, X.P., Tang, J., Jiang, L.F., Li, B., Chen, J.K., Fang, C.M., 2010. Evaluating the impacts of incubation procedures on estimated Q10 values of soil respiration. Soil Biology and Biochemistry 42, 2282–2288. Cheng, W., Virginia, R.A., 1993. Measurement of microbial biomass in arctic tundra soils using fumigation-extraction and substrate-induced repiration procedures. Soil Biology and Biochemistry 25, 135–141. Conant, R.T., Ryan, M.G., Agren, G.I., Birge, H.E., Davidson, E.A., Eliasson, P.E., Evans, S.E., Frey, S.D., Giardina, C.P., Hopkins, F.M., Hyvonen, R., Kirschbaum, M.U.F., Lavallee, J.M., Leifeld, J., Parton, W.J., Steinweg, J.M., Wallenstein, M.D., Wetterstedt, J.A.M., Bradford, M.A., 2011. Temperature and soil organic matter decomposition rates - synthesis of current knowledge and a way forward. Global Change Biology 17, 3392–3404. Conant, R.T., Steinweg, J.M., Haddix, M.L., Paul, E.A., Plante, A.F., Six, J., 2008. Experimental warming shows that decomposition temperature sensitivity increases with soil organic matter recalcitrance. Ecology 89, 2384–2391. Cox, P.M., Betts, R.A., Jones, C.D., Spall, S.A., Totterdell, I.J., 2000. Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model. Nature 408, 184–187. Craine, J.M., Fierer, N., McLauchlan, K.K., 2010. Widespread coupling between the rate and temperature sensitivity of organic matter decay. Nature Geoscience 3, 854–857. Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbon
8