Sage Journals: Discover world-class research

Abstract

Soil respiration represents the largest carbon (C) flux from terrestrial ecosystems to the atmosphere. We created a study site in tropical lowland rainforest and used static chamber method to measure the temporal variations of soil respiration and their relationship with environmental factors at monthly time scale. The temporal variations of soil respiration showed a seasonal pattern related to soil temperature (p < .01) and soil moisture (p < .05). We tested different regression models to explore the relationship between soil respiration and environmental factors. Soil respiration had a better fit with soil temperature than with soil moisture in single-factor models. At different temperatures, the Q₁₀ values from different models changed in rather different ways. We found that the mixed quadratic model composite of soil temperature and moisture had the best-fitting effect (R² = .74) on soil respiration and could better explain the seasonal variation. In a certain soil moisture range close to 15%, soil respiration increased with soil temperature. However, soil respiration became restricted when the moisture was greatly higher or lower than this value. Furthermore, at low soil temperatures (lower than 16°C), higher soil moisture could decrease soil respiration rapidly. Thus, soil respiration in a tropical lowland rainforest is co-controlled by soil temperature and moisture. This study expands our observations of soil respiration in tropical forests and how it responds to environmental factors, which is important for reducing errors in evaluation and scaling up of soil carbon flux in climate change studies.

Keywords

soil respiration tropical rainforest temporal variation environmental factors regression models

Global warming caused by the increase of carbon dioxide (CO₂) concentrations is currently one of the most important issues to be addressed at a global scale (IPCC Climate Change, 2013). Soil respiration is the largest source of CO₂ emissions from terrestrial ecosystems to the atmosphere (Bond-Lamberty & Thomson, 2010; Wu et al., 2011). Climatic warming is hypothesized to increase rates of soil respiration, potentially fueling further increases in global temperatures (Carey et al., 2016). Accurate estimation of soil respiration efflux and accurate identification of the factors controlling soil respiration are very important to understand the ecosystem C cycle under global climate change scenarios (Bond-Lamberty & Thomson, 2010).

Tropical forests contain 40% of the total C in global terrestrial ecosystems (Beer et al., 2010; Jobbágy & Jackson, 2000; Pan et al., 2011), of which, 56% is found in aboveground biomass and 32% in soils (Ngo et al., 2013; Pan et al., 2011). They exchange more CO₂ with the atmosphere than any other biome on Earth (Cavaleri et al., 2015) and, thus, play a key role in the global C cycle (Goodrick et al., 2016). In the coming decades, the tropics will experience unprecedented changes in temperature, rapid increases in atmospheric CO₂ concentrations, and significant alterations in the timing and amount of rainfall (Anderson, 2012; Diffenbaugh & Scherer, 2011; Mora et al., 2013).

In principle, soil respiration is the combination of autotrophic respiration by plant roots and associated microorganisms (i.e., rhizosphere respiration) and heterotrophic respiration by microbes decomposing soil organic matter (Hanson et al., 2000; Högberg & Read, 2006; Ryan & Law, 2005). Soil moisture and temperature are two primary abiotic drivers of root and microbial activities (Davidson et al., 1998; Liu et al., 2016; Lloyd & Taylor, 1994). Understanding their potential moderating effects on soil respiration is essential for predicting the responses of soil respiration to climate changes.

Soil respiration rates have been studied in many of the world’s ecosystems to explain the relationship between soil respiration and environmental factors. In boreal ecosystems and temperate regions, soil temperature is the most important determinant of soil respiration rate (Shibistova et al., 2002; Vargas et al., 2010; Xu & Qi, 2001). Compared with other factors, temperature can directly affect root and microbial metabolic rates (Kuzyakov & Gavrichkova, 2010; Lükewille & Wright, 1997) as well as temporal variations in soil respiration. The C efflux from soil in these regions can be estimated using an empirical function and soil temperature data (Adachi et al., 2009). Different types of models have been used: exponential or Arrhenius equations (Lloyd & Taylor, 1994; MacDonald et al., 1995; Thierron & Laudelout, 1996), linear models (Rochette et al., 1991), and quadratic models (Holthausen & Caldwell, 1980). Although these models have been reported to be successful in fitting data obtained from an individual experiment under some specific circumstances, they suggest different explanations for the response of soil respiration to temperature. The Q₁₀ value defines the temperature dependence or the sensitivity to temperature variations of soil respiration. When Q₁₀ values are derived from different models, they are different both in terms of magnitude and with respect to temperature. Undoubtedly, simulating soil respiration without a good understanding of the variation in temperature sensitivity of soil respiration will limit a model’s utility.

In tropical regions, soil temperatures do not strongly influence the soil respiration rates; instead, most previous studies have highlighted the control of soil moisture on soil respiration (Davidson et al., 2000; Schwendenmann et al., 2003; Sotta et al., 2006). Soil CO₂ efflux can be suppressed in both low and high soil water content (Davidson et al., 2012; Liptzin et al., 2011). Several nonlinear relationships have been proposed to link soil respiration rate and soil water content (Cook & Orchard, 2008; Davidson et al., 2000), indicating optimal conditions for microbial decomposition and root respiration at intermediate moisture conditions. However, these relationships remain empirical, and it is unknown how they vary with soil, climate, and forest type (Rubio & Detto, 2017). Sotta et al. (2004) claimed that short-term variation in soil respiration rates depend on soil temperature, but the soil water contents might be a limiting factor of long-term variation in soil respiration rates in central Amazonian tropical forests.

Therefore, the effects of soil temperature and moisture on soil respiration still need to be expanded. We investigated soil respiration by measuring soil CO₂ flux, soil temperature, and soil moisture at monthly time scale in Diaoluoshan tropical lowland rainforest in Hainan Island. We aimed to address two issues in this study: (a) the relationship between soil respiration and key environmental factors (soil temperature and moisture) in Diaoluoshan tropical lowland rainforest in Hainan Island and (b) the performance of different regression models with the soil respiration data of Diaoluoshan tropical lowland forest and how C efflux was affected.

Methods

Site Description

This study was conducted in Diaoluoshan National Nature Reserve (DNNR) located in the south-eastern part of Hainan Island, China (18°40′ N, 109°54′ E, elevation 255 m, Figure 1). The climate shows strong seasonality because of the tropical marine monsoon climate. The dry and rainy seasons are clearly divided, and more than 80% of the precipitation is concentrated in the rainy season extending from late April to October. Annual rainfall is relatively high and may reach up to 2,160 mm (typhoons might bring some rainfall). The multiyear mean annual temperature is 24.6°C, with the highest value occurring in July and the lowest in January (Zhao et al., 2019).

Figure 1.
Geographic Location of Study Site. DNNR = Diaoluoshan National Nature Reserve.

DNNR has typical zonal forests, including lowland rainforests, mountain rainforests, and hilltop forests. Our study site was categorized as a lowland tropical forest (Yang et al., 1994). The mean canopy height of the forests is approximately 20 to 25 m. The canopy height is lower than that of the inland tropical forests at the same latitude, which might partly be a consequence of typhoons. Before the establishment of the National Nature Reserve, the forest was extensively cut in the 1980s. The dominant species is Vatica mangachapoi, which belongs to the family Dipterocarpaceae. The soil type is dominated by acidic red soil formed by granite or igneous rocks. The soil profile is clear with a mean soil depth deeper than 2.0 m. The soil properties of A-horizon are 9.84 ± 1.97 g · kg⁻¹ total organic carbon, 0.91 ± 0.18 g · kg⁻¹ total organic nitrogen, 5.65 ± 0.17 pH.

Experiment Design

The field campaign was conducted in a 1-ha permanent plot. Three quadrats were selected in the permanent plot to install soil chambers. The chamber consisted of polyvinyl chloride collar of 20 cm in diameter and 10 cm in height, which embedded seamlessly and inserted 3 to 5 cm into the soil, followed by the removal of aboveground biomass. Five chambers were randomly installed in each quadrat, keeping their same position for minimizing interference with the soil environment.

From September 2016 to January 2018, we performed field measurements at monthly intervals and generally completed them before noon. Diurnal variation measurements (7:00–19:00) were conducted with the frequency of each hour on March 18, 2019.

Field Measurements

We used a portable greenhouse gas analyzer, purchased from Los Gatos Research Company (LGR Inc., San Jose, CA, USA), to measure soil respiration. It could store full absorption spectra at a frequency of 1 Hz for further processing or corrections on fluxes. Soil temperature and moisture at 5 cm depth were measured simultaneously with Decagon 5 TM sensor (Decagon Devices Inc., USA). We performed at least three measurements in each chamber, allowing enough time for the analyzer to warm up each time.

Statistics

The one-way analysis of variance was used to determine the differences in mean soil respiration rates during the rainy and dry seasons. A Pearson’s correlation analysis was used to determine whether there was a correlation between soil respiration and soil temperature and moisture and correlated at p < .05. Different regression models were used to find which model could best explain the relationship of soil respiration and environmental factors. Combining the determination coefficient (R²) and Akaike information criterion (AIC) to measure the goodness of fit of the regression models. The larger R² and the smaller the AIC value has the better model fits.

Regression Models

Single-Factor Models of Soil Respiration and Moisture

As shown in Equations (1) to (3), three models that only consider the relationship between soil respiration and moisture were used to fit the relations. Their functions are as follows:

Linear model:
$R_{s} = a + b W$
(1)

Power function model:
$R_{s} = a W^{b}$
(2)

Quadratic model:
$R_{s} = a + b W + c W^{2}$
(3)where a, b, and c are the corresponding fitted parameters for each model, R_s is the soil respiration rate, and W is the soil moisture.

Single-Factor Models of Soil Respiration and Temperature

As shown in Equations (4) to (8), five models that only consider the relationship between soil respiration and temperature were used to fit the relations. Their functions are as follows:

Linear model:
$R_{s} = a + b T$
(4)

Exponential model:
$R_{s} = a e^{b T}$
(5)

Quadratic model:
$R_{s} = a + b T + c T^{2}$
(6)

Arrhenius model (Fang & Moncrieff, 2001):
$R_{s} = a e^{(- \frac{E_{a}}{8.314 T})}$
(7)

Lloyd and Taylor model (Lloyd & Taylor, 1994):
$R_{s} = R_{10} e^{308.56 (\frac{1}{56} - \frac{1}{T + 46})}$
(8)where a, b, c, E_a, and R₁₀ are the corresponding fitted parameters for each model, R_s is the soil respiration rate, and T is the soil temperature.

Q₁₀ Value

The temperature dependence of soil respiration, commonly referred to as the Q₁₀ value, has been the focus of many studies. The value of Q₁₀ is the factor by which the respiration rate differs for a temperature interval of 10°C and is defined as
$Q_{10} = \frac{R_{T + 10}}{R_{T}}$
(9)where R_T and R_T₊₁₀ are the soil respiration rates at temperatures of T and T + 10, respectively (Winkler et al., 1996). The values for different models are obtained with simulated R_T and R_T+10 with the model to be tested (Table 1).

Table 1.
Q₁₀ Values of Each Model at 10°C and 20°C.

Equation
Total
Dry season
Wet season

Q10 at 10°C Q10 at 20°C Q10 at 10°C Q10 at 20°C Q10 at 10°C Q10 at 20°C

Linear NA 2.14 7.67 1.87 0.27 NA

Exponential 3.00 3.00 2.46 2.46 NA NA

Quadratic 1.26 2.58 0.52 3.42 0.18 NA

Arrhenius 16.96 2.57 7.46 1.95 NA 11.76

Lloyd and Taylor 2.30 1.85 2.30 1.85 2.36 1.80

Note. NA means negative value or Q₁₀ > 20.

Mixed Models of Soil Respiration Considering Soil Temperature and Moisture

Equations (10) to (14) were used to study how the two factors of soil temperature and moisture together control soil respiration. Their functions are as follows:

Linear model 1:
$R_{s} = a + b (T + W)$
(10)

Linear model 2:
$R_{s} = a + b T + c W$
(11)

Quadratic model:
$R_{s} = a + b T + c W + d T^{2} + e W^{2} + fTW$
(12)

Power function model:
$R_{s} = a T^{b} w^{c}$
(13)

Exponential model:
$R_{s} = a e^{b T} w^{c}$
(14)where a, b, c, d, e, and f are the corresponding fitted parameters for each model, R_s is the soil respiration rate, T is the soil temperature, and W is the soil moisture.

Results

Seasonal Variation in Soil Respiration and Environmental Factors

Both soil temperature and moisture were measured in each field campaign, while simultaneously measuring soil respiration. As shown in Figure 2, seasonal patterns were observed in soil respiration, temperature, and moisture. Their annual average values were 2.52 ± 0.23 SE μmol CO₂ m⁻² · s⁻¹, 24.02 ± 0.88 SE °C, and 12.88 ± 1.24 SE %, respectively.

Figure 2.
Seasonal Dynamics of Soil Respiration (CO₂ Flux) and Soil Temperature (T_s) and Soil Moisture (M_s).

There is a significantly positive correlation between soil temperature and soil moisture at the depth of 5 cm (p < .01). Warm temperatures coincide with high moisture levels in the wet season, whereas cool temperatures and low moisture levels occur in the dry season. However, in July 2017, the highest soil moisture level was recorded; this caused the soil temperature to reduce to the values lower than those recorded in the adjacent 2 months, indicating that a large amount of precipitation might occur before or during that measurement period, resulting in an increase in soil moisture and a decrease in soil temperature. In general, the characteristics conformed to the typical properties of the monsoon climate, which has obvious wet and dry seasons. It is hot and rainy in summer affected by the warm and humid oceanic airflow, but cold and dry in winter affected by the continental airflow.

Time series of soil respiration measurement varied greatly that ranged from 1.42 to 3.94 μmol CO₂ m⁻² · s⁻¹ and peaked in June 2017. One-way analysis of variance showed significant variations in soil respiration rates during the rainy and dry seasons (p < .001). Emissions were significantly higher and more variable in the wet season and relatively lower and less variable in the dry season. In addition, there were two peaks of soil moisture, with an abnormal decrease in soil CO₂ flux in July and September 2017, compared to the other wet season months.

Diurnal Variation in Soil Respiration

Diurnal variation in soil respiration is shown in Figure 3. The mean rate of soil respiration was 1.97 μmol CO₂ m⁻² · s⁻¹. Based on the measurements performed on March 18, 2019, no significant diurnal change was observed in soil respiration, and it was not related to soil temperature (p = .37 > 0.05, n = 60) because the difference between the maximum and minimum soil temperatures was normally smaller than 2°C. However, soil moisture was negatively correlated with soil respiration (p < .05), ranging from 4.71% to 7.53%.

Figure 3.
Diurnal Dynamics of Soil Respiration (CO₂ Flux) and Soil Temperature (T_s) and Soil Moisture (M_s).

Seasonal Controls of Soil Respiration

During the entire study period, the average soil respiration rate for different seasons was the highest in summer and lowest in winter. This was consistent with the change in soil temperature at 0–5 cm depth (Figure 4A). Meanwhile, the average soil respiration flux in the dry season (November to April) was smaller than that in the wet season (May to October), which was consistent with the change in soil moisture at 0 to 5 cm depth (Figure 4B).

Figure 4.
Soil Respiration in Different Seasons Was Compared With Soil Temperature (A) and Soil Moisture (B).

Relationship Between Soil Respiration and Moisture

Pearson’s correlation analysis showed a significant positive correlation between soil CO₂ fluxes and soil moisture levels in our study (p < .05). Table 2 shows the parameters, determination coefficients (R²), and AIC values obtained by fitting each model. It was difficult to obtain a good fitting effect in the soil respiration model considering only soil moisture. The quadratic model had the highest R² value of .29 and the lowest AIC value. However, the quadratic model fitting (Figure 5) showed that soil moisture has a limitation on soil CO₂ emission. Soil respiration rates increased with soil moisture, but started to drop when soil moisture exceeded approximately 15%.

Table 2.
Fitted Relationships Between Soil Respiration and Soil Temperature and Moisture.

Regression model type Equation Fitted parameters R ² F Akaike information criterion

Single-factor models of soil moisture Linear: $R_{s} = a + b W$ a = 0.05, b = 1.82 .08 4 −11.2

Power: $R_{s} = a W^{b}$ a = 1.20, b = 0.30 .12 202 −12.9

Quadratic: $R_{s} = a + b W + c W^{2}$ a = −0.85, b = 0.50, c = −0.02 .29 8 −22.5

Single-factor models of soil temperature Linear: $R_{s} = a + b T$ a = −2.25, b = 0.20 .52 47 −40.3

Exponential: $R_{s} = a e^{b T}$ a = 0.17, b = 0.11 .59 464 −47.9

Quadratic: $R_{s} = a + b T + c T^{2}$ a = 8.56, b = −0.81, c = 0.03 .62 35 −51.2

Arrhenius: $R_{s} = a e^{(- \frac{E_{a}}{8.314 T})}$ a = 26.08, E_a = 470.69 .53 403 −41.8

Lloyd & Taylor (1994): $R_{s} = R_{10} e^{308.56 (\frac{1}{56} - \frac{1}{T + 46})}$ R₁₀ = 0.84 .51 774 −39.2

Mixed models of soil temperature and moisture Linear 1: $R_{s} = a + b (T + W)$ a = 0.17, b = 0.06 .25 246 −21.2

Linear 2: $R_{s} = a + b T + c W$ a = −2.64, b = 0.24, c = −0.05 .55 422 −43,9

Quadratic: $R_{s} = a + b T + c W$ $+ d T^{2} + e W^{2} + fTW$ a = 6.97, b = −0.55, c = −0.29, d = 0.01, e = 0.02, f = 0.03 .74 300 −72.0

Power: $R_{s} = a T^{b} w^{c}$ a = 0.01, b = 1.87, c = −0.08 .54 411 −42.8

Exponential: $R_{s} = a e^{b T} w^{c}$ a = 0.18, b = 0.12, c = −0.16 .60 322 −50.7

Note. The number of observed effluxes was 45.

Figure 5.
Different Models Fitting of Soil Respiration and Soil Moisture.

Relationship Between Soil Respiration and Temperature

Pearson’s correlation analysis showed a significant positive correlation between soil CO₂ flux and soil temperature in our study (p < .01). In general, soil CO₂ emission rates increase with increasing soil temperature. In the whole study period, soil temperature could provide a better estimation of soil respiration with single-factor empirical models compared to moisture (Table 2). As shown in Figure 6A, linear equation is simply an empirical expression of an increase in soil respiration with increasing temperature without any theoretical basis (R² = .52). The exponential model had an R² value of .59, but a residual analysis (Supplemental Figure 1) suggested that the exponential model might overestimate the respiration rate at high temperatures. The quadratic model produced a best fit with our data (R² = .62, AIC = −51.2). Three parameters are included in the quadratic model, and this may enlarge its applicability. Arrhenius equation is an empirical formula with a good theoretical basis for the relationship between chemical reaction rate and temperature. Nevertheless, the fitting result used in this study was R² = .53, and the soil respiration rate was also greatly underestimated under low temperature conditions. The Lloyd and Taylor model improves the parameters of the original Arrhenius formula based on the measured data. Therefore, the predicted results in this study were relatively poor (R² = .51). Figure 6B and C shows the relationship of soil respiration and temperature in the dry and wet seasons using different models, respectively. In the wet seasons, soil respiration rates varied greatly when the slight variation in soil temperature which continued to be 25°C–28°C. In addition to the Lloyd and Taylor model, the other regression models on the wet season had the approximative determination coefficients and similar fitting effects.

Figure 6.
Different Regression Models Fitting of Soil Respiration and Soil Temperature. A: Relationship between soil respiration and soil temperature fitted by different models throughout the total study period. B and C: Relationship between soil respiration and soil temperature fitted by different models in dry and wet season, respectively.

Effect of Different Regression Models on Q₁₀

Despite the fact that most equations fitted the observed data well and provided similar estimates of soil respiration at different temperatures, the Q₁₀ values derived from these equations change in rather different ways (Table 1). The linear equation gave a high Q₁₀ value at low temperature and then Q₁₀ decreased rapidly with temperature in the dry season. The exponential model provided a non-temperature-dependent Q₁₀ value that was constantly expressed as e^10b, given a value of 3 in the whole study period and 2.46 in the dry season. The quadratic equations showed that the Q₁₀ values increased with increasing temperature, which was contrary to the commonly accepted view that Q₁₀ decreased with temperature. At low temperatures, the Arrhenius model gave a very high Q₁₀ value, which decreased with increasing temperature, consistent with the trend in the dry season. For the Lloyd and Taylor model, the Q₁₀ values were quite close to each other and slowly decreased with temperature, showing less seasonal change.

Combined Effect of Soil Temperature and Moisture on Soil Respiration

The field campaign measurement of soil respiration reflects the combined effects of soil environmental factors. In these mixed models, the quadratic model had the best fit and performed better than any single-factor models (Table 2), which had the highest R² (.74) and lowest AIC value (−72.0) in all models. As shown in Figure 7, within certain temperature and moisture ranges, an increase in both factors would enhance soil CO₂ emissions. However, soil temperature and moisture showed a combined effect on soil respiration. For example, when soil moisture was close to the value of approximately 15%, the soil respiration rates increased with soil temperature, and soil CO₂ emissions were more likely restricted by only moisture when it was greatly higher or lower than 15%. Furthermore, at low temperatures (approximately lower than 16°C), higher moisture levels could decrease soil respiration rapidly.

Figure 7.
Combined Effect of Soil Moisture and Temperature on Soil Respiration by Fitting With the Mixed Quadratic Model.

Discussion

Temporal Variations of Soil Respiration Rates

The soil respiration rates of the tropical lowland rainforest in Diaoluoshan, Hainan Island, exhibited seasonal variations (Figure 2). The temporal variations of soil respiration were sensitive to seasonal changes in soil environmental conditions (Figure 4). Emissions in the wet season were much greater because of the higher soil hydrothermal conditions which were beneficial to the life activities of soil microorganisms and plant roots than in the dry season. In addition, soil hydraulic condition is an important medium for the transfer of soil organic matter. Such patterns of soil respiration in tropical forests have also been observed at other sites such as Thailand (Adachi et al., 2009), Australia (Goodrick et al., 2016), Jianfengling (Jiang et al., 2016), Panama (Rubio & Detto, 2017), Xishuangbanna (Sha et al., 2005), and Amazon (Sotta et al., 2004). The annual soil respiration efflux at our site was 953 ± 87 g C m⁻² · yr⁻¹; this value is lower than that of a tropical lowland rainforest in Panama (1613 g C m⁻² · yr⁻¹; Rubio & Detto, 2017), Thailand (1724 g C m⁻² · yr⁻¹; Adachi et al., 2009), and Amazon (1487 g C m⁻² · yr⁻¹; Sotta et al., 2006), but higher than that of the tropical lowland rainforests in Xishuangbanna, China (831 g C m⁻² · yr⁻¹; Sha et al., 2005) and Hawaii (650–890 g C m⁻² · yr⁻¹; Raich, 1998). Soil is a complex and spatially heterogeneous mixture of minerals and organic pools, including litter, roots, and microorganisms. Each of these components responds differently to environmental variability (P. Li et al., 2013) and is uniquely coupled with other biotic processes, generating a broad spectrum of CO₂ emission rates. Thus, it is not surprising that soil respiration rates vary considerably in different tropical forests.

Our results showed no diurnal fluctuations of soil respiration in Diaoluoshan tropical lowland rainforest between 7:00 and 19:00 h (Figure 3) and were broadly consistent with the results from other tropical forest ecosystems (Sha et al., 2005). On a short-term scale, temperature is the most important factor that determines the rate of soil respiration (Sotta et al., 2004) because temperature is a direct factor affecting root and microbial metabolic rates (Kuzyakov & Gavrichkova, 2010; Lükewille & Wright, 1997). However, soil temperature diurnal fluctuations in the tropics are small, especially in the areas covered by dense canopy, where little radiation reaches forest floors. Thus, in a tropical forest, no obvious diurnal fluctuations might be observed in soil respiration.

Effects of Environmental Factors on Soil Respiration

Soil temperature and moisture are the two most important soil parameters for soil CO₂ emissions, since they control microbial activity and all related processes. In several ecosystems, soil CO₂ fluxes increase exponentially with temperature and are often limited by available soil moisture (Carlyle & Than, 1988; Fang & Moncrieff, 2001; Xu & Shang, 2016). Nevertheless, because the temporal variation of soil temperature in the tropics lies usually within the optimum range of soil respiration, soil moisture is probably the most important abiotic factor influencing soil respiration within tropical forests (Y. Li et al., 2006; Sotta et al., 2006). Rubio and Detto (2017) claimed that soil moisture was the primary driver of temporal variability in tropical forests, with temperature playing a secondary role. However, as Davidson et al. (1998) pointed out, it is difficult, and perhaps impossible, to distinguish the effects of soil temperature and moisture separately. At our study site, the soil CO₂ efflux exhibited a significant relationship with the interaction between soil temperature and soil water content (p < .001, p < .05, respectively). The soil CO₂ efflux was highly limited by soil moisture, and the exponential relationship between soil respiration and soil temperature would no longer apply (Manzoni et al., 2012). High temperatures combined with optimum soil water content might increase soil CO₂ fluxes from soil respiration (Figure 7) in a manner similar to that observed during the wet season in our study. Nevertheless, in July and September 2017, soil CO₂ emission rates decreased when the soil moisture peaked and reached above the optimum water content (Figure 2). Higher water content probably created a barrier for gas diffusion at the soil-atmosphere interface, limiting the escape of CO₂ and supply of oxygen (Liptzin et al., 2011), thereby reducing both, production and diffusion of CO₂ (Davidson et al., 2012; Fang & Moncrieff, 1999).

On the other hand, low soil temperatures occurred simultaneously during the dry season in our study (Figure 2). Low soil temperatures resulted in low root and microbial metabolic rates. The soil water content at 0 to 5 cm depth decreased to 10%, probably resulting in very low root and microbial activity (Silletta et al., 2019). The decrease in soil water availability during the dry season affects several physiological processes, leading to plant dehydration and a substantial loss of root functionality in this ecosystem (Bucci et al., 2013; Scholz et al., 2012). This was also reflected in a decrease in plant transpiration during the drought period (Pereyra et al., 2017), and hence a decrease in CO₂ capture. Furthermore, at low soil moisture conditions, decomposition is limited by soluble C availability (Davidson et al., 2012; Linn & Doran, 1984).

Selection of Soil Respiration Models

As the large temporal variations of soil water content, the single-factor models of soil moisture could not clearly explain and predict soil respiration (Table 2). The relationship between soil respiration rate and soil water content can be expressed as a quadratic function (Figure 5) at our study site. However, this relationship was not consistent in different studies. Adachi et al. (2009) and Sotta et al. (2006, 2007) studied different tropical forests and reported that soil respiration rate and soil water content could be described using a parabolic function. Chambers et al. (2004) reported that the relationship between soil respiration rate and soil water content is curvilinear.

In contrast, our data fitted better with the single-factor models of the effect of soil temperature on soil respiration (Table 2), though there might be large variations due to the higher water content in the wet season (Figure 6). It is difficult to identify the best model because of the close values of the coefficients of determination. A good fit of a model against the experimental data does not necessarily suggest the actual mechanism presumed by the model (Fang & Moncrieff, 2001).

The linear and quadratic models of the relationship between soil respiration and soil temperature are empirical and lack theoretical basis. The responses of soil respiration and mineralization processes to soil temperature are commonly described using exponential and Arrhenius equations. Both equations describe an exponential increase in respiration with increasing temperature, but with a different theoretical basis (Ellert & Bettany, 1992). An exponential increase in soil respiration with respect to temperature is commonly accepted (O’Connell, 1990; Thierron & Laudelout, 1996; Winkler et al., 1996). However, an exponential model could express the relationship between soil respiration and temperature under laboratory conditions, but not in field campaigns. At a high temperature, the sensitivity of soil respiration to temperature might be reduced in a field campaign. Enzymes might be deactivated or destroyed by a further high temperature.

Although the Arrhenius equation has a basis in thermodynamics, it might somewhat oversimplify the response mechanism of soil respiration to temperature (Ellert & Bettany, 1992). This equation uses the reciprocal of absolute temperature to predict the variation in respiration rate, suggesting that it might not be sensitive enough to the variation in soil respiration when temperature is low (Fang & Moncrieff, 2001). Lloyd and Taylor (1994) pointed out that a high value of activation energy under low temperature leads to an unbiased simulation of the equation.

The Q₁₀ value, which defines the temperature dependence or sensitivity to temperature variation of soil respiration, when derived from different models is different with respect to magnitude or temperature (Table 1). Comparing the fitness and Q₁₀ values for different models, it was obvious that a good fit between a model and experimental data does not ensure the estimation of a suitable Q₁₀ from the model. Different models should be carefully dealt with because the Q₁₀ for these models are case-dependent, and an unreasonable Q₁₀ might be derived. A Q₁₀ analysis might be a useful approach for identifying a suitable model for simulating the variation of soil respiration with temperature (Fang & Moncrieff, 2001).

Soil water content was significantly correlated with soil temperature at our site and covaried across seasons (Figure 2). It is very difficult to distinguish the response of soil respiration to soil temperature and water content, as the soils vary with seasons. Therefore, the model considering only one single factor might ignore the influence of another factor to a certain extent. However, the mixed models could reflect the effects of two factors on soil CO₂ flux (Figure 7). Silletta et al. (2019) pointed out that the seasonal variations in soil respiration were mainly explained by the interaction between soil temperature and water content at a Patagonian site in Argentina. In our site, the warmer and wetter soil climate in the wet season increased soil respiration. However, when the moisture exceeds the optimal range, soil respiration becomes limited, offsetting the positive effects caused by the elevated soil temperature and reducing total respiration. At low soil temperatures, a clear trend of the soil respiration decreasing with increasing soil moisture was observed.

Implications for Conservation

The temporal variations and its determinants of soil respiration were investigated in a tropical lowland rainforest in Hainan Island. We found that (a) soil respiration varied significantly with the seasonal change in soil temperature and moisture; (b) soil respiration in the studied tropical lowland rainforest was co-controlled by the two factors of soil temperature and moisture, and the quadratic mixed model could best explain the effects. These two findings are highlighted as important for reducing errors in soil respiration evaluation and scaling up of soil C flux in climate change studies.

Supplemental Material

TRC914902 Supplementary material - Supplemental material for Controls of Temporal Variations on Soil Respiration in a Tropical Lowland Rainforest in Hainan Island, China

Supplemental material, TRC914902 Supplementary material for Controls of Temporal Variations on Soil Respiration in a Tropical Lowland Rainforest in Hainan Island, China by Yi-Bin Cui, Ji-Guang Feng, Li-Guo Liao, Rui Yu, Xiang Zhang, Yu-Hai Liu, Lian-Yan Yang, Jun-Fu Zhao and Zheng-Hong Tan in Tropical Conservation Science

Equation	Total	Dry season	Wet season
Linear	NA	2.14	7.67	1.87	0.27	NA
Exponential	3.00	3.00	2.46	2.46	NA	NA
Quadratic	1.26	2.58	0.52	3.42	0.18	NA
Arrhenius	16.96	2.57	7.46	1.95	NA	11.76
Lloyd and Taylor	2.30	1.85	2.30	1.85	2.36	1.80

Regression model type	Equation	Fitted parameters	R ²	F	Akaike information criterion
Single-factor models of soil moisture	Linear: $R_{s} = a + b W$	a = 0.05, b = 1.82	.08	4	−11.2
Power: $R_{s} = a W^{b}$	a = 1.20, b = 0.30	.12	202	−12.9
Quadratic: $R_{s} = a + b W + c W^{2}$	a = −0.85, b = 0.50, c = −0.02	.29	8	−22.5
Single-factor models of soil temperature	Linear: $R_{s} = a + b T$	a = −2.25, b = 0.20	.52	47	−40.3
Exponential: $R_{s} = a e^{b T}$	a = 0.17, b = 0.11	.59	464	−47.9
Quadratic: $R_{s} = a + b T + c T^{2}$	a = 8.56, b = −0.81, c = 0.03	.62	35	−51.2
Arrhenius: $R_{s} = a e^{(- \frac{E_{a}}{8.314 T})}$	a = 26.08, E_a = 470.69	.53	403	−41.8
Lloyd & Taylor (1994): $R_{s} = R_{10} e^{308.56 (\frac{1}{56} - \frac{1}{T + 46})}$	R₁₀ = 0.84	.51	774	−39.2
Mixed models of soil temperature and moisture	Linear 1: $R_{s} = a + b (T + W)$	a = 0.17, b = 0.06	.25	246	−21.2
Linear 2: $R_{s} = a + b T + c W$	a = −2.64, b = 0.24, c = −0.05	.55	422	−43,9
Quadratic: $R_{s} = a + b T + c W$ $+ d T^{2} + e W^{2} + fTW$	a = 6.97, b = −0.55, c = −0.29, d = 0.01, e = 0.02, f = 0.03	.74	300	−72.0
Power: $R_{s} = a T^{b} w^{c}$	a = 0.01, b = 1.87, c = −0.08	.54	411	−42.8
Exponential: $R_{s} = a e^{b T} w^{c}$	a = 0.18, b = 0.12, c = −0.16	.60	322	−50.7

Footnotes

Acknowledgments

We sincerely thank the national nature reserve of Diaoluoshan, especially Ms. Hai-yan Deng and Mr. Hai-wei Liu for their help with our field work.

Declaration of Conflicting Interests

The author(s) declared no potential conﬂicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by National Natural Science Foundations of China (grant nos. 41771099, 41861023, and 31660142) and C-Talent project of Hainan University (KYQD1626).

ORCID iDs

Yi-Bin Cui

Ji-Guang Feng

Supplemental Material

Supplemental material for this article is available online.

References

Adachi

Ishida

Bunyavejchewin

Okuda

Koizumi

(2009). Spatial and temporal variation in soil respiration in a seasonally dry tropical forest, Thailand. Journal of Tropical Ecology, 25(5), 531–539.

Anderson

B. T.

(2012). Intensification of seasonal extremes given a 2 C global warming target. Climatic Change, 112(2), 325–337.

Beer

Reichstein

Tomelleri

Ciais

Jung

Carvalhais

Rödenbeck

Altaf Arain

Baldocchi

Bonan

G. B.

Bondeau

Cescatti

Lasslop

Lindroth

Lomas

Luyssaert

Margolis

Oleson

K. W.

Roupsard

, . . . Papale

(2010). Terrestrial gross carbon dioxide uptake: Global distribution and covariation with climate. Science, 329(5993), 834–838.

Bond-Lamberty

Thomson

(2010). Temperature-associated increases in the global soil respiration record. Nature, 464(7288), 579.

Bucci

S. J.

Scholz

F. G.

Peschiutta

M. L.

Arias

N. S.

Meinzer

F. C.

Goldstein

(2013). The stem xylem of P atagonian shrubs operates far from the point of catastrophic dysfunction and is additionally protected from drought‐induced embolism by leaves and roots. Plant, Cell & Environment, 36(12), 2163–2174.

Carey

J. C.

Tang

Templer

P. H.

Kroeger

K. D.

Crowther

T. W.

Burton

A. J.

Dukes

J. S.

Emmett

Frey

S. D.

Heskel

M. A.

Jiang

Machmuller

M. B.

Mohan

Panetta

A. M.

Reich

P. B.

Reinsch

Wang

Allison

S. D.

Bamminger

Tietema

(2016). Temperature response of soil respiration largely unaltered with experimental warming. Proceedings of the National Academy of Sciences, 113(48), 13797–13802.

Carlyle

J. T.

Than

U. B.

(1988). Abiotic controls of soil respiration beneath an eighteen-year-old Pinus radiata stand in south-eastern Australia. The Journal of Ecology, 76(3), 654–662.

Cavaleri

M. A.

Reed

S. C.

Smith

W. K.

Wood

T. E.

(2015). Urgent need for warming experiments in tropical forests. Global Change Biology, 21(6), 2111–2121.

Chambers

J. Q.

Tribuzy

E. S.

Toledo

L. C.

Crispim

B. F.

Higuchi

Santos

J. D.

Araújo

A. C.

Kruijt

Nobre

A. D.

Trumbore

S. E.

(2004). Respiration from a tropical forest ecosystem: Partitioning of sources and low carbon use efficiency. Ecological Applications, 14(sp4), 72–88.

10.

Cook

F. J.

Orchard

V. A.

(2008). Relationships between soil respiration and soil moisture. Soil Biology and Biochemistry, 40(5), 1013–1018.

11.

Davidson

E. A.

Belk

Boone

R. D.

(1998). Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biology, 4(2), 217–227.

12.

Davidson

E. A.

Samanta

Caramori

S. S.

Savage

(2012). The Dual Arrhenius and Michaelis–Menten kinetics model for decomposition of soil organic matter at hourly to seasonal time scales. Global Change Biology, 18(1), 371–384.

13.

Davidson

E. A.

Verchot

L. V.

Cattanio

J. H.

Ackerman

I. L.

Carvalho

(2000). Effects of soil water content on soil respiration in forests and cattle pastures of eastern Amazonia. Biogeochemistry, 48(1), 53–69.

14.

Diffenbaugh

N. S.

Scherer

(2011). Observational and model evidence of global emergence of permanent, unprecedented heat in the 20th and 21st centuries. Climatic Change, 107(3–4), 615–624.

15.

Ellert

Bettany

(1992). Temperature dependence of net nitrogen and sulfur mineralization. Soil Science Society of America Journal, 56(4), 1133–1141.

16.

Fang

Moncrieff

J. B.

(1999). A model for soil CO₂ production and transport 1: Model development. Agricultural and Forest Meteorology, 95(4), 225–236.

17.

Fang

Moncrieff

J. B.

(2001). The dependence of soil CO₂ efflux on temperature. Soil Biology and Biochemistry, 33(2), 155–165.

18.

Goodrick

Connor

Bird

Nelson

(2016). Emission of CO₂ from tropical riparian forest soil is controlled by soil temperature, soil water content and depth to water table. Soil Research, 54(3), 311–320.

19.

Hanson

Edwards

Garten

C. T.

Andrews

(2000). Separating root and soil microbial contributions to soil respiration: A review of methods and observations. Biogeochemistry, 48(1), 115–146.

20.

Högberg

Read

D. J.

(2006). Towards a more plant physiological perspective on soil ecology. Trends in Ecology & Evolution, 21(10), 548–554.

21.

Holthausen

R. S.

Caldwell

M. M.

(1980). Seasonal dynamics of root system respiration in Atriplex confertifolia. Plant and Soil, 55(2), 307–317.

22.

IPCC Climate Change. (2013). The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Technical Report. IPCC, Cambridge University Press.

23.

Jiang

Zhou

Zheng

Jiang

Cai

Zhu

Fang

(2016). Soil respiration and its partitioning in different components in tropical primary and secondary mountain rain forests in Hainan Island, China. Journal of Plant Ecology, 10(5), 791–799.

24.

Jobbágy

E. G.

Jackson

R. B.

(2000). The vertical distribution of soil organic carbon and its relation to climate and vegetation. Ecological Applications, 10(2), 423–436.

25.

Kuzyakov

Gavrichkova

(2010). Time lag between photosynthesis and carbon dioxide efflux from soil: A review of mechanisms and controls. Global Change Biology, 16(12), 3386–3406.

26.

Yang

Fang

(2013). Variations of root and heterotrophic respiration along environmental gradients in China’s forests. Journal of Plant Ecology, 6(5), 358–367.

27.

Zou

(2006). Heterotrophic soil respiration in relation to environmental factors and microbial biomass in two wet tropical forests. Plant and Soil, 281(1–2), 193–201.

28.

Linn

D. M.

Doran

J. W.

(1984). Effect of water-filled pore space on carbon dioxide and nitrous oxide production in tilled and non-tilled soils. Soil Science Society of America Journal, 48(6), 1267–1272.

29.

Liptzin

Silver

W. L.

Detto

(2011). Temporal dynamics in soil oxygen and greenhouse gases in two humid tropical forests. Ecosystems, 14(2), 171–182.

30.

Liu

Wang

Lajeunesse

M. J.

Miao

Piao

Wan

Wang

Yang

Deng

(2016). A cross‐biome synthesis of soil respiration and its determinants under simulated precipitation changes. Global Change Biology, 22(4), 1394–1405.

31.

Lloyd

Taylor

(1994). On the temperature dependence of soil respiration. Functional Ecology, 8(3), 315–323.

32.

Lükewille

Wright

(1997). Experimentally increased soil temperature causes release of nitrogen at a boreal forest catchment in southern Norway. Global Change Biology, 3(1), 13–21.

33.

MacDonald

N. W.

Zak

D. R.

Pregitzer

K. S.

(1995). Temperature effects on kinetics of microbial respiration and net nitrogen and sulfur mineralization. Soil Science Society of America Journal, 59(1), 233–240.

34.

Manzoni

Schimel

J. P.

Porporato

(2012). Responses of soil microbial communities to water stress: Results from a meta-analysis. Ecology, 93(4), 930–938.

35.

Mora

Frazier

A. G.

Longman

R. J.

Dacks

R. S.

Walton

M. M.

Tong

E. J.

Sanchez

J. J.

Kaiser

L. R.

Stender

Y. O.

Anderson

J. M.

Ambrosino

C. M.

Fernandez-Silva

Giuseffi

L. M.

Giambelluca

T. W.

(2013). The projected timing of climate departure from recent variability. Nature, 502(7470), 183.

36.

Ngo

K. M.

Turner

B. L.

Muller-Landau

H. C.

Davies

S. J.

Larjavaara

bin Nik Hassan

N. F.

Lum

(2013). Carbon stocks in primary and secondary tropical forests in Singapore. Forest Ecology and Management, 296, 81–89.

37.

O’Connell

(1990). Microbial decomposition (respiration) of litter in eucalypt forests of south-western Australia: An empirical model based on laboratory incubations. Soil Biology and Biochemistry, 22(2), 153–160.

38.

Pan

Birdsey

R. A.

Fang

Houghton

Kauppi

P. E.

Kurz

W. A.

Phillips

O. L.

Shvidenko

Lewis

S. L.

Canadell

J. G.

Ciais

Jackson

R. B.

Pacala

S. W.

McGuire

A. D.

Piao

Rautiainen

Sitch

(2011). A large and persistent carbon sink in the world’s forests. Science, 333(6045), 988–993.

39.

Pereyra

D. A.

Bucci

S. J.

Arias

N. S.

Ciano

Cristiano

P. M.

Goldstein

Scholz

F. G.

(2017). Grazing increases evapotranspiration without the cost of lowering soil water storages in arid ecosystems. Ecohydrology, 10(6), e1850.

40.

Raich

J. W.

(1998). Aboveground productivity and soil respiration in three Hawaiian rain forests. Forest Ecology and Management, 107(1–3), 309–318.

41.

Rochette

Desjardins

Pattey

(1991). Spatial and temporal variability of soil respiration in agricultural fields. Canadian Journal of Soil Science, 71(2), 189–196.

42.

Rubio

V. E.

Detto

(2017). Spatiotemporal variability of soil respiration in a seasonal tropical forest. Ecology and Evolution, 7(17), 7104–7116.

43.

Ryan

M. G.

Law

B. E.

(2005). Interpreting, measuring, and modeling soil respiration. Biogeochemistry, 73(1), 3–27.

44.

Scholz

F. G.

Bucci

S. J.

Arias

Meinzer

F. C.

Goldstein

(2012). Osmotic and elastic adjustments in cold desert shrubs differing in rooting depth: Coping with drought and subzero temperatures. Oecologia, 170(4), 885–897.

45.

Schwendenmann

Veldkamp

Brenes

O’brien

J. J.

Mackensen

(2003). Spatial and temporal variation in soil CO₂ efflux in an old-growth neotropical rain forest, La Selva, Costa Rica. Biogeochemistry, 64(1), 111–128.

46.

Sha

Zheng

Tang

Wang

Zhang

Cao

Wang

Liu

G. G.

Wang

Y. S.

Sun

(2005). Soil respiration in tropical seasonal rain forest in Xishuangbanna, SW China. Science in China Ser D, 48(S1), 189–197.

47.

Shibistova

Lloyd

Evgrafova

Savushkina

Zrazhevskaya

Arneth

Knohl

Kolle

Schulze

E.-D.

(2002). Seasonal and spatial variability in soil CO₂ efflux rates for a central Siberian Pinus sylvestris forest. Tellus B: Chemical and Physical Meteorology, 54(5), 552–567.

48.

Silletta

L. C.

Cavallaro

Kowal

Pereyra

D. A.

Silva

R. A.

Arias

N. S.

Goldstein

Scholz

F. G.

Bucci

S. J.

(2019). Temporal and spatial variability in soil CO₂ efflux in the Patagonian steppe. Plant and Soil, 444(1), 165–176.

49.

Sotta

E. D.

Meir

Malhi

Donato Nobre

Hodnett

Grace

(2004). Soil CO₂ efflux in a tropical forest in the central Amazon. Global Change Biology, 10(5), 601–617.

50.

Sotta

E. D.

Veldkamp

Guimaraes

Paixao

Ruivo

Almeida

(2006). Landscape and climatic controls on spatial and temporal variation in soil CO₂ efflux in an Eastern Amazonian Rainforest, Caxiuana, Brazil. Forest Ecology and Management, 237(1–3), 57–64.

51.

Sotta

E. D.

Veldkamp

Schwendenmann

Guimaraes

B. R.

Paixao

R. K.

Ruivo

M. D. L. P.

Lola da Costa

A. C.

Meir

(2007). Effects of an induced drought on soil carbon dioxide (CO₂) efflux and soil CO₂ production in an Eastern Amazonian rainforest. Brazil. Global Change Biology, 13(10), 2218–2229.

52.

Thierron

Laudelout

(1996). Contribution of root respiration to total CO₂ efflux from the soil of a deciduous forest. Canadian Journal of Forest Research, 26(7), 1142–1148.

53.

Vargas

Detto

Baldocchi

D. D.

Allen

M. F.

(2010). Multiscale analysis of temporal variability of soil CO₂ production as influenced by weather and vegetation. Global Change Biology, 16(5), 1589–1605.

54.

Winkler

J. P.

Cherry

R. S.

Schlesinger

W. H.

(1996). The Q₁₀ relationship of microbial respiration in a temperate forest soil. Soil Biology and Biochemistry, 28(8), 1067–1072.

55.

Dijkstra

Koch

G. W.

Peñuelas

Hungate

B. A.

(2011). Responses of terrestrial ecosystems to temperature and precipitation change: A meta‐analysis of experimental manipulation. Global Change Biology, 17(2), 927–942.

56.

(2001). Temperatures are negatively correlated through a simple linear relationship with R2 values moisture are positively correlated with R² values Q₁₀. Global Biogeochemical Cycles, 15(3), 687–696.

57.

Shang

(2016). Contribution of soil respiration to the global carbon equation. Journal of Plant Physiology, 203, 16–28.

58.

Yang

Lin

Liang

(1994). Forest vegetation in Wuzhishan, Hainan, I. Forest vegetation type in Wuzhishan. Natural Science Journal of Hainan University, 12, 220–234. (in Chinese)

59.

Zhao

Wang

Deng

Wang

Liu

Yang

Tan

(2019). Biomass increment and mortality losses in tropical secondary forests of Hainan, China. Journal of Forestry Research, 30(2), 647–655.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB

Equation	Total		Dry season		Wet season
Equation	Q10 at 10°C	Q10 at 20°C	Q10 at 10°C	Q10 at 20°C	Q10 at 10°C	Q10 at 20°C
Linear	NA	2.14	7.67	1.87	0.27	NA
Exponential	3.00	3.00	2.46	2.46	NA	NA
Quadratic	1.26	2.58	0.52	3.42	0.18	NA
Arrhenius	16.96	2.57	7.46	1.95	NA	11.76
Lloyd and Taylor	2.30	1.85	2.30	1.85	2.36	1.80

Controls of Temporal Variations on Soil Respiration in a Tropical Lowland Rainforest in Hainan Island,China

Abstract

Keywords

Methods

Site Description

Experiment Design

Field Measurements

Statistics

Regression Models

Single-Factor Models of Soil Respiration and Moisture

Single-Factor Models of Soil Respiration and Temperature

Q10 Value

Mixed Models of Soil Respiration Considering Soil Temperature and Moisture

Results

Seasonal Variation in Soil Respiration and Environmental Factors

Diurnal Variation in Soil Respiration

Seasonal Controls of Soil Respiration

Relationship Between Soil Respiration and Moisture

Relationship Between Soil Respiration and Temperature

Effect of Different Regression Models on Q10

Combined Effect of Soil Temperature and Moisture on Soil Respiration

Discussion

Temporal Variations of Soil Respiration Rates

Effects of Environmental Factors on Soil Respiration

Selection of Soil Respiration Models

Implications for Conservation

Supplemental Material

TRC914902 Supplementary material - Supplemental material for Controls of Temporal Variations on Soil Respiration in a Tropical Lowland Rainforest in Hainan Island, China

Footnotes

Acknowledgments

Declaration of Conflicting Interests

Funding

ORCID iDs

Supplemental Material

References

Supplementary Material

Q₁₀ Value

Effect of Different Regression Models on Q₁₀