CO2 fluxes from soil, together with soil water content and temperature have been measured over one solar year in an even-aged beech forest (
Respiration is one of the largest fluxes in the global carbon (C) cycle, emitting approximately 10 times more C than that released through fossil fuel combustion and cement manufacturing (
The factors influencing soil respiration are temperature, soil moisture, photosynthetic supply to roots, substrate quantity and availability (
Given the temperature dependence of soil respiration, a potential positive feedback between increasing temperature and enhanced soil respiration may ultimately accelerate global warming (
From May 2014 to April 2015, within ManFor.BD Life project, CO2 fluxes from soil, together with soil water content and temperature have been measured over one solar year in an even-aged beech forest (
This experimental study was conceived to evaluate the ecological effects of management, with a focus on the influence on carbon uptake/loss from forest soil. We investigated the hypothesis that an innovative silviculture based on tree-oriented silviculture, in addition to being a sustainable treatment which modifies growing spaces and spatial niches to increase heterogeneity of the stand structure, can also reduce C outputs from soil.
The main aim of this study was to evaluate the effect of such silvicultural practices on the CO2 respired from the forest floor. A secondary objective was to analyse the influence of soil temperature and soil moisture on soil respiration.
The study area is located in the Marchesale Biogenetic Reserve (a Natura 2000 site) in Mongiana, the highest side of calabrian “Serre” mountains, within the province of Vibo Valentia (38° 30′ N, 16° 14′ E). The whole reserve is 1257 ha wide, with altitude ranging between 750 and 1170 m a.s.l. and it is managed by the National Forest Service of Italy (CFS). The forest types are beech forest managed as high forest and chestnut stands managed as coppice. In the area there is a small fraction of mixed beech-fir high forest (5%). The understorey is mainly composed of penciled geranium (
The study area is characterized by Palaeozoic granitoid rocks deeply fractured, the morphology is dominated by a mountains landscape with deep, V-shaped valleys (
Within the ManFor C.BD. project, a 30 ha stand of 75-year-old high forest of beech at 1100 m elevation was selected and nine areas from 2.8 to 3.5 ha were subjected to three silvicultural thinnings (3 treatments × 3 replicates; 2012-2013 -
The options were traditional treatments (thinning from below - AT6), innovative treatments based on the tree-oriented silviculture (AI2) and unharvested control (AC3 -
Soil temperature was measured adjacent to each PVC collar at the time of the flux measurement, with a digital soil thermometer (Spectrum Technologies, Aurora, Illinois, USA). The depth of the measurement was 10 cm below the top of the litter layer. Soil water content was measured by time domain reflectometry (TDR) with FieldScout TDR 100 Soil Moisture Meter (Spectrum Technologies, Aurora, IL, USA), with two-rod probes 12 cm long placed vertically into the soil in a randomly selected spot around each collar during each CO2 measurement. The output of the soil moisture meter was the percentage volume water content. Daily precipitation were obtained from a weather station installed at the launch of the project in the core of the high forest beech stand under consideration, at a distance of less than 1 kilometer from plots.
Respiration rate (
where
The relationship between soil temperature and soil respiration rate was fitted with an exponential model (
where
In accordance with
where
One-way ANOVA test was used to compare
After determination of the coefficient of variation (cv%) between measurements at the six collars positions in each circular plot, the mean of respiration, temperature and moisture of the six collars were used to represent every plot.
The effect of treatments on soil temperature and soil water content was analyzed with a one-way ANOVA.
A repeated measures ANOVA, with significant differences determined at α<0.05, was used to test the differences of soil respiration between treatments. A Mauchly’s test of sphericity was carried out to test homogeneity of variances.
The model was set up with the respiration rate measurements (average of the 18 measurements from collars for any treatment) as the dependent variable, the treatments as a “between-subjects” factor, soil temperature and soil moisture as covariates. The interaction between treatment and date was also considered in the model. A
A non-linear regression analysis was used to evaluate the combined effect of soil temperature and soil water content on soil respiration for each treatment. The estimate of the parameters of the nonlinear regression model was based on the Gauss-Newton method (
Statistical analyses were conducted using SPSS statistics ver. 21 for Windows (IBM Corp., Armonk, NY, USA).
Soil respiration varied significantly over the year (p<0.001) for all treatments (AI2: from 0.91 to 4.2 ± 0.85 µmol CO2 m-2 s-1; AC3: from 0.73 to 3.36 ± 0.66 µmol CO2 m-2 s-1; AT6: from 0.98 to 2.90 ± 0.57 µmol CO2 m-2 s-1). Soil temperature and soil moisture also showed a significant seasonal variability (p<0.001).
Respiration was inhibited when soil temperature ranged between 0 and 5 °C with a soil moisture ranging between 10 and 20%. The lowest efflux occurred when soil temperature was under 10 °C and soil water content was about 16% in AI2 and AC3. In AT6 the lowest
The one-way ANOVA showed that no significant effect of treatments on soil temperature was found (F2;708=0.537; p=0.585), but treatments had a significant effect on soil water content (F2;708=3.670; p = 0.026).
Repeated measures ANOVA analysis showed that there were no significant differences in soil temperature between the treatments (p > 0.05), whereas soil water content was statistically different only in AI2 (p<0.001).
During the experimental period
The combined effect of soil temperature and water content gave a better description of the variability in soil respiration in treatment AI2 (R2 = 0.48), compared to AC3 (R2 = 0.43) and AT6 (R2 = 0.38). For each treatment, regression analysis resulted significant, and the parameters of the regressions were all significant (
The developed models are shown in
Sphericity was assumed since Mauchly’s test gave a p-value of 0.396.
High temporal and spatial variation in soil CO2 efflux is a common phenomenon in forests (
Forest management practices such as thinning affect stand biomass and carbon stored in soil, influencing the interaction between the biotic (roots, invertebrates and microorganisms) and abiotic (temperature and moisture) factors of the forest ecosystem. Both increases and decreases in soil respiration after thinning have been observed in previous studies (
The increase in soil respiration in the treated areas in this study could be attributed to a greater increase in heterotrophic respiration from the above- and belowground debris than the reduction in autotrophic (root) respiration (
This study showed that the mean soil respiration increased with thinning intensity, confirming that after harvesting, detrital biomass remaining on the forest floor may contribute to raise soil respiration, also preventing evaporation of water from soil, and leading therefore to an higher microbial activity. This information is useful for forest managers in predicting the consequences of forest management given the efflux of CO2 from the soil surface of our beech forest. However, it would be interesting to carry on measurements in these sites to analyse long term post-harvest effects. Forest management practices, together with other achievements, must be oriented to enhance carbon sequestration; it is fundamental to better know the effects of forest management on CO2 cycle, considering all the components and their interactions. Therefore, studies examining patterns and mechanisms of soil CO2 in response to management practices are essential. Such studies are critical to accurately predict effects of forest management on the C cycle and to develop appropriate forest management strategies aimed at reducing atmospheric CO2 concentrations (
Representation of the nine areas managed with three different treatments (AI2 innovative, AC3 control and AT6 traditional) and the sampling circular plots. Areas where soil respiration measurements were carried out are within the circles.
(a) Averaged daily measurements of soil respiration (µmol CO2 m2 s-1), (b) soil temperature, (c) soil water content with standard errors (error bars) in AI2 (innovative), AT6 (traditional), AC3 (control) treatments.
Soil respiration (µmol CO2 m2 s-1) and total monthly precipitation (mm).
CO2 (µmol CO2 m2 s-1) efflux with standard errors (error bars), recorded when soil water content (SWC) ranged from 0 to 40 % and soil temperature (Soil T) ranged from 0 to 20 °C.
Relationship between soil respiration and soil temperature in AT6 (traditional), AC3 (control), AI2 (innovative) treatments. The points represent the averaged measures taken at the collars for each sub-plot during sampling days.
The combined effect of soil water content and soil temperature on the soil respiration in AC3 (control) AT6 (traditional), AI2 (innovative) treatments.
Main dendrometric characteristics (± standard deviation) of AI2 (innovative), AT6 (traditional), AC3 (control) treatments.
Treatment | Volume(m3 ha-1) | Basal area(m2 ha-1) | No. tree(N ha-1) | |
---|---|---|---|---|
AI2 (innovative) | Before | 338.1 ± 27.7 | 33.9 ± 1.9 | 631 ± 104.4 |
After | 246.1 ± 21.3 | 26.6 ± 1.6 | 533 ± 94.3 | |
AT6 (traditional) | Before | 345.7 ± 19.8 | 34.0 ± 1.2 | 557 ± 138.7 |
After | 301.6 ± 15.6 | 29.7 ± 1.0 | 501 ± 111.5 | |
AC3 (control) | - | 432.9 ± 174.1 | 38.5 ± 11.4 | 334 ± 201.2 |
Parameters and coefficients of the regression analysis (
Treatments | Coefficient | Estimate | SE | SEE | RMSE | R2 |
---|---|---|---|---|---|---|
AC3 | 0.6413 | 0.2073 | 0.896 | 0.846 | 0.429 | |
0.0802 | 0.0127 | |||||
-0.0147 | 0.0025 | |||||
0.0007 | <0.0001 | |||||
AI2 | 0.5049 | 0.1539 | 0.957 | 0.958 | 0.485 | |
0.0887 | 0.0112 | |||||
0.0051 | 0.0013 | |||||
0.0005 | <0.0001 | |||||
AT6 | 0.6784 | 0.2137 | 0.886 | 0.812 | 0.383 | |
0.0706 | 0.0160 | |||||
-0.0152 | 0.0024 | |||||
0.0009 | <0.0001 |