| 1. |
- He, Hongxing, 1987, et al.
(författare)
-
Factors controlling Nitrous Oxide emission from a spruce forest ecosystem on drained organic soil, derived using the CoupModel
- 2016
-
Ingår i: Ecological Modelling. - : Elsevier. - 0304-3800 .- 1872-7026. ; 321, s. 46-63
-
Tidskriftsartikel (refereegranskat)abstract
- High Nitrous Oxide (N2O) emissions have been identified in hemiboreal forests in association with draining organic soils. However, the specific controlling factors that regulate the emissions remain unclear. To examine the importance of different factors affecting N2O emissions in a spruce forest on drained organic soil, a process-based model, CoupModel, was calibrated using the generalized likelihood uncertainty estimation (GLUE) method. The calibration also aims to estimate parameter density distributions, the covariance matrix of estimated parameters and the correlation between parameters and variables information, useful when applying the model on other peat soil sites and for further model improvements. The calibrated model reproduced most of the high resolution data (total net radiation, soil temperature, groundwater level, net ecosystem exchange, etc.) very well, as well as cumulative measured N2O emissions (simulated 8.7±1.1kgN2Oha-1year-1 (n=97); measured 8.7±2.7kgN2Oha-1year-1 (n=6)), but did not capture every measured peak. Parameter uncertainties were reduced after calibration, in which 16 out of 20 parameters changed from uniform distributions into normal distributions or log normal distributions. Four parameters describing bypass water flow, oxygen diffusion and soil freezing changed significantly after calibration. Inter-connections and correlations between many calibrated parameters and variables reflect the complex and interrelated nature of pedosphere, biosphere and atmosphere interactions. This also highlights the need to calibrate a number of parameters simultaneously. Model sensitivity analysis indicated that N2O emissions during growing seasons are controlled by competition between plants and microbes for nitrogen, while during the winter season snow melt periods are important. Our results also indicate that N2O is mainly produced in the capillary fringe close to the groundwater table by denitrification in the anaerobic zone. We conclude that, in afforested drained peatlands, the plants and groundwater level have important influences on soil N availability, ultimately controlling N2O emissions.
|
|
| 2. |
- He, Hongxing, 1987, et al.
(författare)
-
Modeling Nitrous Oxide emissions and identifying emission controlling factors for a spruce forest ecosystem on drained organic soil
- 2013
-
Ingår i: European Geosciences Union, Vienna Austria 07 – 12 April 2013.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- High nitrous oxide (N2O) emission potential has been identified in hemiboreal forest on drained Histosols. However, the environmental factors regulating the emissions were unclear. To investigate the importance of different factors on the N2O emission, a modeling approach was accomplished, using CoupModel with Monti-Carlo based multi-criteria calibration method. The model was made to represent a forest on drained peat soil in south-west Sweden where data of fluxes combined with soil properties and plant conditions were used. The model outcome was consistent with measurements of abiotic (soil temperature, net radiation, groundwater level and soil moisture) and biotic responses (net ecosystem exchange and soil respiration). Both dynamics and magnitude of N2O emissions were well simulated compared to measurements (8.7±2.1 kg N/ha/year). The performance indicators for an ensemble of accepted simulations of N2O emission dynamics and magnitudes were correlated to calibrated parameters related to soil anaerobic fraction and atmospheric nitrogen deposition (correlation coefficient, r≥0.4). A weak correlation with N2O emission dynamics was also found for biotic responses (r≥0.3). However, the ME of simulated and measured N2O emissions was better correlated to the ME of soil moisture (r=-0.6), and also to the ME of both the soil temperature (r=0.53) and groundwater level (r=-0.7). Groundwater level (range from -0.8m to -0.13m) was identified as the most important environmental factor regulating the N2O emissions for present forest soil. Profile analysis indicated that N2O was mainly produced in the deeper layers (≥0.35m) of the soil profile. The optimum soil moisture for N2O production was around 70%.
|
|
| 3. |
- He, Hongxing, 1987, et al.
(författare)
-
Modeling Nitrous Oxide emissions and identifying emission controlling factors for a spruce forest ecosystem on drained organic soil
- 2015
-
Ingår i: Geophysical Research Abstracts. ; 17:EGU2015-10451
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- High Nitrous Oxide (N2O) emission has been identified in hemiboreal forests on drained organic soils. However, the controlling factors regulating the emissions have been unclear. To examine the importance of different factors on the N2O emission in a spruce forest on drained organic soil, a process-based model, CoupModel, was calibrated by the generalized likelihood uncertainty estimation (GLUE) method. The calibrated model reproduced most of the high resolution data (total net radiation, soil temperature, groundwater level, net ecosystem exchange, etc.) very well, as well as accumulated measured N2O emissions, but showed difficulties to capture all the measured emission peaks. Parameter uncertainties could be reduced by combining selected criteria with the measurement data. The model showed the N2O emissions during the summer to be controlled mainly by the competition between plants and microbes while during the winter season snow melt periods are important. The simulated N budget shows >100 kg N ha-1 yr-1 to be in circulation between soil and plants and back again. Each year the peat mineralization adds about 60 kg N ha-1 and atmospheric deposition 12 kg N ha-1. Most of the mineralized litter and peat N is directly taken up by the plants but only a part accumulates in the plant biomass. As long as no timber is harvested the main N loss from the system is through nitrate leaching (30 kg N ha-1 yr-1) and gas emissions (20 kg N ha-1 yr-1), 55% as NO, 27% as N2O and 18% as N2. Regarding N2O gas emissions, our modeling indicates denitrification to be the most responsible process, of the size 6 kg N ha-1 yr-1, which could be compared to 0.04 kg N ha-1 yr-1 from nitrification. Our modelling also reveal 88% of the N2O mainly to be produced by denitrification in the capillary fringe (c.a. 40-60 cm below soil surface) of the anaerobic zone using nitrate produced in the upper more aerobic layers. We conclude N2O production/emission to be controlled mainly by the complex interaction between soil N availability, mediated by mineralization, nitrification, and plant growth together with soil anaerobicity controlled by the groundwater level. The model is currently used for modelling greenhouse gas emissions from drained organic soils over the entire forest cycle, from plantation to harvest. Different land use and plant production are compared like Spruce, Willow and Reed Canary Grass as well as rewetting options.
|
|
| 4. |
- Meyer, Astrid, et al.
(författare)
-
A fertile peatland forest does not constitute a major greenhouse gas sink
- 2013
-
Ingår i: Biogeosciences. - : Copernicus GmbH. - 1726-4170 .- 1726-4189. ; 10
-
Tidskriftsartikel (refereegranskat)abstract
- Afforestation has been proposed as a strategy to mitigate the often high greenhouse gas (GHG) emissions from agricultural soils with high organic matter content. However, the carbon dioxide (CO2) and nitrous oxide (N2O) fluxes after afforestation can be considerable, depending predominantly on site drainage and nutrient availability. Studies on the full GHG budget of afforested organic soils are scarce and hampered by the uncertainties associated with methodology. In this study we etermined the GHG budget of a spruce-dominated forest on a drained organic soil with an agricultural history. Two different approaches for determining the net ecosystem CO2 exchange (NEE) were applied, for the year 2008, one direct (eddy covariance) and the other indirect (analyzing the different components of the GHG budget), so that uncertainties in each method could be evaluated. The annual tree production in 2008 was 8.3±3.9 tC ha−1 yr−1 due to the high levels of soil nutrients, the favorable climatic conditions and the fact that the forest was probably in its phase of maximum C assimilation or shortly past it. The N2O fluxes were determined by the closed-chamber technique and amounted to 0.9±0.8 tCeq ha−1 yr−1. According to the direct measurements from the eddy covariance technique, the site acts as a minor GHG sink of −1.2±0.8 t Ceq ha−1 yr−1. This contrasts with the NEE estimate derived from the indirect approach which suggests that the site is a net GHG emitter of 0.6±4.5 tCeq ha−1 yr−1. Irrespective of the approach applied, the soil CO2 effluxes counter large amounts of the C sequestration by trees. Due to accumulated uncertainties involved in the indirect approach, the direct approach is considered the more reliable tool. As the rate of C sequestration will likely decrease with forest age, the site will probably become a GHG source once again as the trees do not compensate for the soil C and N losses. Also forests in younger age stages have been shown to have lower C assimilation rates; thus, the overall GHG sink potential of this afforested nutrient-rich organic soil is probably limited to the short period of maximum C assimilation.
|
|
