the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Methane emissions from Arctic landscapes during 2000–2015: An analysis with land and lake biogeochemistry models
- 1Department of Earth, Atmospheric, Planetary Sciences, Purdue University, West Lafayette, IN, USA
- 2Department of Agronomy, Purdue University, West Lafayette, IN, USA
- 3Purdue Climate Change Research Center, West Lafayette, IN, USA
- 1Department of Earth, Atmospheric, Planetary Sciences, Purdue University, West Lafayette, IN, USA
- 2Department of Agronomy, Purdue University, West Lafayette, IN, USA
- 3Purdue Climate Change Research Center, West Lafayette, IN, USA
Abstract. Wetlands and freshwater bodies (mainly lakes) are the largest natural source of greenhouse gas CH4 to the atmosphere. Great efforts have been made to quantify these source emissions and their uncertainties. Previous research suggests that there might be significant uncertainties coming from “double accounting” emissions from freshwater bodies and wetlands. Here we quantify the methane emissions from both land and freshwater bodies in the pan-Arctic with two process-based biogeochemistry models by minimizing the double accounting at the landscape scale. Two non-overlapping dynamic areal change datasets are used to drive the models. We estimate that the total methane emissions from pan-Arctic are 35.81 Tg CH4 yr−1 during 2000–2015, of which wetlands and freshwater bodies are 21.38 CH4 Tg yr−1 and 14.45 Tg CH4 yr−1, respectively. The emissions are significantly affected by humidity and vapor pressure, followed by temperature and landscape areal changes. Emissions from wetlands are more sensitive to landscape areal changes than from freshwater bodies while the latter emissions are more influenced by temperature than precipitation. Our sensitivity analysis indicates the pan-Arctic CH4 emissions from freshwater bodies were highly influenced by temperature, but less by lake sediment carbon content from permafrost thaw.
Xiangyu Liu and Qianlai Zhuang
Status: final response (author comments only)
-
RC1: 'Comment on bg-2022-220', Anonymous Referee #1, 14 Dec 2022
General comments:
This manuscript has done parallel modeling work using two models to estimate the methane emission from wetlands and lakes in the Arctic region. Two latest dynamic inundation data and lake distribution data were used to drive the model. The separation of the lake/wetland area in the dataset reduced the main source of uncertainty in previous bottom-up simulations for lake/wetland methane emissions called “double counting”, which is caused by massing up the lake/wetland distributions.
The objectives of this study are meaningful, trying to reduce the uncertainty of bottom-up methane estimation. I find the story inside complete and exciting. Moreover, the detailed analysis of different drivers of methane emission may contribute to the community of methane-related biogeochemistry modeling. The only part missing or not clear from my perspective uncertainty of the model simulation. Thus, I suggest a publication after minor revisions. Below are some of my comments that may help elucidate the strength and limitations of the proposed work.
Specific comments:
Line 16-17: What are the uncertainties?
Line 20-22: As previously mentioned the emissions are significantly affected by humidity and vapor pressure.
Line 27: IPCC 2014 is not the latest one. IPCC6 is published one year ago. What is the newest IPCC6 report number?
Line 29-35: I feel people will get confused about these introductions about wetland emissions. It would be more clear if put them together instead of two paragraphs.
Line 40-42: People may expect a little bit more mechanisms of lake emission. Like why ti will be high in spring?
Line 59-61: 1) As you mentioned the resolution will be the key issue for the double counting issue. Then what is the resolution for the data you used or what model resolution should be highlighted here. 2) Or if resolution is not the matter. You may need want to mention why involving those two datasets or models can avoid double counting.
Line 60: “unpublished data”. Is the data going to be published somewhere? How people are going to access them
Line 100-105: What are the input of vegetation types, soil types coming from?
Line 118-120: This part needs to be highlighted and extended for details.
Line 135: Here should be “wind speed at 10 m”
Line 137-138: Are they the same as what are inputted in ALBM?
Line 154: Should be “(using TEM-MDM)”
Line 158-159: Have you mentioned what data you used for spin-up?
Line 167: Under what scenarios? and we have the latest IPCC6 report, have you checked whether the number is changed?
Line 174-176: Why do you suddenly mention the isotope here? I don't think you have used any of the isotope-related analyses.
Line 182-183: What is the uncertainty of your simulation in each year? Some analyses like considering the uncertainty of input data may be good to be included.
Line 184: How are those uncertainties coming from?
Figure 1 and table 1: Fig1a is duplicated with table 1. Also, table 1 is more suitable for a time-series plot
Line 209-211: I don't understand here.
Figure 2: Please label the simulation type also in the figure instead of only the caption.
Table 2: why do some of the results have no superscript? Also, these superscripts should be explained in a footnote instead of the table title.
Table 3: why 0 values here?
Figure 4: So can I understand this way that the high correlation of radiation is caused by the heating effect of radiation and the high sensitivity of temperature in your model?
Line 264: I didn't see how your simulation results fit in the previous ranges. Could you summarize the results and also your results for comparison?
Line 275: This part is interesting and should be talked about more about why you can attribute this.
Line 288-290: This is a very interesting conclusion since we always know that there are huge differences between bottom-up modeling and top-down modeling results of CH4 emission. I may expect to see emphasizing more explicitly how the uncertainty is reduced based on your study and what ranges you suggested.
Line 296-298: Are they possibly due to your study area being mostly located in boreal so the seasonal cycle of all inputs and methane emissions are strong? Then they should be easily correlated. Their being highly correlated may not be simply explained as they are significant to wetland emission, since another possibility is that they can be confounders, correlated with each other (e,g, SR, and temperature). Causal relations may be considered here
-
RC2: 'Comment on bg-2022-220', Anonymous Referee #2, 27 Jan 2023
The manuscript of Liu and Zhuang covers an interesting topic appropriate for Biogeosciences. The authors addressed the importance of accounting for the methane emissions from wetlands and lakes using two processed-based models, with a few different wetland and lake extent products. The methodology is sound and the results are interesting. However, more in-depth discussions and some clarifications are needed. I have the following main questions:
1. Since reducing the uncertainties from ‘double accounting’ is one of the main focus for this manuscript, I wonder how large the amount was ‘double accounted’ and how large the estimate was improved by your approach. Could you elaborate in the abstract and main text?
2. The Lake with a size < 10 ha is the central component that affects estimating wetland and lake CH4 emissions simultaneously. It is unclear how you treat this since the HydroLakes doesn’t cover the extent for small lakes?
3. The statistical analysis needs to be improved. I am surprised to see shortwave radiation has a much higher correlation with CH4 emissions than the temperature for the Arctic wetlands and lakes on the annual basis. What’s the underlying mechanism for this? Also, these climate variables often co-vary over time. the analysis done in its current way may not reflect the ‘true’ sensitivity of CH4 fluxes to the climate variables.
4. The description of the sensitivity test is not clear. How did you treat the increasing temperature by 5 deg C and increasing precipitation by 15%? Was the increased temperature or precipitation evenly allocated to each month or proportional to its seasonal cycle? This is important because the added precipitation and temperature to different seasons would have different effects on CH4 emissions. Also, how the threshold for temperature and precipitation were chosen.
Specific comments:
Table 2. The way its presented is confusing. As I said, your might want to consider a multiple regression or partial correlation to analyze the sensitivity.
Figure 3. It is difficult to read as it is currently presented. Could you minimize the duplicated information by plotting the time series by different groups? Say Ch4 emissions and climate variables as two groups?
Discussion 4.1 it is not clear to me how your number narrows down the double counting.
4.2 is confusing. If analyzing yearly values can not capture the ‘true’ relationship as the authors said. Why present it in results? To me it looks like you are analyzing the driving variables for different time scales, The 4.2 is for seasonal cycle and the annual analysis is for interannual variations.
Section 4.2 Why vapor pressure has the highest correlation for monthly results. This needs an explanation for the mechanisms of dominant control by humidity and vapor pressure. Are the mechanisms differ with wetlands and lakes?
Line 304 Correlation of 0.77 for short-wave radiation is still high.
Xiangyu Liu and Qianlai Zhuang
Xiangyu Liu and Qianlai Zhuang
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 228 | 84 | 9 | 321 | 2 | 1 |
- HTML: 228
- PDF: 84
- XML: 9
- Total: 321
- BibTeX: 2
- EndNote: 1
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1