the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Sep 2026
Contribution of livestock effluent management to the soil greenhouse gas emission in a Sahelian agropastoral region
Abstract. The assessment of greenhouse gas (GHG, i.e., CO2, CH4 and N2O) emissions from soils is a major component of the carbon footprint of the agro-pastoral ecosystems in the Sahelian region. For this purpose, CO2 emissions were measured in the district of Dantchandou in Niger using a portable analyzer EMG-5 over the annual cycle at nineteen sites from three land use types: shifting crop fields, manured crop fields, and fallow, as well as from three livestock concentration spots: livestock corrals, areas near pastoral ponds and near a pastoral well. Furthermore, CH4, and N2O emissions were measured using a static chamber on a subsample of three land uses and in five different months. The daily and seasonal emission dynamics are defined. The CO2 emissions measured with the EMG-5 are consistently lower at the beginning and end of the day, higher in the early afternoon in un-manured sites, and in the late morning in manured sites. Mean daily CO2 emissions increase with the first rains in June, peak at the beginning and end of the rainy season, and fall sharply following the end of the rains in October. The average daily CO2 emissions from livestock concentration spots reaches during the rainy season 2.64±1.35 gCO2 m-2 d-1 in vicinity of the pond, 24.99±15.73 gCO2 m-2 d-1 in corralling field and 75.09±28.34 gCO2 m-2 d-1 close to the pastoral well, compared to respectively 1.53±1.35, 1.68±0.65 and 4.68±0.75 gCO2 m-2 d-1 during the dry season. The rainy season averages from shifting, manured and fallow fields are 4.3±2.72, 13.8±5.76 and 4.7±2.62 gCO2 m-2 d-1 respectively and only 1.3±0.41, 1.8±0.48 and 1.2±0.35 gCO2 m-2 d-1 during the dry season. The annual mean CO2 emissions are very high in livestock corrals and near pastoral wells (respectively 34.0±10.9 and 92.5 MgCO2 ha-1 y-1) and higher in manured fields (21.2 ±3.9 MgCO2 ha-1 y-1) than in shifting fields, fallow and around ponds (8.3±2.5, 8.7±1.5 and 9.9±3.7 MgCO2 ha-1 y-1 respectively). To account for that spatial variability in assessing the annual emissions at the scale of the district, emissions were weighted by the areas of each land use or livestock concentration spots in the territory. As a result, annual emissions from soils in the district are estimated at 16.5 MgCO2 ha-1 y-1, 175 kg CH4 ha-1 y-1 and 36 kg N2O ha-1 y-1. The gap in CO2 emissions between manured and un-manured sites is used to estimate the contribution of the livestock effluent deposit to the annual CO2 emissions in the territory. In average over the year this contribution amounts 0.48 gCO2 m-2d-1 or 15.3% of total soil CO2 emissions.
- Preprint
(1343 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4903', Anonymous Referee #1, 21 Oct 2026
-
RC2: 'Comment on egusphere-2026-4903', Anonymous Referee #2, 23 Oct 2026
There is not much data supporting this manuscript. Five sampling dates for CH4 and N2O and 10 for CO2. Yes, they sampled multiple times on those dates, but that is very little data. And there are no covariables other than the type of landuse. No soil physical or chemical data for each site, no temperature / moisture data. Just emissions measurements. And I am very skeptical of both the N2O and CH4 results. The results they present as annual emissions are very high, particularly for a semi-arid climate and sandy soils. See below for more feedback and specific comments.
Abstract:
Line 11: GHG emissions are a major component. Remove “assessment of” as the assessment itself is not a component of the C footprint.
Lines 24 to 26: Given your lack of measurements, I’m not sure you should try to calculate annual emissions. It is fair to compare treatments, but taking only 10 measurements across the entire year is insufficient for the calculation of annual emissions.
Line 50: The Graham et al., 2022 paper mentions that enteric methane production from ruminants accounts for 67% of agricultural emissions in SSA. I guess that is more than 30%, but I think you should be more careful in citing this paper.
Introduction: the first paragraph (lines 36 to 54) is poorly written. It starts off talking about typical ruminant farming (i.e. pastoral systems), then switches to what limits yields in crop and rangelands and soil fertility issues, then switches again to present how livestock and soils are GHG sources, and how they are blamed for being sources in SSA and also globally before switching again to mention how GHG concentrations are increasing.
Line 57: is the soil moisture status “permanently low”? what happens after strong rainfalls? And the sentence also seems to imply that low pH supports low soil moisture status.
Line 65: “this surge results…”.
Line 71: while the statement about N2O emissions in the Sahel being linked to N inputs from livestock manure, it seems to come out of nowhere. There needs to be a better link between this statement and the points regarding nitrification and denitrification prior.
Line 78: what do “water surfaces” have to do with your manuscript? And how much livestock effluent is deposited on water? I would delete the “and water surfaces”.
Lines 86-88: This might be better in the “Acknowledgment section”. Not here. Possibly earlier in the introduction as well.
Lines 125-127: So, there is no replication for CH4 and N2O, and only five measurements throughout the year? Is that correct? Can’t do much with that little data.M&M
Did you leave a collar in the ground for the manually sampled chambers? Or did you just push the chamber into the ground at T0? How might pushing the chamber in just before sampling affect the flux measurements?
Did you leave plants inside the chambers? Were the chambers opaque or clear? Did you measure soil / air temperature? How about other variables? Did you evacuate the 10 ml vacuum flasks first? Did you use a single or double septum? Were the septa rubber or something else? Did you add dessicants? Please clarify your methods.
Did you measure soil properties at the measurement sites? Because there is likely a lot of variety in the soils across the region. These properties would also be very useful in determining why there may be variability in the emissions as well.
You don’t mention your statistical methods. Did you do any stats on the emissions? And whether the treatments differed?
Line 115: “sensors”, not “censors”.
Lines 114 to 117: The weather and incoming energy flux measured at the AMMA-CATCH site should be OK, but considering the variable landscape, I’m not sure if the hydrologic sensors and CO2 and water measurements are representative of all of your sites. It would have been really good to have soil temperature and moisture measurements at each site, and on each gas sampling date.
Lines 170 to 173: Did you just use the one standard for N2O? What was the highest concentration? Given that the ECD response is typically non-linear, how did you account for this non-linearity at higher concentrations?
Line 177 to 179: the formula uses “Fgaz” while the explanation below uses “Fgas”. Make these consistent please.
Line 180: why did you use a constant for atmospheric pressure? Wouldn’t it change, particularly between the rainy and dry seasons? I would expect lower pressure during the rainy season.
Lines 202-204: how did you measure the amount of livestock feces that was deposited? And did you do anything to measure the urine as well?
Lines 205 to 208: Five dates are insufficient for calculating annual emissions.Results:
The total N2O and CH4 emissions seem much higher than usual for SSA. I find these values suspect, particularly given the climate and the soil type (lots of sand). I would really like the authors to double check the numbers and the methodologies because these types of emissions are really high. Emissions of ~40 kg N2O-N per ha per year would be very high in intensively managed croplands with high N inputs, so I am very skeptical that you would see this in this type of system.
The units in the results section (and possibly other areas as well) need to be written correctly. The use (or lack of use) of spaces makes it a bit awkward to read. I would have a space before and after the “±”. Also, make sure it is “g CO2 m-2 h-1”
Also, the results need to be written in the past tense. E.g. line 242: “increased” rather than “increase” and line 244 “”reached” rather than “reach”.
Within figures, please keep all Y-axes the same. This makes comparisons easier.
Lines 231-234: did you run any statistics on this? Or is this only talking about trends? Given the lack of replicates, you need to really be careful.
Line 238: “emissions remained low”.
Line 248: Kruskal Wallis test needs to be described in the M&M section.
Figure 4: the x-axis for the Bare Soil is in French still.
Lines 269-270: why did you do this? And did you check for any CH4? It would make more sense to me to look at CH4 from these wet areas.
Line 275: how did you measure the DM of feces deposited on the soil? This was not in the M&M. And also, did you measure the emissions at the same spot of the corral where you measured DM? These may not be deposited evenly across the corral. And why did you not mention the regression in the M&M?
Line 278: You don’t really have enough to determine annual averages. Just say “average”.
Line 281-283: this sentence is very poorly worded. Please revise. Also, didn’t you only measure N2O at one of each site? If you have no replication then you can’t really state that one treatment is “significantly higher” than another treatment.
Table 2: Not really an “annual average” as you haven’t collected sufficient numbers of samples. Just say “Mean daily emissions” or something like that. And, is the “n” value the same for the CO2 and the CH4 / N2O? Also, typically these sorts of results are presented as CO2-C or N2O-N (i.e. the Carbon or nitrogen component of the molecule). Also, the units for N2O are incorrect. They should be mg m-2 d-1. Not grams.
As a result of this, I am also a bit concerned about the scaling up to the whole district. Particularly as Rangeland (which is 22% of the area) was not measured and Fallow (another large percentage of the area) only was measured for CO2. These comments are also for Table 3.
Line 302: 24.4, not 24,4.
Table 3: what is happening with the total CH4 emissions. The CO2 and N2O are sums of the rows above, but that doesn’t work for the CH4. Where do you get 9.90 from? Shouldn’t it be 175.6 (like what is mentioned on line 300)? In general though, the CH4 and N2O emissions seem really high for a semi arid climate with sandy soils and low carbon.
Discussion:
Lines 307 to 314: this section is just presenting results again. Any insights as to why there was no variability during the dry season? And why there was greater variability during the rainy season? Did you measure soil moisture and temperature at each site? Do Ferreira et al discuss why they expected differences and why they did not see any?
Lines 325 to 327: Can you back this statement up? Do you have data on the quantities and nature of the OM, litter and fecal material? If you do, please share that data. You only present the fact that some fields had manure applied. Not how much or what the results of a chemical analysis or anything like that. You also don’t present any information on the amount or type of OM or plant litter. So, it really doesn’t look like you have the supporting data for this statement.
Lines 365-367: these are results, and should not be in the discussion section.
Lines 366 to 370: You state that the N2O emissions were low, but your values in Table 3 are very high. And your values are actually very high compared to the emissions in the Butterbach-Bahl et al., 2020 paper. The emissions in that study ranged from 9 to 10,266 ug N2O-N m-2 hr-1 in the bomas (mean = 643) and from -0.7 to 15.9 ug N2O-N m-2 hr-1 (mean = 5.5) for the savannah sites (fallow). The 5.5 ug N2O-N m-2 hr-1 would equal about 0.5 kg N2O-N ha-1 yr-1, which is about 1% of what the current authors assumed for N2O emissions from fallow and rangeland. I would assume the fallow from the current study and the savannah sites in the Butterbach-Bahl study would be similar, so, the assumption by the current authors to assign 40 kg N2O /ha /yr seems way too high.
Lines 379 to 409: It would be great to actually look at some of the spatial variability, but you don’t’ have any data to understand why that variability occurs. You don’t know the amount of C and N added with the manure application, you don’t have any information on variability in soil C across sites, or soil N. Just whether or not manure was added. Which does not allow for much investigation.
Lines 446: According to the Li and Kelliher paper, CH4 oxidizing bacteria worked best in acidic conditions, with the oxidation potential decreasing by 50% when the pH increased from 4.4 to 6.4. The paper by Su et al. also mentions that it is important to consider the initial soil pH because decreases in soil pH can either increase or decrease the amount of CH4 oxidation. So you may want to change this part of the discussion. Low acidity also tends to reduce methanogens as well.
Line 476: please don’t start a sentence with a number.
Conclusion: it is too long and repeats too much of the discussion. The conclusion needs to be shorter and with more of a synthesis than a repetition of what was said earlier.Citation: https://doi.org/10.5194/egusphere-2026-4903-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 385 | 118 | 66 | 569 | 75 | 68 |
- HTML: 385
- PDF: 118
- XML: 66
- Total: 569
- BibTeX: 75
- EndNote: 68
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study evaluated the GHG emissions from different land use type and livestock concentrated area in a Sahelian agropastoral region. Large uncertainties in GHG estimation still exist in SSA due to the limited number of studies. However, there are several concerns on this study from my sides:
(1) Though, it was clearly that field campaign was difficult in this area. Using a frequence of 8 times per year for CH4 and N2O to do the annual estimation certainly would lead to large uncertainties or even misleading conclusion. Such as line 278: daily CH4 emissions are four times lower in livestock corrals than in shifting crop field?
(2) In general, the soil acts as CH4 sink. Manure application may results in some short CH4 emissions, but it is still hard to turn the soil from CH4 sink into CH4 source considering the amount of manure applied and the short-term manure induced CH4 emissions. In your study, both fields (manured and shifting fields) acted as very large CH4 emissions. It would be great that you double check your data, especially the unit. The same for the N2O in figure 3, from -250 µg N2O m-2 h-1 to 1000 µg N2O m-2 h-1. This does not make sense for soils, soils in such region could not act as such large N2O sink. Regarding the soil fertility and low N input, it would be very hard to find such a big peak. Kindly double check your units.
(3) Though studies in SSA are scarce, numerous studies have reported the diel patterns of the GHG fluxes. The daily dynamics in this study did not provide new information. I suggested the authors to compare difference across land use type and seasonal effect using those data directly, not the cumulative emissions.
Specific comments:
Abstract:
CH4 and N2O data should be mentioned.
Regarding the measurement frequency, it is not recommended to do the comparison using the mean value of six time in rainy season and two times in dry season.
Introduction:
Line 55-72 Here, the livestock effluent on soil GHG emissions should be stated in details.
M&M:
2.2 To make you study site and measurement more easily understood, in table 1, add the measurement information.
Figure 1 consider added the landscape photo for each land use type.
Results:
Using the mean daily dynamics to do the comparison across land use types and seasonal effects.
Fig. 3 double check the unit and improve the quality of the figures
3.4 any other data you can use to compare for the upscaling emissions?
Discussion
Not necessary to discuss the daily dynamics and focus on the seasonal variability and land use type.
Conclusion
This is too long for a conclusion, make it shorter.