| 15 Sep 2026
Satellite Constraints on European Landfill Methane Emissions Reveal Underestimation in Reported Inventories
Abstract. This study evaluates the consistency between landfill methane emissions reported in European Union bottom-up inventories—the European Pollutant Release and Transfer Register (E-PRTR) and the Industrial Emissions Directive (IED)—and independent satellite-derived estimates compiled by the United Nations Environment Programme’s International Methane Emissions Observatory (IMEO). Additional emission estimates from GHGSat observations are included to extend the satellite-based assessment. Our study reveals that satellite detections remain limited, with cloud cover representing a major observational constraint, and primarily capture a subset of the largest emitters, preventing comprehensive characterization of landfill methane emissions across Europe. Among the datasets considered, GHGSat provides comparatively better spatial coverage and improved capability for detecting moderate emitters. The comparison between European Union inventory data and satellite-derived observations reveals a systematic underestimation of landfill methane emissions in bottom-up inventories. Measurements from NASA's high-resolution Earth Surface Mineral Dust Source Investigation (EMIT) instrument indicate methane emission rates that exceed reported inventory values by up to one order of magnitude. Additionally, satellite observations reveal pronounced seasonal variability in landfill emissions that is not captured by annual inventory reporting. Although the available observations cannot establish whether this bias is representative of the entire European landfill sector, if confirmed across a larger and representative sample of facilities, such systematic underestimation could indicate that the European anthropogenic methane budget is significantly underestimated. These findings highlight the need to integrate facility-level reporting with satellite observations to better constrain landfill methane emissions, improve emission verification, and support the assessment of progress toward EU methane-mitigation and climate-neutrality targets.
General comments
The manuscript compares facility-level landfill methane emissions reported to the European E-PRTR/IED Industrial Emissions Portal with satellite-derived emission estimates. The satellite estimates are compiled by UNEP’s International Methane Emissions Observatory (IMEO), from TROPOMI (via CAMS MHE/SRON and Kayrros) and the imaging spectrometers EMIT, EnMAP and PRISMA, complemented by GHGSat observations. The main conclusion is that reported landfill emissions are substantially underestimated, by about an order of magnitude according to EMIT, and that landfill emissions vary strongly with season.
The topic is relevant and timely. Facility-level verification of EU landfill reporting is a real gap, and bringing E-PRTR/IED together with the IMEO catalogue and GHGSat data is a useful idea. However, the manuscript is still mostly a descriptive inventory of third-party detections. The manuscript would benefit substantially from (i) a clearer article structure with a dedicated Data and Methods section, and (ii) a clearer description of the datasets and their roles, in particular the distinction between satellite observations and the derived emission products compiled by IMEO. In addition, some methodological aspects need attention before the quantitative conclusions can be fully assessed. I hope the comments below are useful to the author in strengthening the study.
1. Manuscript organization
The overall structure still makes it difficult to follow how the study progresses. Section 2 (“Data, Methods and Analysis”) combines methodological descriptions, results and interpretation. Some examples:
This mixing hides the analytical workflow and makes the study harder to reproduce. Separate sections would improve readability considerably and let readers evaluate the study design and findings more easily. I suggest:
The Discussion (Sect. 3) and Conclusions (Sect. 4) also repeat material from Sect. 2 in several places. For example, Sect. 3.1 repeats Sect. 2.1, and the detection-limit and cloud-cover caveats appear in Sects. 2, 3 and 4. Restructuring would also make it possible to shorten these parts.
2. Description and harmonization of the satellite datasets
The manuscript needs a clearer description of the datasets and the data-processing workflow. Sects. 2.2, 2.2.1 and 2.2.1.1 overlap substantially and could be streamlined. For example, the integration of multiple satellite instruments within IMEO is described at L135–155, L157–169 and L170–181, and the TROPOMI share is given at L195–197 and again in the Conclusions (L661–662).
Harmonization. It remains unclear whether, and how, the different satellite products were harmonized before the analysis. The author lists the products included in IMEO, but more information is needed on spatial and temporal harmonization. Specifically:
Given these differences in spatial resolution, measuring frequency, detection threshold and retrieval methodology, a detailed explanation of how comparable quantities were obtained is needed to understand how meaningful comparisons were made. This applies in particular to Fig. 4, Fig. 7 and Fig. 8. If the emission estimates were used as reported by each provider, without harmonization, please state this explicitly and discuss what it implies for the comparisons.
3. Suitability of TROPOMI for landfill detection
The discussion of the TROPOMI-based detections raises several questions.
The detectability argument (L237–240). The author notes that reported landfill emissions are below the TROPOMI detection threshold, yet TROPOMI-based detections are reported and interpreted as “independent evidence that reported emissions may be systematically underestimated”. Given TROPOMI’s coarse spatial resolution compared with the small physical extent of most landfill facilities, this conclusion seems counterintuitive and needs stronger justification.
Lack of correlation (L535). More generally, the manuscript reports no statistically significant correlation between the TROPOMI-based estimates and facility-level reported emissions (L535–539), and Fig. 8a appears to confirm this. The Conclusions also note that TROPOMI’s resolution “precludes firm conclusions regarding the consistency between the two emission estimates at this scale” (L723–725), which I fully agree with. If the dataset does not adequately resolve facility-scale emissions, it is difficult to see why TROPOMI-based detections and their positive bias relative to the inventory should be taken as evidence that existing inventories or reporting frameworks underestimate emissions. I suggest either giving a stronger rationale for this interpretation or presenting the TROPOMI-based results mainly as screening information on where large emission events occur, rather than as quantitative evidence of under-reporting.
4. Use of high-resolution satellite products
Given the limitations of TROPOMI for facility-scale analysis, the author may wish to consider focusing the quantitative analysis, particularly the comparison with reported emissions, on the higher-resolution datasets. These are EMIT and EnMAP (DLR), together with PRISMA and GHGSat where sample size allows, and they appear better suited to the spatial scale of landfill emissions. At present, many conclusions rely on indirect inference because some lower-resolution datasets cannot resolve the sources under investigation.
TROPOMI-based estimates could still be kept illustrating screening capability and detection patterns. However, a clear justification of why each dataset is included, and of its intended role in the analysis (detection, quantification, or comparison with the inventory), would make the study more coherent and the conclusions more robust.
Technical corrections