Peatland Emission Reduction Potential from partial/seasonal re-wetting and other management measures

Peatland Emission Reduction Potential from partial/seasonal re-wetting, and other management measures Rebekka Artz, Stuart Smith, Doug Wardell-Johnson, Mohamed Jabloun, Mike Rivington, and Keith Matthews James Hutton Institute Report produced under a Calldown request. September, 2023; Updated to reflect more up to date research in July 2025 Disclaimer The findings and recommendations in this report are the result of a research study, and do not reflect the views or position of Scottish Ministers. Executive Summary Around 75% of Scotland’s peatlands are currently estimated to be degraded and contributing 6.34 Mt of carbon dioxide equivalents to our territorial greenhouse gas emissions. A very significant proportion of these emissions are estimated to originate from peat under agricultural use, ranging from rough (wild) grazing for deer, sheep and dairy cattle grazing, and a small area under cropland use. The current Inventory data are particularly sensitive to uncertainties in the estimated areas and the emission factors of peat, particularly for the more intensive (high emitting) land use categories of extensive and intensive grassland. Further research in this project and a recently published report (Parker et al., 2025) on the area estimates of the ‘on paper’ higher emitting land uses (cropland, intensive and extensive grassland) on peat suggested that the true areas involved may be lower than currently estimated. Further evaluation of the emission factors of intensive and extensive grassland on peat suggests that emissions in Scottish settings could be lower than currently estimated, due to the generally lower stocking densities/offtake and potentially also due to lower water table depths in the generally wetter and cooler climatic settings. Emissions of modified bog (also part of the LULUCF Grassland reporting class), however, could be higher than currently reported due to an omission of a biomass offtake correction in the current Inventory calculations. All UK land is considered managed land, hence ‘wild’ deer grazing is still considered management and even near natural bog is currently included. While direct partial rewetting measures may not be applicable to Scotland, realistic (rather than ‘on paper’) emissions reductions may well be possible with livestock/deer management on peat. Scenarios have been run, which suggest some, highly uncertain, benefits of managing densities of grazers at minimum agricultural activity levels (0.05 LU ha-1). Some very small co-benefits of reducing cattle on peat could filter through to the Agriculture UKGHGI sector, as a small proportion of biomass C being converted to methane is included in that sector. How this is accounted for (possibly via Tier 1 total livestock numbers, not split by soil type) would require further research. The UK Inventory activity data and emission factors thus will require updating for Scottish scenarios to better reflect ‘real life’ baseline emissions and mitigation potential. Work is ongoing in UNC-F3-5 to improve the emission factor component. The findings of a recent RRF project (Parker et al., 2025) suggested that the true area of grassland on peat may be in the region of 3000-20,000 ha instead of the current 100kha estimate in the UK Inventory. Both these projects should result evidence that could feed into the Inventory in the next few years. An incidental finding of this project suggested that there might be additional full rewetting through AECS measures that may not yet be accounted for in annual Inventory submissions. These estimated ~3kha will require further evaluation, and comparison with Peatland ACTION submissions to ensure avoidance of potential double counting. Background Around 75% of Scotland’s approximately 1.9 million hectares of peatlands are estimated to be in a degraded condition 1 . While peatlands in good condition are small net carbon sinks that accumulate peat slowly over millennia, degradation turns such carbon sinks into considerable net carbon sources that contribute to climate change. Current estimates of the contribution of damaged peatlands to climate change in Scotland are that they release 6.34 Mt carbon dioxide equivalent (2021 UK GHG Inventory figures 2 ) out of a total of 41.6 Mt territorial greenhouse gas emissions, enough to nearly offset the entire removals provided by forestry 3 . In addition, the changes in our climate that are already being experienced may lead to reductions in groundwater levels across large scales, threatening to further increase emissions from degraded peatlands as well as impacting the sink strength of remaining peatlands in good condition 4 . Scottish Government committed to restoring 250,000 hectares of degraded peatland by 2030, in its Climate Change Plan update and current Programme for Government, and committed to a £250 million ten-year funding package in 2020 to achieve this target. This funding package was developed on the basis of estimated average cost of capital investment to complete initial rewetting (i.e., excluding ongoing management costs) at that time. Capital costs have, however, risen rapidly since then due to inflationary pressures. It is now evident that either additional investment (from either public, private, or blended sources) is required, or alternatives to full rewetting are required. To date, most of the focus in peatland restoration has been on achieving full rewetting of land that has clearly observable restorable features that are contributing to drainage. Such areas may have drainage ditches (grips), erosion gullies, peat banks that have been the result of domestic peat extraction, or conifer plantations, inclusive of the planting mounds and furrows. Much less attention has been given to land under more productive agricultural use, for example where former bog or fen habitat has been drained and converted to extensive or intensive grassland, or, in extremis, cropland. Although such areas are relatively less common in Scotland, and tend to be confined to margins of former lowland raised bogs and fens or at the margins of blanket bogs in uplands, current estimates of their emissions are that they are quite significant carbon sources per hectare (Table 1). The 2021 UK Greenhouse Gas Inventory submission estimates that a significant proportion of the 6.34 Mt of estimated CO2 equivalents originate from peat currently under grassland 5 or cropland uses. This has led to policy teams seeking to understand whether there are options to reduce emissions from high-emitting peatland areas without requiring full rewetting efforts. The aim of this rapid review was, therefor, to assess the potential to partially rewet peatland in Scotland, in order to provide further evidence that could help to support development of policy measures and targets for peatlands within the Agriculture Reform Route Map and the next Climate Change Plan Update. 1 https://uk-air.defra.gov.uk/assets/documents/reports/cat07/1904111135_UK_peatland_GHG_emissions.pdf 2 Direct communication from Dr Hannah Clilverd, UKCEH, July 2023. 3 https://www.gov.scot/publications/scottish-greenhouse-gas-statistics-2021/ 4 https://www.nature.com/articles/s41558-021-01059-w 5 The UKGHGI includes the following land uses on peatland in the grassland category: Eroding Modified Bog, Modified Bog, Extensive Grassland, Intensive Grassland, Rewetted Modified Bog (from Modified Bog), Rewetted Bog or Fen (from Eroding Modified Bog, Intensive and Extensive Grassland. Table 1. Estimated areas of peat soil in Scotland under each condition category (2021 UK GHGI activity data, provided by Dr Hannah Clilverd, UKCEH, July 2023) compared with the baseline in 1990 (Inventory data for 1990 were not available, showing data from Evans et al, 2017, which may differ slightly) Condition class 1990 baseline area (ha, from Evans et al, 2017) 2021 area (ha) Tier 2 EF (2021) in t CO2e ha-1 y-1 Forest 332,746 359,612 N/A (Tier 3 reporting) Cropland (peat > 40 cm) 8,181 5,148 37.17 Intensive Grassland 78,641 70,605 22 Extensive Grass 31,794 32,336 15.88 Industrial Extraction 2,881 2,434 18.86 Domestic Extraction 44,923 44,655 15.18 Eroding Modified Bog Drained 74,147 8,333 18.86 Eroding Modified Bog Undrained 198,116 28,647 17.72 Modified Bog Drained 188,326 213,943 3.32 Modified Bog Undrained 496,498 641,243 2.51 Rewetted Bog 0 16,740 3.42 Rewetted Fen 0 2,044 3.31 Rewetted Modified Bog 0 29,287 0.32 Settlement Not reported 6,483 N/A Near Natural Bog 490,497 490,216 0.32 Near Natural Fen 0 0 -0.36 Total 1,947,750 1,951,727 What is partial/seasonal rewetting? Partial or seasonal rewetting are terms used for any measure that seeks to raise the water level in a peatland to a lesser level, or for a shorter duration each year, than rewetting normally aims for. Where full rewetting aims to create a water table that is near the surface all year round across all of a site, partial rewetting measures include a) reducing the drained area of an agriculturally productive field, or b) moderating drainage conditions by either fully rewetting during only the winter months or by moderating the drainage depth towards shallower water tables across the site. Freeman et al (2022 6 ) and others describe various partial (in area terms) or seasonal (summer drainage, winter rewetting) measures that could theoretically be applied to Scottish degraded peatlands. Any measure that produces a shallower mean annual water table should result in a reduction in emissions (Evans et al., 2021 7 ) as there is a strong relationship between the mean annual effective water table and gaseous emissions of carbon dioxide and methane. Other researchers have shown very similar relationships, although the fit of the models to the data varies slightly (see e.g., Freeman et al., 2022idem). 6 https://onlinelibrary.wiley.com/doi/epdf/10.1111/gcb.16152 7 https://www.nature.com/articles/s41586-021-03523-1 In Scotland, the vast majority of peatland drainage has occurred with open drains and so the potential to partially rewet such areas is limited to reducing the depth of drainage with e.g., partial infilling of the drains or dams that are leaky through their surface layers. In practice, the costs of these solutions are probably similar to full rewetting. In agriculturally more productive drained peatlands, a solution currently trialled in e.g., the Netherlands and southern England is the use of submerged drainage piping, installed at a fixed height within the peat. This allows for active water level management, where in winter excess (flooding) water is pumped from the drains to rivers, and in summer the site is partially rewetted to the level of the drain by allowing flow from the river to the drains. Such systems have not been tested in Scotland, nor are there likely to be significant enough areas of land available for this solution to produce meaningful emissions reductions at a reasonable cost. The installation itself would incur a carbon cost, as would ongoing active pumping. At present, there is limited evidence of the impacts on the carbon emissions that such partial rewetting measures have, with one study that showed no impact on carbon emissions in sites where the water table was extremely deep and the peat layer quite limited (i.e., the partial rewetting did not protect the peat layer) 8 and two studies that showed reduced net carbon dioxide losses 9 , 10 although in one of these latter studies this was offset by extremely high nitrous oxide emissions. The mean effective water table is the water table within the peat soil layers of a soils, meaning that this relationship can also be applied to so-called ‘wasted’ peatlands 11 , where the former deep peat has been removed or oxidised to leave a depth of less than the 50 cm per national definition. Unfortunately, such ‘wasted’ peat has not been mapped adequately in Scotland 12 , however there are likely to be substantive areas of such wasted peat around former raised bog margins and which are often under cattle grazings. In the ongoing RESAS funded UNC grassland on peat project, the contemporary extent of peat as estimated by the current UKGHGI at around 200 locations was compared with historical 6-inch mapping provided by the Ordnance Survey in the 1860s 13 . In these maps, then still uncultivated bogs, fens, marshes and other rough grassland and moorland were mapped at astounding detail. A recent effort by Macfarlane and colleagues at JHI in collaboration with the National Library of Scotland used machine learning tools to produce a mapped output from the digitised historic maps 14 . Further site-based checks of a small number of these locations within the UNC grassland on peat project suggests that peat can indeed still be found in some such areas, and therefore future mapping of peat will also require more precise evaluation of the edges of peat to locate wasted peat that is not currently included in the UKGHGI. An ongoing RESAS-funded Strategic Research Project has been developing a new 10 metre resolution digital soil map for Scotland data product that aimed to improve on the UKGHGI accuracy 15 . The quality of the peat predictions, however, was only moderate when compared with the Peatland ACTION peat depth database, and therefore this data product cannot be recommended for alternative use. 8 https://bg.copernicus.org/articles/18/3881/2021/ 9 https://bg.copernicus.org/articles/19/5707/2022/ 10 https://www.sciencedirect.com/science/article/pii/S0168192323000035?via%3Dihub 11 https://digital.nls.uk/pubs/e-monographs/2014/701.pdf 12 https://www.iucn-uk-peatlandprogramme.org/sites/default/files/202306/Use%20of%20Peat%20Depth%20Criteria%20-%20Accounting%20for%20the%20Lost%20Peatlands_1.pdf 13 https://maps.nls.uk/os/6inch/os_info3.html 14 Moorland in Scotland, 1840s-1880s - Georeferenced Map Overlay viewer - Fraser Macfarlane - National Library of Scotland 15 Soil Property, Carbon Stock and Peat Extent Mapping at 10 m Resolution in Scotland Using Digital Soil Mapping Techniques - Robb - 2025 - European Journal of Soil Science - Wiley Online Library Future fates of peatland carbon stores Global-scale model projections currently predict that even peatlands currently still in a near-natural state will become net carbon sources under the high-end (RCP8.5) climate projections, whereas they remained as sinks under the strong mitigation scenario (RCP2.6) 16 . The RCP2.6 scenario has also been tested with and without peatland protection and restoration policies in a global scenario, which additionally suggest that the rewetting of 60% of degraded peatlands globally, over the coming decades, could help turn the entire land system back into a net carbon sink by 2100 17 . An additional consideration is that wildfire incidence is increasing with climate change, and observations suggest that wildfires reduce the strength of remaining intact peatland to act as carbon sinks and enhance emissions from degraded peatlands 18 . Furthermore, shallower or more fragmented peatlands are more vulnerable to high severity fires and larges losses of peat carbon stocks 19 , 20 . For near natural peatlands, warmer spring temperatures and water table drawdown in late summer in particular, appear to be driving changes in emissions 21 . We therefore also identified peatland areas in Scotland that may be under likely increased climatic stress, on the basis of climatic water balances (Fig 1). Fig. 1. Change direction in mean monthly water balance over the period 1990-2019, compared with a 1960-1989 reference period, for peatland areas as per the UKGHGI model extent. Areas in light blue remained in a surplus water balance, whereas areas in dark blue went from a deficit to a surplus. Areas in red were either remaining in deficit (light red) or changed from a surplus water balance to a deficit (dark red). 16 https://www.cell.com/one-earth/pdf/S2590-3322(21)00726-0.pdf 17 https://iopscience.iop.org/article/10.1088/1748-9326/abae2a 18 https://www.nature.com/articles/s41558-023-01657-w 19 https://iopscience.iop.org/article/10.1088/1748-9326/aba7e8 20 https://www.sciencedirect.com/science/article/pii/S0048969721002783 21 https://www.nature.com/articles/s41558-022-01428-z Peatlands in the eastern parts of Scotland as well as around the lower lying areas of the Northers and Western regions are already increasingly affected by climatic water deficits. We also checked the projected climatic water balance, using three ensemble members of UKCP18 climate projections. These are based on the high emissions (RCP8.5) scenario and therefore represent a more pessimistic scenario than global climate change is currently tracking, however it is useful to see a range of plausible worst-case outcomes. For this report, we selected three model ensemble members that represented a mid-range of plausible future climates under this scenario (see Annex 1). Projected water balances suggest, in all cases (Fig 2) that pressures on peatlands in all but mountain regions will increase during the next few decades. Climatic water balances are calculated using the differential between precipitation and reference evapotranspiration (ETo, calculated using the Priestley-Taylor equation with a grass reference). Actual evapotranspiration may be lower in peatlands than reference evapotranspiration, however these mapped areas of potential deficits do suggest that there may be scope to look at regional prioritisation of partial or full rewetting efforts in areas projected to go into climatic water deficits. In particular, it may be useful to focus efforts on areas that are either currently high emitting (see next section) and/or have high densities of carbon locked in the top 1 m of their soil columns (Annex 2). Ensemble member 04 Ensemble member 05 Ensemble member 08 Fig. 2. Change direction in mean monthly water balance over the period 2020-2049, compared with a 1960-1989 reference period, for peatland areas as per the UKGHGI model extent. Areas in light blue are projected to remain in a surplus water balance, whereas areas in dark blue are projected to change from a deficit to a surplus. Areas in red were either projected to remain in deficit (light red) or change from a surplus water balance to a deficit (dark red). Plots are shown by month for ensemble member 4 (top; >2°C warming, very slight reduction in precipitation), ensemble member 5 (middle; >2°C warming; 2.5% reduction in precipitation) and ensemble member 8 (lower panel; ~1.9°C warming and 8% increase in precipitation). How much degraded peatland in Scotland could partial/seasonal rewetting theoretically be applied to? Current, full, rewetting efforts are focusing on the vast area of peatland that has identifiable drainage features such as those that have undergone drainage, are eroded, or have been afforested or cutover. This leaves the still very significant areas of croplands and grasslands (estimated at 108kha, Table 1 and Fig 3) as well as the modified undrained peatlands (641 kha) currently used as low intensity grazing land for sheep and/or sporting areas for e.g., deer. In addition, there may be a significant area of wasted peat in Scotland that no longer meets the 50 cm threshold, but may be emitting at quite high rates if the remaining peat layer is above the mean water level. At the moment, some wasted peat areas are mapped within the Scottish activity data for the UKGHGI but little ground verification of the validity of these areas has been undertaken and so it is likely that other areas of wasted peat are currently missed by the modelling approach to defining peat extent. In Inventory terms, such areas of grassland that are not included in the organic soil Inventory are currently assumed to be undisturbed and hence is assumed not to carry any emissions (Section 6.4.1. of the NIR, Pers Comm Dr Hannah Clilverd). Hence, any significant under-mapped wasted peat should be addressed in future revisions of the Inventory activity data. Leaving aside the considerable uncertainties over available area, the most likely management options would be to incentivise stock densities for agriculturally used peatlands to the range of the minimum carrying capacity (see also below) and instigate partial rewetting of fields that have a peat component, for example, by re-routing or sluicing of existing perimeter drainage. It is unknown whether lowland peat soils under agriculture have been routinely drained with subsurface drainage pipes in Scotland 22 , 23 . Where these exist, however, an argument could be made to remove or modify these structures so that they act as a means to partially rewet the area (see previous sections). Fig 3. Spatial distribution of the cropland (left) and extensive/intensive grassland (right) on peat as per the modelled peat extent in the UKGHGI. Individual polygons have been exaggerated with a very wide border for visibility of their relative locations across the country, individual fields in these categories tend to be quite small (see Table 1 for total area figures). 22 https://www.rspb.org.uk/globalassets/downloads/documents/farmingadvice/lowlandagriculturallanddrainagesystemsfunctionandmondificationforwetlandconservation_254843.pdf 23 https://www.climatexchange.org.uk/media/1636/report_on_agricultural_drainage_and_greenhouse_gas_abatemen t_in_scotland.pdf The above limits are broadly compatible with the current minimum agricultural activity rules, including the cited minimum stocking density of 0.05 livestock units per hectare (but which can be lower if it can be demonstrated that the carrying capacity is lower or if an environmental management agreement is in place 46 ). As stated in the SRUC guidance, stocking densities should take into account what the carrying capacity of the area in question is within any given season/year. In a summer drought year, or a cold/wet summer year, for example, primary productivity may be lower than average and hence stocking densities would require to be reduced to stay below the overall productivity level of the system in order to avoid net emissions. Using a combination of the carbon offtake rates per animal, summarised above, and the threshold value of 0.05 LU for minimum agricultural activity, this would translate into biomass offtake rates of 0.1 t CO2e ha-1 y-1 for minimum agricultural activity cattle grazings; a range of 0.5-1.9 t CO2e ha-1 y-1 for minimal sheep grazings (lower value as per Worrall et al model, upper value as per mean of Table 2), and 1.8 t CO2e ha-1 y-1 for grazing by deer at the minimum agricultural activity. These values for deer, however, are double the density where damage may occur to open habitats such as near natural blanket bog, natural moorland on shallower organic soils or peatland moderately degraded to heather moorland 47 . At such halved deer densities, biomass carbon offtake would be in the range of 0.9 t CO2e ha-1 y-1. Behaviour of unfenced grazing communities, however, can be extremely hard to quantify and predict within a changing landscape, and rewetting projects in particular can produce a temporary flush of lusher vegetation that are very attractive to grazing and browsing animals. In generic terms, therefore, effective partial rewetting measures may also require low herbivore stocking densities to minimise biomass offtake and reduce the export of C component, sward management to reduce evapotranspiration losses while maximising GPP, and/or ditch management to improve water holding capacity on the peatland components of businesses. A word of caution, however, will need to be said for the nuisance of lower stocking densities for different peatland plant communities, particularly in areas where Molinia is expanding. Molinia caerulea can rapidly form near monocultures when grazing is reduced or removed, due to its tussocky growth form, and may require, for example, short grazing periods during the summer to re-open the sward in order to achieve a more diverse vegetation layer (Grant et al. 1996 48 ). Out of all vegetation scenarios explored by Worrall & Clay (2012) 49 in their GHG modelling, Molinia dominated communities had the highest carrying capacity to sustain grazing. Molinia dominance also presents an increased fire risk, particularly in spring when there is a large potential fuel load available due to the standing dead biomass from the previous year. The potential reasons for its expansion are manyfold, but two major factors are increases in nitrogen deposition post-industrial revolution, possibly then coupled with the nutrient enrichment that the initial grazing provided, feedbacks of reduced site wetness as caused by the high grazing pressure and/or ongoing changes in climatic conditions 50 . Use of livestock in 46 https://www.ruralpayments.org/publicsite/futures/topics/all-schemes/basic-payment-scheme/basicpayment-scheme-full-guidance/eligible-hectares-and-minimum-activity---bps/#612981 47 https://www.scotlink.org/wp-content/uploads/2020/01/Herbivore-Impacts-Upland-Red-Deer-DensitiesCarbon-Sequestration-and-Storage-in-the-Upland-Red-Deer-Range-%E2%80%93-a-Report-for-ScottishEnvironment-Link%E2%80%99s-Deer-Task-Force-.pdf 48 https://www.jstor.org/stable/2404769 49 https://www.sciencedirect.com/science/article/pii/S0048969712011783?via%3Dihub 50 https://besjournals.onlinelibrary.wiley.com/doi/10.1002/2688-8319.12113 conservation grazing, however, also needs to be balanced against animal welfare needs 51 . Other options that include cutting instead of grazing could also be considered 52 , although these are labour and carbon-intense alternatives. Analysis of potential partial rewetting scenarios Based on the collated findings of this calldown project of a) likely overestimates of the area of cropland and grassland on peat (though potentially counter-balanced by as yet unknown areas of wasted peat), b) the possibility to rewet peatland edges currently under grassland use, and c) the possibility to reduce emissions further through reductions in the biomass offtake by grazing animals, we explored a series of hypothetical ‘what-if’ scenarios. The first of these scenarios reduced the currently estimated area of cropland by half, and the currently estimated area of intensive grassland and extensive grassland on peat by 80% (i.e. set to 20,000 ha as per the upper estimate in Parker et al., 2025idem). Given Parker et al.’s findings that intensive grassland on peat was a relatively rarer occurrence than extensive grassland, we partitioned the 20,000ha into 4,000 ha intensive and 16,000ha extensive grassland on peat. We reallocated the reductions in cropland and grassland peat area to modified bog in order not to compromise the overall peatland area estimate in the Inventory (Table 4). This resulted in an estimate of total current emission from peat of around 5 Mt of CO2e. In the second scenario, we added the assumption that intensive grassland use on peat does not realistically exist in Scotland on meaningful scales, on the basis of our analysis of the activity data themselves and further supported by the IACS claims on the areas suggested to be intensive grassland in the UKGHGI. The area assumed to be under intensive grassland use was thus reallocated to extensive grassland, which carries a lower emission factor. Our second scenario reduced the estimate of current emissions by a further 0.1 Mt CO2e and should therefore also be viewed as within the plausible range of current emissions. In Scenario 3, we explored the potential impacts of reducing stocking densities on the remaining extensive grassland and modified bog to levels at the minimum agricultural activity level of 0.05 LU. A very large assumption in this scenario is that it is possible to reduce the biomass offtake on particularly upland peat areas to such a degree without adverse effects. In this scenario, we also included the possibility of the land area of peat under grassland being overestimated. This scenario has the ‘on-paper’ potential to lead to a reduction of 1.8 Mt CO2e compared with Scenario 2, but would require further research to check whether such a reduction is possible in the light of food security, biodiversity and Just Transition concerns. Scenario 4 then additionally takes all remaining extensive grassland out of productive use and assigns it to rewetting, while also converting any remaining cropland to wet agriculture (paludiculture). Due to the associated transition emissions of preparing land for a different land use, we assigned the (lower) emission factor for cropland on wasted peat for this last scenario, rather than assuming that such transition would result in emissions savings similar to full rewetting. There is very limited published literature to date on emissions from paludiculture projects. A recent Sphagnum 51 https://www.sciencedirect.com/science/article/pii/S0167880911002830?via%3Dihub 52 https://www.sciencedirect.com/science/article/pii/S000632070300483X?via%3Dihub paludiculture study from Northern Germany suggested CO2 emissions in the range of 4-10 t CO2 ha-1 y-1, but the study did not quantify site preparation of management emissions 53 . This fourth scenario could have the potential to reduce emissions by a further 0.3 Mt CO2e. Finally, we also show the option of fully rewetting all of the currently mapped degraded peatland, assuming the lower estimates for grassland on peat of 20kha. Some residual emissions will remain due to rewetted bog carrying some emissions for probably at least a decade (see also RESAS JHI-D32, CentrePeat, preliminary findings of a period of ca 10-20 years for transition to net zero emissions in rewetted peatlands). In summary, it is likely that a) the current estimates are significantly higher than the real-world scenario and b) notwithstanding these potential errors in estimates, some emissions savings might be feasible by reducing livestock or wild grazing levels on peatlands, or by rewetting areas currently estimated to carry high per hectare emissions. As described in more detail above, the assumptions for these Scenarios require more thorough checks of the actual existence of grassland extent and its condition on peat, as also summarised above. Stocking densities or wild grazer use rates on peat are highly uncertain and hence the analysis in Table 4 should, in general, be considered no more than an initial exploration of options. It is unclear how critical any current stocking levels on peat are to the overall business holdings in terms of their viability, or indeed whether any such areas they already contribute to conservation efforts via management agreements. Table 4. Exploration of hypothetical partial rewetting scenarios, versus current baseline and full rewetting counterfactuals. See text for full description of the scenarios. Emissions cannot reach net zero due to rewetted bog carrying a higher emission factor than near natural peatlands and it is currently uncertain whether the natural end point can be reached within a meaningful timeframe. Scenario Cropland Intensive Grassland Extensive Grassland Total estimated Scottish peatland emissions (Mt CO2e yr-1) Current UKGHGI 5,148 70,605 32,336 6.34 Scenario 1 – incorporating magnitude of estimated current mapping error 2,578 4,000 16,017 ~5.0 Scenario 2 – as per 1 but assuming IG non-existent 2,578 0 20,017 ~4.9 Scenario 3 – as per 2 plus reduction in grazing offtake on MB and EG to minimal agricultural activity levels 2,578 0 20,017 ~3.1 Scenario 4 – as per 3 plus d converting CL to paludiculture (indicated by brackets) and rewetting any residual EG areas (2,578) 0 0 ~3.0 Full rewetting of all cropland, peat extraction and grassland areas (using Parker et al., 2025 maximum) 0 0 0 ~1.2 53 https://www.sciencedirect.com/science/article/pii/S0048969723015590?via%3Dihub Current land cover uncertainties The above sections detailed the considerable uncertainties in the UKGHGI condition classifications, which were derived largely from Land Cover of Scotland 1988 data. Annual updates to this figure are currently provided by manual subtraction of rewetted areas as provided by NatureScot Peatland ACTION and/or through deforestation, and through correcting for any new forestry planting on seminatural land. This means that the current Inventory method does not apply a correction for any deintensification of land use that may have occurred since 1990 on categories such as cropland or intensive/extensive grassland. We can see this as the area of intensive and extensive grassland in particular does not change much between 1990 and 2021 in Table 1 (we cannot check whether the reduction in cropland may be due to conversion to e.g., forestry). We therefore overlaid the UKGHGI baseline data with the 2015 and 2019 IACS claims database, to see how well known current IACS claim land use attributes match the condition category in the UKGHGI. A each IACS claim must be associated with 1 field or 1 common grazing, so as long as the underlying UKGHGI model assumption of an area being peat is correct, IACS land use information is supplementary information on current condition. There is still some level of caveat though, as there can be fields with multiple claims and therefore multiple land uses. For example, a field with Spring Barley and Field Margins will be claimed as mostly spring barely, with some permanent grassland for the boundaries, but will show up as just spring barley for this analysis. For this analysis, the land use with the largest area per field was used to denote the whole field, so in the worst case the UKGHGI portion may be located on the bit of the field where the “other” land use is. This is more of a problem in the uplands where the fields are larger, so more likely to overlap only a part of the field. Fig 4. Condition category matches of 2019 (upper panel) and 2015 (lower panel) IACS land use claims overlaid with the UKGHGI baseline, shown as a proportion of the total area per class, for cropland, intensive grassland and extensive grassland. The areas in ha for each matching category are given in the figures per ‘pie slice’. Colour scheme is as follows: IACS land use code = cropland (yellow); intensive grassland (bright green), extensive grassland (light green), grey shades (unknown IACS assignment or no cover of IACS data), woodland (brown, virtually invisible slices in the figures). Taking these caveats into account, the resulting apportions to UKGHGI categories are shown in Figure 4. On balance, it is highly likely that some de-intensification has taken place since 1990, and that current land cover of cropland and intensive grassland on peat, at least, are likely to be lower, and extensive grassland on peat higher, in area than at the 1990 baseline. We cannot state this with a high degree of confidence due to the caveats mentioned above and due to the approximately quarter of UKGHGI locations that do not have any IACS claim information to match with. Existing measures for peatland management, excepting full rewetting via e.g., Peatland ACTION/Peatland Code There are a few existing agri-environmental measures already in existence for land where full rewetting via the Peatland ACTION programme and/or the Peatland Code are not currently intended, but which include measures to manage peatlands. Peatlands do not currently feature specifically in the Basic Payment Scheme, where minimum activity rules may be an impediment to upland farmers (e.g. meeting the average stocking level requirement, or the alternatives of an environmental management agreement/agri-environmental commitment). Within the present Agri-Environment Climate Scheme (AECS), there are limited options for peatland management, within both upland and lowland settings (https://www.ruralpayments.org/topics/all-schemes/agri-environment-climatescheme/management-options-and-capital-items/#604269). Uptake of existing AECS measures In order to look at one example of the current level of uptake of measures, we extracted the uptake of the following AECS measures from 2016 to 2020. No GIS intersect was applied, i.e., we did not check whether the extracted data matched modelled peat extent (or condition class) in the UKGHGI. 1. Lowland Bog Management 2. Wetland Management 3. Management of Buffer Areas for Fens and Bogs 4. Moorland Management Data prior to 2016 could not be extracted with the available resources on this project as those had been historically shared with JHI as a bespoke load for a specific ask from SG, and so isn’t part of our standard data access. In addition, the IACS system changed in 2015, so it would be necessary to look at RP and LMO measures to ensure the full picture further back in time. Our analysis included a summary of capital items and annual recurrent items. For this analysis, there are also number of caveats to consider. In some cases, JHI data access allows us to identify a single field, but for others the data only allow identification of a business (we can then get the fields associated with that business, but not which field the specific measure was applied to). For the capital items claimed (Annex 3 for full table), few could be unequivocally assigned to constituting a full or partial rewetting measure, with the exception of ditch-blocking capital items (creation of dams). JHI access permissions to the IACS claim database also allow us to enumerate the number of “units” within each claim, so if a linear option is recorded, we are able to extract the number of meters (such as of a fence), for area options, the hectares, or for most other capital items, the “count”. This latter option applies for drain-blocking measures. For the period 2016-2020, a total of 34,531 dams were recorded. Although the spatial data of the actual locations of these dams do not exist within the IACS database, the measure stipulates a minimum 30 m distance between dams. Therefore, this number of dams would have resulted in a conservatively estimated 1036 km of blocked drains, or, if converted to area using a 30 m buffer along a linear feature, an estimated 3108 ha of drain blocking in addition to work carried out under Peatland ACTION. Further work could verify this estimate by overlaying the relevant IACS fields with high resolution aerial imagery and extracting the relevant blocked drains via either manual or automated (e.g., object-oriented) feature extraction. This data would need to be further QC’ed by checking ay areas against peat extent models/maps and removal of any accidental double claims with e.g., Peatland ACTION (e.g., AECS measure followed by further, more closely spaced, drain blocking under Peatland ACTION funding at a later time point). Other capital item claims identified in Annex 3 have a more tenuous connection with rewetting. Control of scrub may result in a partial recovery of the water table on peatland sites, given that a system component that contributes to evapotranspiration losses of water is removed, however empirical evidence of this management having such an impact is currently lacking. Similarly, there are several capital items that may result in localised (e.g., stock bridges, fencing, bracken treatment) or larger scale (open range deer management) reductions in grazing offtake as well as reducing evapotranspiration losses. Again, empirical evidence of such impacts is currently not available. There were also two capital items with low uptake related to wetland creation, which included field drain breaking and pipe sluicing. Without spatial information that would allow us to reference this management type as to whether it occurred on peat, it is not possible to say with even low certainty that this could be a partial rewetting measure. In general, uptake of the capital items related to specific peatland or moorland management measures was low during 2016-2020 (Annex 3). Similar to the capital items, uptake of the recurrent element was relatively low (Annex 4) By how much or how quickly might emissions fall? There are no precedents in Scotland for partial rewetting where evidence may be obtained from. On timelines for emissions reductions, we don’t have much evidence relevant to Scotland even from peatland restoration projects, although analyses of Peatland ACTION hydrological datasets suggest that projects in climatically wetter areas fare better (Lucy Elliff, pers comm from findings of an as yet unpublished commissioned NatureScot report; https://www.nature.scot/doc/peatland-actionmonitoring-strategy). There are very few datasets within NatureScot’s Peatland ACTION monitoring network where before-after changes have been monitored and also there are no ongoing comparisons of unrestored versus rewetted scenarios. There are no studies to date that have examined the greenhouse gas emissions impacts of partial rewetting; there are still very few studies to date to evidence the emissions savings from ‘full’ rewetting in a UK setting. Most of the evidence is with regards to botanical recovery 54 , which is often used as a proxy for emissions based on data from chamber-based evaluations in temperate European sites 55 , but this assumption that botanical recovery can be used as a proxy for greenhouse gas emissions has not been tested in full against landscape-scale, eddy covariance-based data in a UK setting. For a single location in Scotland, restored a blanket bog after afforestation led to the site performing as a carbon dioxide sink after >15 years had passed since the felling and rewetting work, whereas a younger (10 years since rewetting) site was still functioning as a net carbon dioxide source 56 . Similar observations have been made in a Canadian study, where formerly extracted peatlands were restored using the moss transfer technique 57 . In both of these cases, it is important to note that the rewetting technique generates a lot of labile carbon (from wood chips and needles, or straw, respectively), which may contribute to the lengthy time until the site returns to a net carbon dioxide sink function. In a German site that was previously used for peat extraction, net losses of carbon were still found 18 years after rewetting 58 . In contrast, ditch blocking of a formerly drained but never extracted site in Ireland led to immediate cessation of carbon dioxide losses in the year after rewetting, following by increasing sink strength over the next 5 years 59 . Some of the carbon gains are offset by increases in methane production in all peatland restoration projects. In the Irish example, increased methane production was indeed also observed post-rewetting, but not to such an extent as to offset carbon dioxide uptake42. The net production of methane is inversely related to the depth of the water table 60 , and so, with appropriate water level management, the negating impact of methane on overall carbon balances can be mitigated by aiming for water levels that result in net carbon sink function. Overall, it is therefore likely that rewetting minimally degraded (drained/modified) peatlands is likely to have an immediate impact of emissions reductions to net zero or indeed carbon sequestration, whereas more degraded starting conditions (agricultural sites, forestry, erosion) may require a longer timeframe before significant reductions will be achieved. 54 https://www.conservationevidence.com/actions/1756 55 https://link.springer.com/article/10.1007/s10750-011-0729-x 56 http://mires-and-peat.net/media/map23/map_23_05.pdf 57 https://iopscience.iop.org/article/10.1088/1748-9326/ab56e6/meta 58 https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020JG005960 59 https://onlinelibrary.wiley.com/doi/full/10.1111/gcb.16359 60 https://www.nature.com/articles/s41586-021-03523-1 Annex 1: Climate change precipitation and temperature anomaly for all 12 projections (as numbered in the plots) used under RCP8.5 for 2020-2049 and 2050-2079 periods with respect to 1960-1989 baseline. Reproduced from Project D5-2 report (dated 23rd of June 2023). Annex 2. Map of the total carbon storage in the top 1 m of peat soils, as per Aitkenhead and Coull (2019) 61 . Direct reproduced from journal article (not for distribution). 61 https://bsssjournals.onlinelibrary.wiley.com/doi/abs/10.1111/ejss.12916 Annex 3. Uptake of capital items under relevant peatland or wetland AECS schemes AECS capital item type by Scheme Option (in bold) Total number of business holding submitting claims (may contain successive year claims for the same business) Lowland bog management with grazing 72 Control of scrub - Follow up treatment 1 Control of scrub/woody vegetation - Primary treatment - Heavy vegetation category 3 Control of scrub/woody vegetation - Primary treatment - Intermediate vegetation category 3 Control of Scrub/Woody vegetation - Primary treatment - Light vegetation category 6 Control of scrub/woody vegetation - Removal from site of the cut vegetation 3 Ditch blocking - Peat dams 10 Ditch Blocking - Plastic Piling Dams between 1 and 2m wide 5 Ditch Blocking - Plastic Piling Dams over 2m wide 6 Ditch Blocking - Plastic Piling Dams up to 1m wide 1 Fence removal 3 Gate - Stock fence 10 Muirburn and heather cutting 1 Stock bridges for bog, fen or wetland management - Large, greater than 2.6m span 1 Stock bridges for bog, fen or wetland management - Small, up to 2.6m span 1 Stock fence 14 Stock fence - Shared boundary 3 Temporary electric fencing 1 Lowland bog management without grazing 21 Control of scrub - Follow up treatment 1 Control of scrub/woody vegetation - Primary treatment - Heavy vegetation category 2 Control of scrub/woody vegetation - Primary treatment - Intermediate vegetation category 2 Control of Scrub/Woody vegetation - Primary treatment - Light vegetation category 5 Ditch blocking - Peat dams 6 Ditch Blocking - Plastic Piling Dams over 2m wide 1 Ditch Blocking - Plastic Piling Dams up to 1m wide 1 Fence removal 1 Gate - Stock fence 1 Stock fence 1 Management of buffer areas for fens and lowland bogs 28 Ditch blocking - Peat dams 1