Explanatory Memorandum tothe Water Resources (Control of Agricultural Pollution) (Wales) Regulations 2021.This Explanatory Memorandum has been prepared by the Department for Economy, Skills and Natural Resourcesand is laid before Senedd Cymruin conjunction with the above subordinate legislation and in accordance withStanding Order 27.1.Minister/Deputy Minister’s DeclarationIn my view, this Explanatory Memorandum gives a fair and reasonable view of the expected impact of the Water Resources (Control of Agricultural Pollution) (Wales) Regulations 2021.I am satisfied that the benefits justify the likely costs.Lesley Griffiths MSMinister for Environment, Energy and Rural Affairs27January2021

1.DescriptionThe Regulations establish measures to protect the environment from pollution caused by agricultural activities.The Regulations impose limits on the amount of nitrogen from fertilisers which may be applied to land;a requirement to undertake nutrient management planning;controls on where, when and how nutrients are applied and ensuresthe storage of manure is appropriate forit to be utilised efficiently.The Regulations also includesprovisions to include a review of proposals (if any are submitted within 18 months of the Regulations coming into force) on an alternative suite of measures to those in these Regulations to prevent or reduce pollution caused by agriculture.2.Matters of special interest to the Legislation, Justice and Constitution Committee.None3. Legislative backgroundThe Regulations are made using the powers conferred by sections 92 and 219(2)(d) to (f) of the Water Resources Act 1991.The Regulations make provision in accordance withCouncil Directive 91/676/EEC concerning the protection of waters against pollution by nitrates from agricultural sources (OJ No. L 375, 31.12.91, p. 1) and aspects of Directive 2000/60/EC establishing a framework for Community action in the field of water policy (OJ No. L 327, 22.12.2000, p. 1).Sections 92 of the Water Resources Act 1991 gives the Welsh Ministers the power to make regulations for preventing and controlling any poisonous, noxious or polluting matter for the purpose of preventing or controlling the entry of the matter into any controlled waters. Functions of the Secretary of State under section 92 and section 219 were transferred to the National Assembly for Walesunder article 2 of, and Schedule 1 to, the National Assembly for Wales (Transfer of Functions) Order 1999 (S.I. 1999/672). As regards section 92, functions were transferred in relation to those parts of Wales which are outside the catchment areas of the rivers Dee, Wye and Severn. In relation to those parts of Wales which are within those catchment areas,functions under section 92 areexercisable by the National Assembly for Wales concurrently with the Secretary of State. By virtue of section 162 of, and paragraph 30 of Schedule 11 to, the Government of Wales Act 2006functions under sections 92 and 219 now vest in the Welsh Ministers.These Regulations are being made under thenegative resolution procedurein accordance with section 219(1) of the Water Resources Act 1991.34.Purpose andintended effect of the legislationAgricultural activities are one of the main causes of water pollution and ammonia emissions which are detrimental to public health, the environment, biodiversity and the economy. While many farms in Wales operate to high standards, comply with the regulatory baseline and follow good practice guidance, many do not. The Regulations target agricultural activities which present a risk of pollution to reduce the level of environmental pollution caused by poor practice.The Regulations will protect water (and air quality)from poor agricultural practice by reducing losses of pollutants from nutrients across the whole of Wales. Currently, regulations for the protection of the environment from agriculturalpollution arelimited. In the absence of an improved regulatory baseline,detrimental impactson the environmentand the resilience of ecosystems necessary to enhance and protect biodiversityand public health will continue to occur. Wales’ agricultural industry may also be harmed if the regulations are not introduced, particularly where compliance or regulatory equivalence is necessary for trade purposes. The regulations will fulfil existing requirements under the Nitrates Directive and Water Framework Directive to minimise this risk.The Regulations will enable more efficient use of nutrients and enable the agricultural industry to demonstrate improved production standards. The Regulations will also protect farms performing to good or high standards from the reputational damage to the industry caused by poor practice elsewhere. The measures in the Regulations are expected to reduce losses of pollutants to the environment each year by approximately 2,000tonnes, an environmental benefit equating to £300m. Thisincluding nitrates, phosphorus, ammonia and nitrousoxide.The biggest impact on nitrate losses is attributed to increased slurry storage, phosphorus and nitrous oxide losses from not spreading at high risk times and ammonia from integrating fertiliser and manure applications. Due to the large range of potential environment costs associated with these pollutantsand the variability of farm types and practices, there can be no certaintyofthe cost benefit ratio. While the impact of the measures will be minimal for farms already compliant with existing regulations and which follow good practice guidance, the greatest costs are attributed to those businesses not compliant with existing regulatory measures and which do not follow good practice recommendations. In this respect, the Regulations are proportional and aligned to the polluter pays principle.4Regulatory Impact AssessmentIntroductionThis report is an impact assessment of a potential policy change to implement measures to address agricultural pollutioninWales. Wales’ natural resources are among our most valuable assets. They provide essential services including food, water and land. These are as fundamental to the long-term success of our economy as they are to the quality of our natural environment and the well-being of our communities.These resources are under pressure from challenges, including agricultural pollution. A significant proportion of Wales’ nutrient input to the environment originates from diffuse pollution, individual small sources of pollution which collectively cause asignificant impact. Agricultural activities are one of the main causes of water pollution and ammonia emissions which are detrimental to public health, the environment, biodiversity and the economy. Acute point-source pollution incidents also effect water quality and can cause significant losses in biodiversity in large stretches of the aquatic ecosystem. It can take many years for full recovery to be achieved following large scale incidents, if at all. While the primary intention of the proposal is to reduce water pollution from agriculture the approach will be advantageousto other policy aims such as reduced atmospheric emissions. The proposed measures are designed to avoid pollution swappingandprevent or minimise increased losses of nutrients to the environment(including greenhouse gases, phosphorus and ammonia)as a result of measures primarily focussed on reducing losses of nitrogen.The following key policy options are considered in this impact assessment,with the measures under each option listed in Table 1-1:•Option 1–Doing nothing:2.4% of Wales remains designated as Nitrate Vulnerable Zones (NVZs).There would be no change to the existing situation.This option provides the baseline against which the costs and benefits of the following options will be assessed.Option 2-Apply measures to the whole of Wales with a review clause to consider the introduction of earned autonomy. •Option 3–Designate additional areas as NVZs (8% of Wales). •Option 4–Introduce regulations across the whole of Wales; with 8% designated as NVZ and different measures elsewhere; with a review clause for earned autonomy. 5Table 1-1 Measures and spatial applicability under the different policy optionsOption 2Option 3Option 4All WalesProposed NVZ AreaProposed NVZ AreaRest of WalesUse a fertiliser recommendation systemIntegrate fertiliser and manure nutrient supplyDo not apply manufactured fertiliser to high-risk areasAvoid spreading manufactured fertiliser to fields at high-risk timesIncrease the capacity of farm slurry stores to improve timing of slurry applications (5-month storage requirement)Increase the capacity of farm slurry stores to improve timing of slurry applications (4-month storage requirement)Do not apply manure to high-risk areas Do not spread slurry or poultry manure at high-risk times Do not spread FYM to fields at high-risk times For each of these options, it was assumed that compliance with the measures would increase from the current practice (which may be compliance with existing regulation and is described within this report) to full compliance with the new measures. The impacts of adding the following measures to option 4 were also considered:•Do not apply manufactured fertiliser to high-risk areas•Avoid spreading manufactured fertiliser to fields at high-risk times (no person may spread nitrogen fertiliser if the soil iswaterlogged, flooded, snow covered, frozen or has been frozen for more than 12 hours in the previous 24 hours and weather conditions must be taken into account –no closed period applies).However, the definition of ‘high risk times’ for fertiliser applications in this report (see Section 2.2.4) negated the need to model these measures, as option 4 with the measures included is effectively the same as option 2. The costs and environmental impacts of implementing the measure ‘high risk times’ for applications of manufactured fertiliser are very uncertain as they will mainly depend on 6soil and weather conditions in early spring (i.e. February and March). If fertiliser applications are delayed until after the end of March there is an increased risk that cropyields will be affected as a result of sub-optimal crop nutrient supply. In order to assess the uncertainty associated with implementing the ‘high risk times’ measure, two options were considered, for each scenario where option ‘a’ avoided fertiliser applications between October and March and option ‘b’ avoided fertiliser applications between October and February.The options considered reflect the requirements of European Directives, including the Nitrates Directive and Water Framework Directive,as well asretained EU Lawand the responses to relevant published consultations. The responses to the consultation on theReview of the Designated Areas and Action Programme to Tackle Nitrate Pollution in Wales were the key element of the policy development. Responses to other related consultations, including on the storage of silage and slurry and the sustainable management of natural resourceswere also considered. The consultations referred to can be accessed using the following links:https://gov.wales/nitrate-vulnerable-zones-waleshttps://gov.wales/review-water-resources-control-pollution-silage-slurry-and-agricultural-fuel-oil-wales-regulationshttps://gov.wales/taking-forward-wales-sustainable-management-natural-resourcesConsultation with stakeholders has taken place through the Wales Land Management Forum sub-group on agricultural pollution, as well as with individual stakeholders. This includes affected individuals and internal consultation withWelsh Government officials to ensurepolicy alignment.Minutes of meetings of the Wales Land Management Forum sub-group and a progress report on the work of the sub-group can be found using the following link:https://naturalresources.wales/guidance-and-advice/business-sectors/farming/wales-land-management-forum-sub-group-on-agricultural-pollution/?lang=enThe resulting regulations will be reviewed every four years but this will be dependent on our future relationship with the European Union over the coming months and years. The Welsh Government will continue to work with stakeholders, including the Wales Land Management Forum sub-group, as part of the review process. 71Methodology and Assumptions1.1MethodologyA modelling approach was used to estimate the potential effects of different policy scenarios on pollutant loads as well as farm costs. The modelling work consisted of two main parts:a)Using the Farmscoper tool (Gooday et al., 2014) to predict the effects of the proposed measures on pollutant losses as well as on farm costs as relevant to each policy option.b)Using the MANNER-NPKtool (Nicholson et al., 2013) to model the effects of avoiding high risk times for high available N manures (cattle slurry, pig slurry, broiler litter and layer litter) in accordance with the proposed measures.Three additional components of work were undertaken to fully account for the costs of measures and monetise the estimated pollutant reductions:a)Estimate the costs associated with increased slurry storage capacityb)Estimate the costs associated with record keeping and manure and nutrient planningc)Review the damage costs associated with the different pollutantsThe range of potential implementation and damage costs was accounted for with a sensitivity analysis. For some of the key measures (either those with significant costs or greater uncertainty in the costs), high, medium and low cost estimates were produced.The review of damage costs also produced a central estimate and upper and lower bounds for each pollutant. The sensitivity analysis thus considered the consequences of using the high, medium or low implementation costs, and the high, medium and low environmental damage costs. The pollutants considered are nitrate, phosphorus, ammonia and nitrous oxide.The assessment considers the management of livestock manures only and not other organic materials (e.g. biosolids, digestate and compost)1. The assessment does not specifically consider the impacts of the measures on organic farming as this makes up a very small proportion of the agricultural land in Wales.1.1.1FarmscoperThe Farmscoper model is a decision support tool used to assess diffuse agricultural pollutant loads on a farm and quantify the impacts of farm pollution mitigation options on theselosses. It was developed by ADAS with Defra and EA funding and has been used both internally within those organisations and in a number of external projects lookingat the impacts of regulation and agri-environment schemes (e.g. Gooday et al., 2015; Collins and Zhang, 2016; Gooday and Whitworth 2017; Collins et al., 2018; Elliott et al., 2019). The tool allows for the creation of unique farming systems, based on combinations of livestock, cropping and manure management, and the assessment of the cost and effect 1. The N loading from other organic materials (e.g. biosolids, digestate and compost) is estimated at less than 3% of total N inputs (BSFP, 2018).8of one or more mitigation methods from a library of over 100 methods contained within the tool, many based upon the Mitigation Method User Guide (Newell-Priceet al., 2011). The tool can be used to simulate losses from multiple farming systems, to allow predictions at catchment scale or larger. A more detailed description of the model is presented in Appendix 1. The Farmscoper tool was parameterised using June Agricultural Survey (JAS) data from 2018 for Wales. The JAS was used to determine average cropping and livestock for different farm types and sizes. The farm types considered were the 9 robust farm types (RFT), with the Cattle and Sheep LFA RFT further subdivided into Specialist Sheep, Specialist Beef and Mixed; the farm sizes considered were based on standard labour requirement. Separate farms were made for land inside and outside of the proposed NVZ area. The total number of farms in Wales, by type and size, is shown in Table 1-1. Additional management information for these farms was taken from national stratified surveys including the 1stand 2ndWelsh Farm Practice Surveys (Anthony et al., 2011; Anthony et al., 2016), the Defra Farm Practice Surveys and the British Survey of Fertiliser Practice.Table 1-1: Number of farms in Wales by farm type and farm size (based on standard labour requirement)HobbySmallMediumLargeTotalCereal304622529420General Cropping7619923127Horticulture76927925830Specialist Pig220221225Specialist Poultry9813228521,093Dairy1882053079141,614LFA –Specialist Sheep2,1628635351,1264,686LFA –Specialist Beef81613824311,009LFA –Mixed Livestock3,8031,0666278606,356Lowland Cattle and Sheep1,7504031561912,500Mixed Livestock892110481191,169Other4,28533694634,778Total16,2463,2631,8643,43424,807Pollutant losses were calculated for each of these different farms under each of the soil and climate zones recognised by Farmscoper, with the results expressed as losses per 9hectare. These losses were then mapped back on to the LPIS field parcels (where every field parcel had been assigned to a farm type, size, climate and soil type and either inside or outside the proposed NVZ area).The LPIS dataset contained information for 610,000 field parcels, covering a total area of 1.69m hectares. For 545,000 of these parcels, a farm ID was provided, for 23,470 different farms. Of these farms, 14,663 could be directly linked to JAS farms (which accounted for 1.20m hectares). A further 5,898 farms (62,759 fields; 0.18m hectares) were joined to JAS farms by matching as close as possible the LPIS area of a farm with the JAS area for all unmatched farms, with matches constrained by the Small Area (a spatial designation) allocated to each JAS farm and the Easting and Northing provided. A total of 4,246 of the JAS farms (out of 24,807) were unaccounted for, and these were classed as either ‘other’ or hobby farms (which accounted for over 85% of the unmatched Jas farms) which have low nutrient use and very few livestock. In the creation of the ‘average’ farms used in the Farmscoper modelling, the livestock numbers were scaled to ensure the total livestock numbers across Wales (and within the proposed NVZ area) remainder close to the JAS totals when distributed across the LPIS parcels. Although there is some uncertainty about the accuracy of the mapping of the farm data, the methodology was designed to preserve JAS livestock numbers and LPIS land areas. As the results in this report are being summarised atnational (or proposed NVZ) scale, the spatial uncertainty has limited impact on the overall modelled outputs.Changes in pollutant losses predicted by Farmscoper due to measure implementation depend on (i) the effectiveness of the measures at reducing pollution and (ii) the current (and future) uptake of the measures. Parameterisation of these values are based upon the scoring system shown in Table 1-2, with a central value selected that represents the range within which the impact or implementation is expected to be2,-the values selected to parameterise the different mitigation measures are described in the following sub-sections. Farmscoper uses a source apportionment coordinate system, so the impact of a mitigation measure may be targeted at one (or more) of the coordinates –for example buffer strips may reduce losses by 50% in surface runoff, but have no impact on losses in drain flow or to groundwater.2This could reflect, for example the uncertainty in survey data or its applicability, or the variation in evidence for effect. The use of a scoring system allows for easy comparison between the different pollutants and multiple mitigation methods within Farmscoper.10Table 1-2Confidence ranges and central values used by Farmscoper for estimating current implementation of measures and impact potentialCategory Implementation or Impact (%) Uncertainty Range Description A --None B 2 0 to 10 Very Low C 10 2 to 25 Low D 25 10 to 50 Moderate E 50 25 to 80 High F 80 50 to 95 Very High G 100 100 Total 1.1.2MANNER-NPKThe MANNER-NPKmodel (details presented in Appendix 2) is a decision support tool designed to show the impact of different application timings and methods on losses of nitrate, ammonia and nitrous oxide (Nicholson et al., 2013). MANNER-NPKwas used to model the impacts on N loss of introducing the closed period for spreading high N available manure (cattle slurry, pig slurry, broiler litter and layer litter) across the whole of Wales or relevant NVZ areas. The MANNER-NPKdecision support tool is recognised as the industry standard tool for estimating crop available nutrient supply, nitrate leaching and ammonia volatilisation losses following manure applications. It was used to derive the ‘look up’ tables in AHDB’s Nutrient Management Guide (AHDB, 2020) which detail crop available N supply from contrasting manure application timings and methods

11fertiliser recommendation system ensures that the necessary quantities of nutrients are available when required for uptake by the crop. Nutrients are only applied when the supply of nutrients from all other sources is insufficient to meet crop requirements. As a result, the amount of excess nutrients in the soil is reduced to a minimum. Use of a recommendation system should also ensure that the soil is in a sufficiently fertile state to maximise the efficient use of nutrients already in the soil, or supplied from other sources such as fertilisers/organic manures. Maintaining an appropriate balance between different nutrients (i.e. NPK) is also important to maximise the efficient uptake of all nutrients and reduce environmental losses to a minimum. (i)NitrogenMost agricultural soils require applications of nitrogen from fertiliser and/or organic materials on an annual basis to ensure optimum crop growth. Most of the mineral nitrogen in the soil is present as nitrate, which is mobile in the soil. Any nitrate that is present in the soil at the start of the winter is unlikely to be taken up by crops as growth slows due to cold temperatures and reduced light intensity. When excess winter rainfall occurs, and water drains through the soil the nitrate is at risk of being lost from the soil by leaching. Figure 1-1: Impact of manufactured fertiliser nitrogen applications on winter wheat yields and nitrate leaching losses (Lord and Mitchell, 1998)Nitrogen applications to arable crops that supply less than economic optimum will result in sub-optimal crop yields and quality whilst applications that exceed crop requirement will increase the risk of nitrate leaching (Figure 1-2; Lord and Mitchell, 1998; Figure 2-2 Johnson et al., 2011). 12Figure 1-2: The effect of nitrogen fertiliser applications on drainage water nitrate concentrations and nitrate leaching losses (Johnson et al.,2011)Nitrous oxide emissions occur from soils as a result of the microbially mediated processes of nitrification and denitrification. Factors that affect nitrous oxide emissions include soil moisture content, temperature and mineral nitrogen content. Generally nitrous oxide emissions are related to nitrogen inputs from manures and fertilisers with elevated emissions where nitrogen supply exceeds crop requirement (Figure 1-3; Cardenas et al.,2010). Figure 1-3: The effect of manufactured fertiliser nitrogen application rate on nitrous oxide emissions at 3 contrasting grassland sites (Cardenas et al.,2010). 13(ii)PhosphorusA large proportion of phosphorus (P) in soils is bound in forms that are not readily available to the plant or at risk of leaching to water (i.e. fixed or residual P), because of the strong affinity that some soil substances (clays, iron-Fe/aluminium-Al/calcium-Ca) have for P (Holford, 1997). Consequently, managing crop available P supply is based on maintaining sufficient amounts in the soil for the needs of a crop rotation rather than an individual crop. AHDB’s Nutrient Management Guide (RB209) uses a soil P index system (based on the Olsen extractable P levels in topsoil) to provide guidance on P supply from manufactured fertilisers and organic materials. For grassland and most arable crops the target soil P index is 2 (16-25 mg/l Olsen P). For soils below the target index it is recommended to apply P at rates that exceed crop offtake to ensure optimum crop yields and to build up soil reserves. Where soils are at target index, fertiliser rates should match crop offtake to maintain soil fertility at optimum levels and where soil P levels are above target index, P fertiliser applications are not recommended as they represent an unnecessary cost and increase the risk of P losses to water (Figure 1-4; Poulton et al.,2013, Heckrath et al.,1995 Withers et al.,2017). Figure 1-4: The impact of Olsen extractable P levels on crop yields and soluble P losses to water (Poulton et al., 2013, Heckrath et al.,1995). Graph taken from Withers et al.,(2017).The extent to which soil is saturated with P will influence the risk of P lossesto water. The soil saturation capacity depends on the quantities and forms of Fe, Al and Ca present in the soil and P is more strongly bound in the order Fe>Al>Ca (Withers, 2011). Risks of P loss to water have been reported to greatly increase once P saturation exceeds a threshold of 20-30% (Heckrath et al., 1999, Kleinman et al., 2000; Nair et al., 2004). P saturation threshold broadly equates to Olsen soil P indices of 3, 4 and 5 for sand, loam and clay soils, respectively. Consequently, soils with P indices above these levels represent an increased risk of P losses to water.14At the farm level, the impact of fertiliser recommendation schemes on increasing nutrient use efficiency and reducing diffuse pollution will vary depending on the current level of nutrient use. Data from the British Survey of Fertiliser Practice (2018) indicate 88% of tillage land and 52% of grassland in England and Wales received applications of manufactured fertiliser nitrogen in 2017. Fertiliser phosphate was applied to only 44% of tillage land and 30% of grassland in England and Wales. The survey data suggest that this method is likely to have a small overall impact on fertiliser use. The average field rates for nitrogen reported in BSFP are similar to those typically recommended in AHDB’s Nutrient Management Guide (RB209) for arable crops and application rates on grass are typically lower than those recommended in RB209 (Figure 1-5; Figure 1-6). Also, data suggest that applications of phosphate and potash fertiliser have declined over recent years (Figure 1-7) with little scope for further reductions.Figure 1-5: Average nitrogen fertiliser rates appliedto tillage crops across England and Wales (Taken from BSFP, 2018)15Figure 1-6: Total nitrogen fertiliser use across tillage and grassland in Great Britain (BSFP, 2019)Figure 1-7: Average field phosphate and potash rate across England and Wales (BSFP, 2019)Anthony et al (2012) reported that the majority of farmers in Wales either used their own knowledge (64%) or took professional advice (35%) when estimating fertiliser requirements and that only 4% claimed to use RB209 or any software directly 16themselves revealing scope for improvement. Anthony et al (2016) reported that 39-57% of surveyed farmers in Wales used a fertiliser recommendation system.Based on expert knowledge, responses to the survey and modelling results it was determined that the use of a fertiliser recommendation systems would result in no reduction in phosphorus losses and between a 5 –10% reductions in nitrate losses from arable and grassland systems. These calculations were based on the following assumptions:•For manufactured phosphorusthe use of a fertiliser recommendation system was interpreted as ensuring a balance of annual fertiliser input and crop off-take to maintain an appropriate soil phosphorus index.•For nitrogen it was assumed that the use of a recommendation system would enable an improvement in the precision of applications rather than a reduction in the quantity applied. The NITCAT and N-CYCLE models (Lord, 1992; Scholefield et al., 1991) were then used to calculate the effect of a 25% reduction in the average error in estimating optimum nitrogen for the crops on each of representative farm types.After discounting livestock manure phosphorus input-offtake balances were all negative and indicated that fertiliser applications could not be reduced in the absence of manures. There was, therefore, no direct impact of this mitigation method on the use of manufactured phosphorus fertiliser.Anthony et al., (2012) conclude that more precise use and application of manufactured nitrogen fertiliser is likely to reduce nitrate lossesfrom the combined fertiliser and soil nitrogen supply by between 5 and 10% for both arable and grassland.Newell Price et al.,(2011) suggested that the use of fertiliser recommendation systems had the potential to reduce nitrogen and phosphorus losses to water and ammonia and nitrous oxide emissions to air by c.5%.Williams et al., (2017) suggest that where excess nutrients areapplied implementing a nutrient management plan can reduce fertiliser costs and risks of water and air pollution. However, where insufficient nutrients are applied a nutrient management plan may lead to increased fertiliser use which may increase absolutelosses to the environment but reduce losses per unit of production. Information from Welsh Farm Practice Survey (Anthony 2012) reported 43% of farmers have a soil nutrient plan suggesting there was scope to improve the precision of fertiliser application rates for each year.Representation in ModellingBased on the information described above, Farmscoper assumes that losses associated with nitrogen fertiliser will be reduced by 10%, whilst those with phosphate fertiliser by 2%, reflecting the lower potential for changes in P fertiliser usage.Farmscoper assumes that improvements in nutrient use efficiency that come from matching crop available nutrient supply to crop demand and soil nutrient status, ensuring optimal fertiliser timings and the maintenance of soil pH at target levels will reduce average fertiliser inputs by 5% on arable farms and increase average 17productivity of grassland by 10% compared with baseline, which equates to savings of approximately £5 ha-1and £11 ha-1respectively. Current implementation of this measure in Farmscoper is assumed to be 50% as a baseline, with rates higher inside NVZ areas and lower on extensive grazing systems. This is based on Defra Farm Practice Survey (2012), which found 16% and 48%of farmers use the Tried and Tested paper based planning system or PLANET software, respectively, and the 2ndWelsh farm Practice Survey (Anthony et al., 2016) which found 57% of Dairy farmers used a fertiliser recommendation system, but only 40% of cattle and sheep farmers did.1.2.2Integrate fertiliser and manure nutrient supplyDescriptionOrganic materials are valuable sources of plant nutrients and if used effectively they can reduce the need for applications of manufactured fertilisers to meet optimum crop needs (Table 1-3). Fertiliser recommendation systems (e.g. RB209, PLANET, MANNER-NPK and other supplementary guidance) provide guidance on how to make full allowance of the nutrients applied in organic manures and reduce manufactured fertiliser inputs accordingly. Laboratory analysis of manures can provide better understanding of manure nutrient contents and supply. MANNER-NPKinformation on application rates, timings and methods can be used to quantify crop available nutrient supply and provide estimates of nitrogen losses to water and air following application. The nitrogen fertiliser replacement value of organic manures can be increased by applying manures in spring to reduce nitrate leaching losses. For slurries, the use of precision application techniques can reduce ammonia emissions and ensure that applications are spread evenly across known bout widths. In order to maximise the nitrogen value of slurry and poultry manures it is usually necessary to apply them in spring to minimise nitrate leaching losses. The use of low emission spreading techniques such as trailing hose on arable land and trailing shoe and shallow injection of grassland will reduce ammonia losses and further increase the nitrogen value of slurry.For solid manures it is likely that applications will supply more phosphate and potash than is used by a crop in a single year. Consequently, annual applications of manure to the same field can increase soil P contents to levels where there is an increased risk of P losses to water. Targeting manure applications to fields where soil P and K status are below target indices will maximise manure fertiliser replacement values.18Table 1-3: Nutrients supplied by spring application timings of different organic materials (based on typical manure analysis figures in AHDB’s Nutrient Management Guide (RB209))Manure typeApplication Rate (t/ha)Crop Available N (kg/ha)Total P2O5(kg/ha)Total K2O(kg/ha)Crop Available SO3(kg/ha)Pig Slurry3563638412Pig FYM352521028030Cattle Slurry40364812810Cattle FYM402412832014Poultry Manure87220014438Biosolids Cake20333601224The impact of this measure on reducing diffuse pollution will depend to what extent farmers are already accounting for nutrients supplied by organic materials when planning their manufactured fertiliser use. The BSFP (2019) suggests that where farmers have used organic materials manufactured fertiliser nitrogen and phosphate applications were reduced by c. 20 kg/ha N and c.15 kg/ha P2O5, respectively. The savings in nitrogen fertiliser use as a result of integrating manures into nutrient management plans will represent an annual saving once the method has been adopted. However, the P and K value of the manure applications will depend on the P and K status of the soil. Where soils are deficient in P and K i.e. atsoil index 0 and 1, then the available crop P and K fraction of the manure should be accounted for. When soils are P and K index 3, there is no requirement for fertiliser P and K for grass and arable crops, consequently the P and K applied by the manures will have no value. Information from the PAAG suggest that c. 30% of soils in Wales exceed target levels and will not require annual P and K inputs from either manufactured fertilisers or organic materials to support optimum crop growth.In order to identify the maximum and minimum cost benefit for this measure two scenarios (i) accounting for manure N only and (ii) accounting for all manure nutrients have been assessed for each option.Representation in ModellingFarmscoper assumes that fertiliser losses could be reduced by up to 25%, depending upon the amount of manure applied relative to the amount of fertiliser currently used.Farmscoper assumes a saving is made due to reduced fertiliser usage, which is estimated at £6 per tonnes of FYM and £3 per m3of slurry and £28 per tonne of poultry manure. These figures are based on current fertiliser prices for nitrogen, phosphorus and potash, assumed nitrogen efficiency and nutrient availability. However, there is uncertainty surrounding the fertiliser replacement value of the manure. It is possible to account for the nitrogen fertiliser replacement value of the manures as in the vast majority of agronomic situations annual applications of nitrogen are required for optimal crop growth. In contrast the phosphate and potash value of the manure applications will depend on the supply of these nutrients from the soil with no requirement for manufactured fertiliser P and K inputs to arable and grass crops when soils are at or above soil index 3 and for horticulture, potatoes and maize crops when soils are at or above index 4. Accounting for only the nitrogen in manures reduces the savings in slurry to £0.6 per m3and poultry manure to £4 per tonne. There is not assumed to be a reduction in the number of fertiliser applications, which could result in an additional cost saving. The 2ndWelsh Farm Practice Survey (Anthony et al., 2016) found that the percentage of farmers using professional advice or manure testing, and standard values such as RB209 to assess the nutrient value of spread manures was 19 and 11% respectively. However, the majority of farmers (73%) assessed the nutrient value of spreads manures using own knowledge and experience, whereas 20% of farms did not assess at all. Of these, 50% solely rely on own knowledge or experience when assessing the nutrient value of spread manures. The Defra Farm Practice Survey (2012) found 57% of farms assess or calculate the value of their manures, and only 24% tested the nutrient content by taking samples. Based on this current implementation of this measure in Farmscoper is assumed to be 50% as a baseline, with rates higher inside NVZ areas and lower on extensive grazing systems. Information provided by Menter a Busnes (Cate Barrow, Pers |comm) suggest that since 2016 c. 3,000 nutrient management plans have been completed via Farming Connect. This may suggest that the 50% baseline in Farmscoper is an underestimation of the implementation of this measure. However, details of farm type, size and nutrient use for the farmsand information on whether farmers are following the plans is not available. 1.2.3Do not apply manufactured fertiliser to high-risk areasDescriptionDo not apply manufactured fertiliser to field areas where there is a significant risk of fertiliser getting into surface water.This could include sloping land or areas where there are direct flow paths to watercourses,for example, areas with a dense network of open drains, wet depressions (flushes) draining to a nearby watercourse, or areas close to road culverts/ditches. The risk of pollution is reduced by not applying fertiliser at any time to hydrologically well-connected areas where it could easily be transferred to a watercourse.Not applying fertiliser to crops will significantly reduce yields as there will be insufficient crop available nutrient supply to support optimum crop growth.The following evidence suggests that ‘high risk areas’ occupy approximately 5% of the agricultural area:•Compaction due to machinery: Anthony et al (2012a) found this was reported on 25% of dairy farms and 10% of cattle and sheep farms. The compacted area within such fields is estimated at 1-2% of the total area.•Poaching from livestock: Gooday et al (2015) reviewed a range of evidence which suggested that 3% of field areas had visible poaching damage from livestock. Observations suggest that an area of poaching around a livestock feeder or trough can cover 20m around the feeder, which equates to c2% of a 5-ha field.•Anthony et al (2012b) surveyed areas of soils with tile drainage and the area of land affected by evidence of drain failure. The proportion of land affected by sustained waterlogging ranged from 2% on arable farms to 13% on upland cattle 20and sheep farms. As a proportion of all soils (not just those with tile drainage), the affected area would be a smaller percentage.There are no surveys which provide information on the amount of fertiliser applied to steeply sloping land. In this study it is assumed that nationally very small amounts of fertiliser are applied to steeply sloping land due to the practicalities involved, and so any additional impacts from avoiding these areas have not been accounted for.Representation in ModellingFarmscoper assumes losses associated with fertiliser on the ‘high risk areas’ are entirely negated.The 2ndWelsh Farm Practice Survey (Anthony et al., 2016) found a baseline of 56% of dairy farmers had a soil nutrient management plan, but only 25% of cattle and sheep farms in SDAs. Farms in Glastir, Tir Gofal or Tir Cynnal were more likely to have soil nutrient management plans. Farmscoper assumes a baseline of 50% of ‘high risk areas’ are avoided, with values greater inside NVZs and lower on cattle and sheep farms. Farmscoper assumes a 50% yield reduction for arable crops and 30% reduction in grass yields over 5% of the agricultural area as a result of implementing this measure, which equates to £210 ha-1and £600 ha-1for high risk areas on grassland and arable land respectively. There would also be the need to identify high risk areas, typically through the creation of a nutrient management plan. Costs of this are dealt with separately (see Section 1.4.1).1.2.4Avoid spreading manufactured fertiliser to fields at high-risk timesDescriptionDo not spread manufactured fertiliser at times when there is a high-risk of surface runoff or rapid movement to field drains i.e. when soils are ‘wet’. Do not spread N fertiliser between September and February when there is little or no crop uptake and there is a high-risk of nitrate leaching loss (unless there is a specific crop requirement during this period).Fertiliser timing affects the potential for mobilisation of nutrients from land to water. Avoiding spreading fertiliser to fields at high-risk times reduces the availability of N and P for loss in surface runoff or drain flow. Surface runoff is most likely to occur when rain falls on sloping ground, when soils are ‘wet’, frozen or snow covered. The rapid preferential flow, through the soil, of N and P from applied fertilisers is most likely to occur from (drained) soils when they are ‘wet’ and rainfall follows soon after application. Avoiding N fertiliser application in the autumn/winter reduces the amount of nitrogen available for leaching by over-winter rainfall.The risks of water pollution following application of manufactured fertilisers will vary according to soil type which controls the pathway for water and nutrient loss and on the soil moisture content. Nitrate is mobile in soils and is at risk of being leached from the soil when drainage occurs. Phosphorus is more immobile in soils and the risks of phosphorus leaching is highest when soils are saturated with P or where rapid transfer of P from the soil to water occurs following applications of fertiliser P or organic manures.21On sandy soils (which occupy less than 5% of Wales; Figure 1-8), drainage occurs slowly over winter by piston displacement in the unsaturated phase, with wetting fronts moving to depth at rates of a few metres a year depending on drainage volumes and the pore volume of the soil and base rock. Consequently, the highest risk of water pollution on sandy soils is following nitrogen fertiliser applications in the autumn/winter period when the nitrate supplied is unlikely to be taken up by crop growth.Figure 1-8: RB209 soil classification for WalesOn undrained clay and medium loam soils, surface runoff is likely to occur inrapid response to rainfall events, because of the impermeable nature of the soil matrix(Goss et al.,1978). Where an effective drainage system is present, much of the water that would otherwise be lost as surface runoff, will move rapidly from the soil surface through macropores that have developed naturally or have been created through the installation of pipe drains, mole drains or subsoiling fissures, with transit times influenced by rainfall volume and intensity (Goss et al.,1983). On these soil types which occupy the majority of the productive land in Wales (Figure 1-8) the highest risk of water pollution following fertiliser application is likely to be when soils have a soil moisture deficit of less than 10mm –i.e. drainage will occur when hydrologically effective rainfall (I.e. rainfall-evapotranspiration) exceeds 10mm.As part of this study the IRRIGUIDE water balance model (Bailey and Spackman, 1996) was used to quantify high risk times by estimating daily soil moisture deficits for two soil types (sandy loam and clay loam), two crop types (grass and winter wheat) for 9 locations chosen to be representative of contrasting agroclimatic zones across Wales (i.e. Aberystwyth, Llangefni, Bangor, Wrexham, Fishguard, Haverfordwest, Welshpool, Newport Pembrokeshire and Newport Gwent). The model was run using 30 year (1987-2018) average climate data for each site. The model uses information on volumetric 22moisture content, crop cover, rooting depth and weather data to estimate evapotranspiration and soil drainage. Figure 1-9: Average estimated monthly soil moisture deficit in Wales (based on 30 year average climate data from 9 sites across Wales)The model runs indicate that, on average, soils across Wales were close to field capacity at the end of February (i.e. soil moisture deficit close to 0)with soil moisture deficits of less than 10 mm predicted at the end of March (Figure 1-9). There was some annual variation in soil moisture deficit at the end of March. In 6 years out of 30 (1990, 1997, 2002, 2003, 2012 and 2019) soil moisture deficits greater than 10 mm were predicted at the end of March. These drier years contrasted with 6 years (1992, 1994, 2006, 2009, 2010 and 2015) when soils were at field capacity ordrainage was occurring at the end of March. There was little difference in estimated soil moisture deficits between soil and crop types reflecting the low growth rates during the winter months for grass and arable crops. This suggests that applying fertiliser during February and March could beconsidered‘high risk’.There are few studies that have investigated the impact of fertiliser application timings by date on crop yields. Generally, the guidance provides information to ensure that sufficient nitrogen is applied to support crop growth at critical points in the growing season. E.g. In cereals, when the crop is growing rapidly between growth stages 30 and 39 (Stem extension and Flag leaf emergence; Figure 1-10) which usually occurs betweenthe end of March to the middle of May depending on soil and weather conditions.23Figure 1-10: Growth stages for Winter wheat (taken from AHDB’s Wheat Growth Guide(www. ahdb.org.uk/wheatgg)For winter wheat, AHDB’s Nutrient Management Guide recommends that ‘where more than 120kg/ha N is required 40 kg/ha N should be applied between mid-February and early March. The balance of the application should be applied in one or two dressings during early stem extension. Where more than 120 kg N/ha remains to be applied, half should be applied at the start of stem extension (not before April) and half at least two weeks later (not after early May)’. Information provided by ADHB’s Wheat Growth Guide suggests that delaying fertiliser applications until early April is unlikely to have a significant impact on wheat yields in most years. Similarly, on grassland soil and weather conditions are likely to have more significant impact on grass growth than delaying fertiliser applications until early April. However, where weather conditions prevent uptake of nitrogen in April and May (e.g. continued period of dry weather following application) there is a risk that cereal and grass yields will be reduced. Delaying fertiliser applications until the end of Marchis likely to significantly affect crops established in late winter/early spring which have a requirement for fertiliser to be applied in the seed bed. For example, early potatoes, which are typically planted in south-west Wales at the end of January or beginning of February would be particularly disadvantagedif the measure prevented fertiliser applications in February.The crop is usually grown on c. 500-1000 ha.Representation in ModellingIn order to assess the uncertainty of this measure on operational costs and environmental benefits yields two versions of this measure were modelled for each option. For Option ‘a’ fertiliser applications were not allowed from October to March, and Option ‘b’ fertiliser applications were not allowed between October to February.Surveys suggest that a small amount of N and P fertiliser is applied before March (c. 6% of total applications; BSFP, 2018), so the impacts of this restriction window (to end of February) on losses to water were small –a 2% reductionin N and P losses in runoff or drain flow shortly after application (as opposed to residual losses post-harvest for nitrogen, which will be unchanged). With the restrictions lengthened into March, reductions a 10% reduction in nutrient losses from fertiliser applications during this period was assumed.24As the modelling that underpins Farmscoper is based upon fertiliser timing information derived from the British Survey of Fertiliser Practice, current implementation is captured in the modelling and so the implementation of the mitigation measure is set to 0.With restrictions on fertiliser applications to the end of February, it was assumed that crop yields were unaffected, so there was no cost associated with option b. With restrictions to the end of March (option a), Farmscoper assumes a 10% reduction on crop yields one year in 10 to reflect yield reductions that may occur from sub-optimal crop available nutrient supply early in the growing season.1.2.5Increase the capacity of farm slurry stores to improve timing of slurry applicationsDescriptionExpand slurry storage facilities for the collection and storage of slurry, to allow spreading at times when there is a low-risk of runoff and when there is an actively growing crop to utilise nutrients applied in theslurry. The storage provides increased flexibility in land application timing, so there will be fewer occasions when a lack of storage forces slurry applications to occur when here is a high-risk of nitrate leaching, surface runoff or drainflow to water i.e. when soils are ‘wet’.The current statutory requirement for farmers outside Nitrate Vulnerable Zones is to comply with the Water Resources (Control of Pollution) (Silage, Slurry and Agriculture Fuel Oil) (Wales) Regulations 2010 (SSAFO) which requirestorage of 4 months excreta production and an allowance for the highest rainfall expected in 5 years (M5). This method assumes that increasing slurry storage capacity to 5 months excreta production plus M5 rainfall will reduce the likelihood that slurry will be applied to land under conditions which are likely to increase the risk of water pollution.Representation in ModellingFarmscoper assumes that losses of ammonia from manure storage will be increased by 25% due to the increased amount of manure beingstored being increased by 25% (there is potentially a marginal further increase due to the increased surface area of the store), but losses of ammonia from manure spreading will consequently be decreased. The impacts of improved timing of manure applications, facilitated by the increased storage, are described in Section 1.2.7.The costs of implementing this measure were calculated separately (see Section 1.4.2).1.2.6Do not apply manure to high-risk areasDescriptionDo not apply manure to field areas where there is a high-risk of direct loss to watercourses, e.g. directly adjacent to a watercourse, borehole or road culvert, to shallow soils over fissured rock or widely cracked soils over field drains, to areas with a dense network of open (surface) drains, spring lines or wet depressions (flushes). These areas have a high-risk of rapid transport of manure-borne pollutants to watercourses, so manure applications (particularly of slurry) should be avoided wherever possible. 25‘Avoiding high risk areas’ for manure applications is assumed to affect the same area as for fertiliser applications i.e. 5% of the agricultural area. However, it is assumed that there is no impact on crop yields as a result of introducing this measure as the likelihood that manures were the sole source of nutrient inputs to support crop growth in these areas is small.Representation in ModellingFarmscoper assumes short term incidental losses associated with manure on the ‘high risk areas’ are reduced by 80%. Losses are not entirely negated (unlike not applying fertiliser to high risk areas) as the manure will still be applied somewhere.To avoid double counting costs from reduced yields associated with sub-optimal nutrient supply from both this measure and ‘Do not apply manufactured fertiliser to high-risk areas’, and the difficulty in determining what proportions of crop nutrient demands are met by fertiliser or manure (within these high risk areas), the potential yield penalty has been attributed solelyto ‘Do not apply manufactured fertiliser to high-risk areas. In this study, the only costs implementing this measure are associated withthe need to identify high risk areas, typically through the creation of a manure management plan. Costs of this are dealt with separately (see Section 1.4.1).Implementation of this mitigation measure is 80%, with higher rates inside NVZ areas and lower rates on cattle and sheep farms. The Defra Farm Practice Survey (2012) found 65% of grazing livestock farms and 90% of dairy farms had a manure management plan. The 2ndWelsh Farm Practice Survey (Anthony et al., 2016) found a baseline of 83% of dairy farmers had amanure management plan, but only 50-60% of cattle and sheep farms. Farms in Glastir, Tir Gofal or Tir Cynnal were more likely to have soil nutrient management plans.1.2.7Do not spread slurry or poultry manure at high-risk timesDescriptionDo not apply slurryor poultry manure to fields at times when there is a high-risk of surface runoff e.g. in winter when soils are ‘wet’ or frozen hard, or when heavy rain is expected in the next few days. Do not apply slurry or poultry manure to fields at times when there is a high-risk of rapid percolation to field drains e.g. in winter and spring when soils are ‘wet’. Do not apply slurry or poultry manure to fields late in the growing season (i.e. autumn/early winter) when there is no crop to utilise the added N. Slurriesand poultry manures have ‘high’ readily available N contents (>30% of total N). As is the case for manufactured fertiliser applications the risks of nitrate and phosphorus losses to water following slurry applications will vary according to soil and crop type, soil moisture content and rainfall in the days/weeks after application. Data reported by Chambers et al. (2000) suggest that up to 20% of total nitrogen supplied by slurry and poultry manure applied to free-draining soils before the establishment of winter cereals can be lost by nitrate leaching (Figure 1-11). Similarly nitrate leaching following autumn applications to grassland were 15% of total N applied compared with less than 5% from late winter/early spring timings (Figure 1-12). Nitrate leaching occurs following slurry / poultry manure applications in autumn/early winter as a result of readily available N being added to the soil at a time when there is little N uptake by crops. The amount of N lost by leaching is controlled by the amount of readily available N supplied and the 26volume of drainage after application. Nitrate leaching losses from farmyard manure are lower than from slurry and poultry manure applications reflecting their lower readily available N content.Figure 1-11: Nitrate leaching losses following contrasting application timings of slurry/poultry manure and farmyard manure to free-draining arable soilsFigure 1-12: Nitrate leaching losses following contrasting cattle slurry applications to free draining grassland soils (Chambers et al., 2000)On clay and medium soils the risks of water pollution are greatest when slurry applications are made to soils that are ‘wet’. Defra project WQ0118 investigated the effect of contrasting slurry application timings on drainage water quality at three sites in England over 4 drainage seasons. The project showed that when slurries were applied to soils with moisture deficits of less than 10 mm, and rainfall occurred within 2 weeks of application, drainflow ammonium-N and phosphorous concentrations increased (Figure 1-13and Figure 1-14) and contaminated drainage water was observed (Figure 1-15).27Figure 1-13: Ammonium-N concentrations in drainage water following contrasting slurry application timings to drained clay soils (Defra project WQ0118)Figure 1-14: Total dissolved phosphorus concentrations in drainage water following contrasting slurry application timings to drained clay soils (Defra project WQ0118)0123456701-Oct-0713-Nov-0726-Dec-0707-Feb-0821-Mar-0803-May-0815-Jun-0828-Jul-08DateNH4-N (mg/l)Grass AutumnGrass Early SpringGrass SummerEarly spring (6 Mar)13 mm rain on 10 Mar40 mm rain on 15-16MarEC FFD limit 0.78 mg/l NH4-NGrass - summer slurry (4 Jul)62 mm rain in following 7 days01234501-Oct-0713-Nov-0726-Dec-0707-Feb-0821-Mar-0803-May-0815-Jun-0828-Jul-08DateTDP (mg/l)Grass AutumnGrass Early SpringGrass SummerEarly spring (6 Mar)13 mm rain on 10 Mar40 mm rain on 15-16MarGrass - summer (4 Jul)62 mm rain in following 7 days28Figure 1-15:Drainage water samples 10 days after March slurry application to drained clay soils withc.10mm soil moisture deficit.The project suggested that in order to minimise the risks of diffuse water pollution, over-winter slurry storage capacity should be sufficient to prevent applications to soils when soil moisture deficits were below 20 mm (Table 1-4).Table 1-4: Risk management guidelines for slurry application timing (from Defra project WQ0118)Soil moisture deficit (mm)Risk>20Low10-20Moderate<10HighThe IRRIGUIDE (Bailey and Spackmann, 1996) modelling carried out as part of this study suggests that soil moisture deficits across Wales at the end of March were c.10mm which suggests that slurry applications in March would pose a high risk of ammonium-N and phosphorous contamination of surface waters. As the risks of nitrate leaching losses are greatest following autumn application timings it can be suggested that high risks times for water pollution for slurry and poultry manure applications run from the beginning of October until the end of March. This indicates that in order to minimise the risks of applying slurry at high risk times 6 months storage capacity is required. Information from Natural Resources Wales (Andrew Chambers, Pers Comm) suggest that there were 180 and 160 surface water pollution incidents from agriculture in 2018 and 2019, respectively. Some of these incidents are likely to be caused by failures of slurry management including leaking slurry stores and the application of slurry to soils when there is a high risk of runoff or drainage water contamination which may be a result of insufficient storage capacity. Increasing slurry storage capacity to 6 months is likely to reduce the risk of point source as well as diffuse water pollution.Representation in ModellingThe impacts of this measure for nitrate, ammonia and nitrous oxide were calculated using the MANNER model, which explicitly accounted for the impacts of changing timing 29from a baseline distribution of timing derived from the British Survey of Fertiliser Practice. The MANNER modelling is described in more detail in the Appendix.For phosphorus, Farmscoper assumed a reduction in short term losses from manure of 50%.The costs of this measure are solely associated with additional storage to facilitate improved manure timing, which are calculated separately (see Section 1.4.2).1.2.8Do not spread farmyard manure to fields at high-risk timesDescriptionAvoid spreading (straw-based) FYM to fields at times when there is a high-risk of surface runoff or drain flow, for example, where rain falls shortly after applying FYM to‘wet’ soils i.e. those with a soil moisture deficit of less than 10mm. There is a risk of pollution if solid manures are spread under conditions where heavy rain following application could transport nutrients to surface water systems. The high dry mattercontent and low readily available nutrient content of farmyard manure result in a lower risk of pollution than following applications of slurry. It will not add sufficient water to the soil to initiate surface runoff or preferential flow to field drains; ‘Fresh’ FYM has a higher content of readily available N, and generally presents a greater risk of pollution than ‘old’ FYM that has been stored for several months.Representation in ModellingFarmscoper assumes a reduction in short term losses from manure of 25%.As the modelling that underpins Farmscoper is based upon manure application timing information derived from the British Survey of Fertiliser Practice, current implementation is captured in the modelling and so the implementation of the mitigation measure is set to 0.There are no significant costs associated with this measure.1.3Assumptions Used for Cost and Benefit Estimates1.3.1Variables of InterestSome policy scenarios will increase capital costs to farmers as well as farmers’ time input and operational costs. There are also potential benefits to farmers from reduced manufactured fertiliser costs. The environmental savings for fertiliser nitrogen were estimated as part of savings in operational costs within the environmental modelling. The environmental benefits from increased manure nutrient use efficiency include potential reductions in three types of pollution: (i) Greenhouse Gas (GHG) emissions to air; (ii) ammonia emissions to air; and (iii) nitrate-N and total phosphorus losses to water. 1.3.2Societal Benefits –WaterAn estimated 3.8billion m3of water is used in Wales each year with the majority used for electricity generation and public water supply (Morris, J. & Camino, M., 2011). The value of the water used in Wales each year has been estimated at £57million based on Natural Resources Wales standard unit charges of c. £15/1000m3. In addition, Wales has 11 lowland and 10 upland wetland sites (inland marshes and peat bogs) covering 303,458 ha which provide flood control, recreation andbio-diversity benefits which have been estimated to be worth c. £643/ha per year giving a value of c.£2.2 million/year. Wales is also an important provider of freshwater fishing activities with market value for fishing rights of £90 million. The freshwater fishing industry also supports an estimated 700 jobs (Maule, G. 2018).Water pollution from agriculture affects different stakeholders (Defra, 2014) including:•Water companies must use costly processes to remove agricultural pollutants to produce safe drinking water •Members of the public obtain reduced recreational value from use of watercourses, e.g. angling •Members of the public suffer increased risk of illness when bathing •Members of the public obtain reduced non-use benefits from watercourses due to ecosystem damage from agricultural water pollution and eutrophication of freshwater and marine water •Commercial shellfisheries and fish farms suffer an increased risk of contaminated produce from unclean water and therefore a loss of sales •The tourism sector could suffer losses from beaches that are closed due to failing bathing water standards •Other farmers suffer loss of revenue due to potential health risks if polluted water is abstracted unknowingly and applied to sensitive crops, such as salad. Poor water quality may also prohibit the planting of certain crops The value of economic benefit from reducing agricultural pollution has been reported in a number of studies. Metcalf et al (2012) surveyed households from across England and Wales in order to assign a value to the implementation of measures to meet Water Framework Directive targets for water quality. The study suggested the value placed on improving water quality ranged between £2,263 to £39,168 per km2depending on the population density (areas with higher population densities put greater value on the measures) the location of the improvement and the ecological scope of that improvement.Estimates derived from information reported by O’Gorman and Bann(2008) suggested that in Wales, costs associated with agriculture’s contribution to bathing water failures and the impacts of less than good quality river water were c.£1.5 million/year.Defra (2016) suggest that it is inappropriate to assign single average figures to describe the environmental benefit of reductions in agricultural water pollution due to the geographic and temporal variation in pollutant concentrations. The damage caused by the pollutant will also vary according to the size of the water catchment, the degree to which it is used by humans or supports wildlife and the baseline water quality. It is suggested that a range of values is used to quantify the economic impacts of reductions in nutrient losses to water.There are a range of environmental damage costs reported for nitrate and phosphorus loss to water in different environmental impact assessments. Defra 2016 quote a central value of 33p/kg (range 0-48p/kg) for nitrate and £19.89/kg (range £4.20-£35.06/kg) for phosphorous. In contrast, figures reported in Defra project LM0304) suggest central 31values of 43p/kg (range 24-62p/kg) for nitrate and £12.79 for phosphorus (range £2.77-£22.66/kg). In this project we have chosen to use the figures recently published in Defra’s Enabling Natural Capital Approach (ENCA) Databook which gives central values of 97p per kg (range 69p-£1.26/kg) for nitrate and £30.00 (range 26.66 to 33.34 /kg) for phosphorus. The ENCA methodology sets the standard for studies quantifying the impacts of agricultural practices on Natural Capital.1.3.3Societal Benefits –AirCarbonGHG emissions is measured as the equivalent amount of carbon dioxide (CO2e). Methane (CH4) and Nitrous Oxide (N2O)are converted to CO2e using their respective conversion factors of 25 and 298. The standard unit used is equivalent tonnes (tCO2e).The carbon valuation methodology evolved over time. In December 2007, the approach to carbon valuation adopted the use of the shadow price of carbon (SPC) as the basis for incorporating carbon emissions in cost-benefit analysis and impact assessments. It is based on estimates of the lifetime damage costs associated with greenhouse gas emissions, known as the social cost of carbon (SCC), and it takes more account of uncertainty compared to the SCC approach adopted previously. GHG values are based on the economic cost of mitigating a unit of carbon. The carbon value will vary depending on the sector from which the emissions occur. There are two types of sectors: the traded sector (which is defined as those activities covered by the EU Emissions Trading System (EU ETS) with a market price for carbon) and non-traded sector (which included all other sectors not covered by the EU ETS). Agriculture is included in the non-traded sector. The changes in GHG emissions from the agricultural sector are valued at the non-traded carbon prices published by The Department for Business, Energy & Industrial Strategy (BEIS; Table 1-5).Table 1-5: Non-traded carbon prices for year 2021-2040 (£/CO2et) in 2018 prices.YearLowCentralHigh20213570106202236721072023367310920243774111202538751132026387611420273977116202839791183220294080120203040811212031448813220324896144203352103155203455111166203559118178203663126189203767133200203870141211203974148223204078156234Source: BEIS modelling.AmmoniaVarious valuation methodologies have been used for air quality appraisals, which include: impactpathways approach (IPA), damage cost approach(a set of monetary impact values per tonne of emission), activity costsapproach(monetary value per KWh energy used)and abatement costs approach. Abatement costs approach should be used when the policy/project is expected to push emission concentrations above legal limits. This approach is used to assessthe cost of offsetting measures (the "abatement cost") only for theamount of air quality that breaches therelevant obligation. Activity costs approach is often used in policies associated with fuelconsumption, particularly when change in fuel is known but changes in pollutant emissions are unknown. IPA is the best practice approach but resource intensive. This approachis best suited for project that are more than £50million with the main objective of the policy or project being changes in air quality.Damage cost method is an approach developed by Defra to enable proportionate analysis when assessing relatively small impacts on air quality (NPV <£50m). This approach is deemed to be most appropriate to be used in this appraisal assessing the impact of policy changes in NVZ regulations. Damage costs are a set of impact values, measured as per tonne of emission by pollutant, which are derived using the more detailed IPA in order to estimate the societal costs associated with small changes in pollutant emissions. The damage cost methodology has evolved over the years and currently includes values for impacts on human health, productivity, amenity, environmental health and ecosystem services.33Defra updates and publishes ammonia damage cost prices each year, the most recent published (2020) range of ammonia price (central value) is £7,923 with a range from £1,521 to £24, 476 in 2017 prices. The damage cost value for ammonia in the previous year (2019) was £6,064 (central value) with a range between £1,133 and £18,867. The increase in ammonia value reflects the most recent re-evaluation of damage costs relating to human health and the inclusion of wider ecosystem service impact. The most recentammonia damage cost data (2020) was used in this appraisal. 1.3.4Summary of prices used for valuation of environmental benefitsFor valuation of GHG emission savings, the central cost of carbon for non-traded GHG emissions in the UK is used (£68 per tonne of CO2e in 2019), the full range of the monetary cost estimate is £34-£102 per tonne of CO2e in 2019 (Table 1-6). The central estimate of forecast prices for carbon has been used for each year over the period from 2021-2040. Similarly, the damage cost estimate associated with ammonia emissions fall over a large range. The central estimate used in the analysis is £7,923 per tonne but the full range is £1,521to £24,476 per tonne in 2017 prices. The central value of these estimates has been used to quantify the environmental benefits in terms of reduced ammonia, phosphorus, nitrate-N and GHG emission savings as well as the low and high end values to illustratethe value range of environmental benefits.Table 1-6: Variables impacted on and their monetary valuePollutantCentral Value (£/t)Value Range (£/t)Data sourceGHG £68£34-£102Non-traded CO2values in 2018 prices. Source: Department for Business, Energy and Industrial Strategy. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/793632/data-tables-1-19.xlsx(Table 3)Ammonia £7,923*£1,521-£24,476Defra Air Quality Damage Cost Guidance (2020).National averages in 2017 prices. https://www.gov.uk/government/publications/assess-the-impact-of-air-quality/air-quality-appraisal-damage-cost-guidanceNitrate-N £970£690-£1,260ENCA services databookPhosphorus £30,000£26,660-£33,340ENCA services databook*Ammonia value increased significantly from year 2018 because of a re-evaluation of the damage costs, especially relating to human health and the inclusion of wider ecosystem service costs.
34As the prices for different pollutants were based on different reference year, the prices were then adjusted to the 2018 price base year (which is the latest base year that was used across all pollutant prices from various valuation sources in the table above) using Gross Domestic Product (GDP) deflator for this appraisal.The analysis assumes that there is full compliance with the measures. Should compliance be less than that, then costs and benefits will both be less but the net monetary effect will be in the same direction. 1.4Costs of implementing measuresFor most of the mitigation measures, costs of implementation were taken from the Farmscoper modelling. However, costs of record keeping and for slurry storage were calculated separately as described in the following sub-sections.1.4.1Administrative cost of record keeping and nutrient planningThe completion of records and plans required by the proposed measures is likely to add additional administrative costs to farm business. For some farms the measures will have little or no impact as they may already be keeping records as part of existing land management or farm assurance schemes. The estimates for the administrative costs associated with nutrient management planning are based on a number of assumptions as outline below:•Farmer’s time is costed at £20 per hr3. •74% of dairy farms, 55% of cattle and sheep farms (outside of SDAs) and 46% of cattle and sheep farms (inside of SDAs) already have a soil nutrient plan (Anthony et al., 2016). •40 hrs is the typical time to create a nutrient management plan (Johnson et al., 2012). Because of the large proportion of farms in Wales that are small, this value was scaled by farm size. This value was assumed to represent a farm of 24 to 40 ESU (Economic Size Units; Table 3-5), with the time scaled by average area for the other farms sizes (resulting in 6 hours for the smaller farms and 70 hours for the largest farms). This suggests that the average cost of a farmer produced nutrient management plan was £800/farm (range £130-£1400) which is lower than typical charge of between £1,000 and £2000 for plan produced by a FACTS qualified adviser (Mel Holloway, Pers Comm). •Average annual time is an additional 20 hours from the survey (Johnson et al., 2012). This value was also scaled by farm size which was equivalent to £400/farm which is lower than the £700-£900/farm typically charged by a FACTS qualified adviser (Mel Holloway, pers comm.).3£20/hr is judged to be representing the average cost rate. The hourly rate is ranging from £12.11(farm managers’ time) to £40/hr using a consultant. £12.11 is the average hourly ratefor managers and proprietors in agriculture and horticulture in Wales [source: Office for National Statistics (ONS), 2019. Earnings and hours worked, region by occupation by four-digit SOC: ASHE Table 15.5a -Hourly pay -Gross (£) -For all employee jobs]. According to Nix 2020 pocket guide (p.168), farm management cost at Grade 6 is £15.96/hr.35•71% of manure on dairy farms, 19% on cattle and sheep farms (outside of SDAs) and 11% on cattle and sheep farms (inside of SDAs) is managed as slurry (Anthony et al., 2012a).•Half of the farms with slurry storage would need professional planning to build or expand their facilities. The cost of this would be £3,500 per application (Kenny Dhillon, pers comm).•Current impacts of NVZs have been ignored due to the small proportion of farmers within the existing areas (and as these may have been included in the survey figures used)Costs were calculated by farm type and farm size (Table 1-7) using European Size Units (ESU), and accounting for those farms inside the existing NVZ area. The overall costs for slurry storage and associated costs for applying for planning permissions, as well as planning time for nutrient management plan are the same for Option 2 and Option 4 (Table 1-8). The extra time put into nutrient management planning comesfrom the requirement of implementing the measure ‘using a fertiliser recommendation system' by farm both inside and outside NVZs. The estimated costs for planning include: £4.3m (before discounting) for on-going additional planning on all farms and £4m (before discounting) upfront costs for those farms currently without a plan. There is a further cost of £3.5m (before discounting) in planning fees for the additional slurry storage facilities. For option 3 (Table 1-9), the costs are lower and include: £0.13m (before discounting) for on-going additional planning on all farms and £0.09m (before discounting) upfront costs for those farms currently without aplan. There is a further cost of £0.16m (before discounting) in planning fees for the additional slurry storage facilities.The cost assessments assume that farms that are under 8 ESUs have low levels of nutrient inputs from fertilisers or manures and do not need detailed nutrient management plans.These farms are defined in the NVZ guidance as:•In the calendar year have 80% of the agricultural area of the holding is in grass•The total amount of nitrogen in organic manure applied to the holding, whether directly by animal or as a result of spreading is no more than 100 kg/ha N •The total amount of nitrogen in manufactured fertiliser nitrogen applied to the holding is less than 90 kg/ha N •No organic manures are brought on to the holding36Table 1-7: Number of active farms in Wales by farm type and farm size (defined by Economic Size Units) from 2019 June Agricultural Survey DataFarm Type< 88 -2424 -4040 -100> 100Total Number of Active FarmsCereal128596071102420General cropping4730122216127Horticulture64428403187830Dairy1041,08634341491,614Cattle and Sheep LFA5,3223161,5311,7073,17512,051Cattle and Sheep Lowland1,279562762316582,500Mixed76979801271141,169Pigs2132316225Poultry95579123981,093Other458252281614,778Total14,0431,7402,0702,5784,37624,8071.4.2Slurry storage costsSlurry storage volumes were calculated by integrating total livestock counts for Wales from the 2018 June Agricultural Survey, with livestock properties and management practice data in order to calculate annual average slurry storage requirements by month. Initial excreta volumes by livestock type were taken from NVZ guidance documents. This excreta was apportioned by month between fields, yards and housing. Excreta in housing was apportioned between solid manure and slurry systems according to results of the 2ndWelsh Farm Practice Survey (Anthony et al., 2016), which found over 70% of manure on dairy farms was managed as slurry, but only 10-20% on cattle and sheep farms. There was no solid manure generated on yard areas -excreta deposited was either managed as slurry, dirty water or simply not collected (based on data in 1stWelsh Farm Practice Survey (Anthony et al., 2012a), which found approximately 62% was collected in slurry stores on dairy farms and 20% on cattle and sheep farms). The contribution of rainfall to slurry storage requirement was based on the highest rainfall expected in 5 years (M5) assuming annual rainfall for Wales of 1460mm. An area of yard was specified per animal, by livestock type (0.9 m2per sheep, 4.3 m2for beef cattle and 6.4 m2for dairy cattle; Webb et al., 2001), with a proportion of this arearoofed and guttered. Any rain falling on the un-covered area was assumed to be sent to slurry store, dirty water or uncollected as per fractions mentioned above. T1460 mm. For dairy animals, an additional allowance of 25 litres per day per cow was made for water used in washing the dairy parlour, which was all assumed to be sent to the slurry store. This allowed for a total volume of slurry generated per month to be calculated, and thus storage capacity required to store manure for a specified period. Fromthis, a surface area of the slurry storage could be determined, and this allows for the calculation of additional volume of material to be managed resulting from rain falling into the storage facility (which was assumed to be uncovered). With all calculations undertaken on a monthly basis, the impacts of storing an additional month or two of material can be determined.Understanding the current level of slurry storage capacity in Wales is difficult because of the lack of detailed survey data. Surveys of slurry storage capacity in England and Wales (Smith et al., 2000; 2001; 2001) reported average capacities of 3.5 months for pig slurry; 3.3 months for beef slurry; and 3.8 months for dairy slurry. These values include the effect of some farms reporting no slurry storage. Natural Resources Wales (NRW) have recently (2019) surveyed slurry storage capacity on 230 dairy farmsin Wales (Andrew Chambers, pers. comm.). The milking herd size weighted average storage capacity was a comparable 4.1 months. As the Water Resources (Control of Pollution) (Silage, Slurry and Agriculture Fuel Oil) (Wales) Regulations 2010 (SSAFO) require 4 months excreta production and an allowance for the highest rainfall expected in 5 years (M5) the baseline assumed that farms were complying with SSAFO regulations. Previous studies calculating slurry storage requirements have followed Defra and Welsh guidance at the time which recommended using average annual rainfall to calculate the contribution that rainfall made to slurry storage volumes. Following consultation with Welsh Government the contribution of rainfall to slurry storage 40requirement in this study was calculated using M5 rainfall which is typically c. 10-20% higher than average rainfall. The calculated baseline and additional storage capacity and additional costs required to increase slurry storage capacity to comply with the measures: (i) ‘Increase the capacity of farm slurry stores to improve timing of slurry applications’ –i.e. 5 months storage and (ii) ‘Do not spread slurry or poultry manure at high-risk times’ –i.e. 6 months storage are given in Table 1-10. The costs for above ground stores (i.e. constructed with a concrete base with either steel or concreate walls) has been assumed at £50/m3and the cost of earth-banked lagoon stores has been assumed at £40/m3(Nix, 2019). It is likely that costs will vary between farms according to the configuration of the farm steading, and availability of labour and materials etc.Table 1-10: Capital costs of increasing slurry storage requirements (50% of yard area roofed).AreaSlurry storage volumeAdditional storage requirementAdditional CostsAbove ground tankAdditional costsLagoonMillion m3£ millionBaseline+5 months6 months5 months6 months5months6 months5months6 monthsNVZ0.911.081.270.170.358.5317.926.8214.3292% of Wales5.53 6.517.610.982.0749.02103.739.2182.96All Wales6.457.608.881.152.4357.54121.746.0397.30 + Baseline assume compliance with SSAFOFor the whole of Wales baseline slurry storage capacity estimates were c. 6.5 million m3 comparedwith c.7.6 million for 5 months storage and c.8.9 million for 6 months storage with dairy slurry accounting for c.85%, beef slurry 15% and pig slurry less than 1% of total volumes. The cost of the additional storage was estimated at between 46 million and 57 million for 5 months and £97 million and £122 million for 6 months storage, respectively (Table 1-10).Costs of the additional storage requirement in the NVZ area were estimated at between £6 million and £8million for 5 months and £14 million and £18 million for 6 months storage respectively. For the area outside the proposed NVZ area the costs of additional were estimated at between £39 million and £49 million for 5 months and £83 million and £104 million for 6 months storage respectively. The lower costs associated with additional requirements in these areas compared to the whole of Wales reflect the smaller number of animals and consequently lower slurry volumes.Yard runoff and water running from roofs can make a significant contribution to slurry storage capacity requirements, especiallyin areas of high rainfall. Baseline estimates assume that 50% of dirty yard areas are covered and no allowance is made for water collected from roofed areas. The assumptions are based on evidence from Defra farm practice survey (2006) which states that 40% of concrete yards are uncovered and Aitken et al. (2001) reported that rainfall falling on 65% of yards produced contaminated runoff. 41The estimates that yard runoff water contributes around 20% of total annual slurry volumes collected. Further estimatesof slurry storage capacity and capital costs were carried out with the area of dirty yard roofed increased to 75%. The additional capital costs associated with roofing the yards was estimated based on a cost of £80/m2(Nix, 2019; confirmed by Charles Bentley, Pers Comm.) and the slurry storage costs were adjusted to account for the lower storage requirement. The reduction in slurry spreading costs as a result of the reduced yard runoff component was also quantified.The capital costs of increasing the area of roofed yard from 50% to 75% was estimated at £115 million for the whole of Wales, £15 million for the proposed NVZ area and £100 million for the area outside the proposed NVZ (Table 1-11). The additional roofing reduced the capital cost of an additional 5 months slurry storage by c.£15 million for all Wales, c£14 million for the area outside the proposed NVZ and c.£0.5 million for the NVZ area. Additional roofing reduced the capital cost of an additional 6 months storage by c. £17 million for all Wales, c£15 million for the area outside the proposed NVZ area and c.£2million for the NVZ area. Overall costs of roofing increased capital costs for 6 months storage by c. £97 million for the whole of Wales, £83 million for the area outside the proposed NVZ and c.£14million for the proposed NVZ area. Roofing increased overall capital costs for 5 months storage by c.£100 million for the whole of Wales, £86 million for the area outside the proposed NVZ and £15million in the proposed NVZ area. The additional capital costs were partly offset by savings in annual slurry spreading costs of £135k/ year in the proposed NVZ area, £900k/year in the area outside the proposed NVZ area and £1million/ year across the whole of Wales.Table 1-11:Capital costs for slurry storage capacities and increasing covered yard area to 75% (costs based on average of tin tank and earth banked lagoon)Area5 months capacity6 months capacityRoofStorageTotalRoofStorageTotalNVZ1572215163192% of Wales1003113110077177All Wales115371521151032181.4.3Sensitivity analysis of cost and benefit assessmentsThe range of potential implementation and damage costs was accounted for with a sensitivity analysis. For the following measures with the most significant costs and greatest uncertainty high, medium and low cost estimates were produced:•Do not spread slurry or poultry manure at high-risk times,•Integrate fertiliser and manure nutrient supply•Do not apply manufactured fertiliser to high-risk areas•Avoid spreading manufactured fertiliser to fields at high-risk timesThe review of damage costs also produced a central estimate and upper and lower bounds for each pollutant. The sensitivity analysis thus considered the consequences of using the high, medium or low implementation costs, and the high, medium and low damage costs. 42For the uncertainty analysis, the high, medium and low costs for do not spread slurry and poultry manure at high risk times were represented by:•High: Increasing the covered dirty yard area from 50% to 75% and rebuild 50% of slurry stores to hold 6 months slurry production•Medium: Increasing the covered dirty yard area from 50% to 75% and extend slurry storage capacity from 4 to 6 months•Low: Extend slurry storage capacity from 4 to 6 months.The high medium and low costs for increase slurry storage were represented by:•High: Increasing the covered dirty yard area from 50% to 75% and rebuild 50% of slurry stores to hold 5 months slurry production•Medium: Increasing the covered dirty yard area from 50% to 75% and extend slurry storage capacity from 4 to 5 months•Low: Extend slurry storage capacity from 4 to 5 months.For Integrate fertiliser and manure nutrient supply:•High: Only the crop available N in manure is accounted for•Medium: The crop available N in manure is accounted for, and 30% of the available P and K•Low: All of the available N, P and K is accounted forFor Do not apply manufactured fertiliser to high-risk areas:•High: areas occupy 10% of fields•Medium: areas occupy 5% of fields•Low: areas occupy 2% of fieldsFor Avoid spreading manufactured fertiliser to fields at high-risk times:•High: 10% yield loss occurs 1 year in 5•Medium: 10% yield loss occurs 1 year in 10•Low: 10% yield loss occurs 1 year in 151.5Time horizon and discounting rateThe costs and benefits of the policy scenarios are assessed over a 20-year period (which is assumed as the lifetime of slurry stores) from year 2021 to year 2040. The non-amortised value of capital costs was used inthe NPV calculations, assuming zero residual value at the end of the 20-year policy period. A discounting rate of 3.5% was used in this impact assessment in line with the HMT’s Green Book4guidance to estimate the Net Present Value (NPV) of costs and benefits of different policy scenarios. The initial year is 2021.