1. Abstract

Overview

EEA’s report “Water and agriculture: towards sustainable solutions” (EEA 2020a) provides a thematic assessment of agricultural practices in Europe and their implications for water resources. Central to this assessment was the description of four key pressures on water originating from agricultural activities: Pollution from nutrients; pollution from pesticides; water abstraction for irrigation; agricultural land use in the floodplain.

This paper describes the data, methodology and results of an analysis that shows variation of farming-related pressures exerted on freshwaters depending on agricultural landscape types. Adopting the present-day farming landscape typology of Levers et al. (2018) (who delineated different types of agricultural land use intensities), it derives 15 broad European agricultural regions (BEAR), specifies their pressure-profiles, and derives the composite multi-pressure index. The index includes the four aforementioned pressures, using datasets available with European coverage aggregated at the level of more than 30 000 river catchment units. Further analysis on linking agricultural production to the pressure index is given in the annex.

Key results

In total, 15 different BEARs were defined, which show different combinations and intensities of pressures on water from agriculture. The most abundant BEARs were ‘Extensive grassland and fallow farmland‘ (covering 25 % of catchments) and ‘Western intensive cropland‘ (covering 17 % of catchments).

The most intense composite multi-pressure indices were identified for the ‘Mediterranean intensive cropland’ located in parts of Spain, Italy and Greece (Figure 1). In this BEAR, most pressures, including pollution from pesticides, water abstraction for irrigation and agricultural land use in the floodplain are particularly high, whereas nitrogen surplus is lower. The regions with the lowest composite multi-pressure indices were ‘Northern and Highland livestock farming’ and ‘Extensive grassland area and fallow farmland’.

 Figure 1: Composite agricultural pressure index classifying the average intensity of multiple pressures from agriculture on water bodies in a catchment

Consultation questions

For this consultation we would like to ask you to comment on the approach and results:

  1. Do you think it is useful to establish a composite multi-pressure index of agriculture on water at pan-European scale?
  2. Do you think the methodology used is appropriate for this type of analysis (regarding BEAR delineation and definition of the composite multi-pressure index)?
  3. Can you agree with the results of the multiple pressure analysis in your country (in general, and in detail regarding specific areas)?
  4. How could such an analysis be subject to regular updates, and at which frequency?
  5. What further information could be included to support or improve the analysis?
  • troddway (Wayne Trodd) 31 Aug 2021 11:43:01

    Comments on Report from Environmental Protection Agency and Dept. Agriculture, Food and Marine

    Introduction

    The European Environment Agency (EEA) has developed a composite index of the pressures arising from agriculture on water. The assessment is based on datasets that are readily available for all member states. The output is a series of EU-wide maps of the impacts of four key pressures from agriculture on water: nutrients, pesticides, abstraction for irrigation and landuse in the flood plain. An additional multiple pressure map has been developed that is a composite of all four pressures. The EEA has invited feedback and comments from the National Reference Centre (NRC) representatives on water and agriculture on the approach, and in particular, on the usefulness of the composite index assessment. The following comments on the applicability of the approach and the outputs in the Irish context are provided from representatives of the Environmental Protection Agency and the Department of Agriculture, Food and Marine.

    Comments on the report

    • Ireland’s main environmental pressure from agriculture, by far, is nutrients. The composite pressure index masks the level of effort needed to address the nutrient issue, and therefore does not really add value to the adoption of single pressure indices.
    • Figure 2 page 6 indicates incorrectly that almost the whole of Ireland has intensive livestock. Ireland has 135,000 farmers and the majority of these would have extensive livestock holdings, especially in the western half of the country. In Ireland, 27% of basic payment applicants, who are located on about 1/3 of land, have 70% of bovine livestock, based on 2019 data. A very small area is arable, so about 60% of the country is extensively farmed.
    • Figure 3 showing the distribution of Ireland's nitrogen surplus is reasonably accurate and reflects more appropriately the distribution of intensive farming in Ireland (in contrast to the inaccurate distribution of intensive farming in Figure 2).
    • Ireland's livestock industry is almost completely grass based with animals only indoors over winter. The model seems to assume that grasslands are low intensity which is not the situation over all parts of in Ireland.
    • The overall pesticide risk is reasonable although it is overestimated in places. Only 7% of Ireland's agricultural area is arable, and most of that is in the south and south east. Ireland's most widespread pesticide risk is from herbicides for grassland weeds.
    • Ireland does not have a significant irrigation industry. Irrigation is only used for vegetables which covers approximately 1.5% of the agricultural area. That is perhaps why there are no data. Our recommendation would be that Ireland should be in the lowest category.
    • Ireland does not have major river floodplains such as those present in Europe – our flood plains are typically small as 75% of our river network is stream order 1 or 2. A clear description of the methodology used to generate Figure 6 would be helpful so that it can be assessed why so much of our landscape is located in the highest top two categories. Hydromorphological issues are the second most important pressure from agriculture after nutrients, but the locations of the affected areas are more widely distributed and localised that those depicted in Figure 6.
    • The report appears to be based on 2006 data? If our understanding is correct, data from the period before the abolition of the dairy quota will show a very inaccurate position for the current impacts from agriculture in Ireland.
    • The purpose of the outputs of this report are not clear. Further refinement of the assessment outputs and conclusions would be necessary if it is intended that it should be used in decision making.

     Responses to the specific questions asked

    1. Do you think it is useful to establish a composite multi-pressure index of agriculture on water at pan-European scale?

    The individual pressure indexes are likely to be most useful for member states, and for taking the conversation forward to measures. It is difficult to see the practical value of the composite map.

    1. Do you think the methodology used is appropriate for this type of analysis (regarding broad European agricultural regions delineation and definition of the composite multi-pressure index)?

    The precise methodology is difficult to understand without reading all the reference material. The document would benefit from a comprehensive, yet easy to read summary of the methods.

    1. Can you agree with the results of the multiple pressure analysis in your country (in general, and in detail regarding specific areas)?

    The methodology has not captured the reality that well in Ireland. Perhaps this is because our main agricultural pressure is from grass fed livestock, which results in excess nitrogen from high intensity farming in the south and south east, and excess phosphorus from lower intensity farming elsewhere.

    1. How could such an analysis be subject to regular updates, and at which frequency?

    Updates to these maps would be useful following each river basin management plan reporting period.

    1. What further information could be included to support or improve the analysis?

    Suggest the team download the WFD waterbody level impact assessment data from the WISE reporting to validate the assumptions and modelled outputs.

  • farrereg (Regis Farret) 14 Sep 2021 18:02:00

    Question 1:  Is the index useful? Not really, because a composite index is not easy to link to policies. Also, it does not seem so robust for the methodology (see remarks below), some underlying indicators seemquestionalble (esp. hydromorphology) and there is a mix between pressures & impacts.

    Question 5 : better identify groundwater among the indicators, in particular for withdrawals (suggestion: in withdrawals, identify the shares taken from surface water and groundwater)

    Other questions and technical comments:

    Nitrogen : w do not find the same results for France, particularly South-West and Brittany

    Pesticides : il is an impact indicator and not pressure indicator  

    Withdrawals: The map does not seem entirely accurate to us, even if the overall order of magnitude is good. According to our figures, the withdrawals per hectare are more important in the South, the South-West and the North-East, and do not stand out in the Paris basin and the North of the Loire (note that the national average abstraction figure for France, present in the global report on page 44, corresponds well to our figures, which would indicate that the underestimation may come from the data on the irrigated agricultural surface and its distribution, rather than from the abstraction data). In order to make a more detailed comparison, it would be necessary to have the modelled abstraction data and the irrigated agricultural area data used. It should be noted that the data concerning ""landscape classifications"" represent the conditions in 2006 (p. 5), which may explain certain discrepancies with other data.

    Hydromorphology : Please clarify the objectives of this indicator. It seems ambitious and gathers many themes that are very difficult to combine. Additionally, the proxy that is eventually chosen is not convincing.

    The results of the multiple pressure analysis do not match with our results for the Landes de Gascogne area (SW of the France). This area is delineated as "forest" and without agricultural pressure, whereas we identified agricultural pressure due to irrigated maiz crops (important withdrawals, use of fertilizers and pesticides).


  • silvaagu (Agueda Silva) 14 Sep 2021 18:50:07
    1. Do you think it is useful to establish a composite multi-pressure index of agriculture on water at pan-European scale?

    The application of this index at European level is interesting as it gives an idea of the agricultural pressure on water resources in terms of quality and quantity. It very useful to establish a composite multi-pressure index where the analysis is differentiated with the separate analysis of the values and the spatial distribution of the individual pressures included in the calculation of this Index, to better understand the main contributions. Thus, it must be associated with monitoring the specific indicators that give rise to it, with a view to keeping information available that clarifies the main pressure factors. Furthermore, agriculture is a key driver of pressures that lead to less than good water bodies status.

    1. Do you think the methodology used is appropriate for this type of analysis (regarding BEAR delineation and definition of the composite multi-pressure index)?

    It is important to clarify the eventual reason that justifies the absence of the parameter phosphorus in the calculation of the Index (“Diffuse pressures: nutrients”), since this parameter is penalizing and responsible for the eutrophication of water bodies.

    The methodology should be a little more detailed in terms of plant protection products, as it is not clear which pesticides are used in the methodology, the study only indicates the number of 332.

    Regarding water consumption for agriculture, it is also not understood how the figures were calculated.

    It is important to better understand how the various classes and thresholds were established in Table 1: Indicators and pressure levels classified into four pressure intensity classes. In terms of pesticides, the unit of analysis and allocation is not well understood.

    However, it is necessary to take into account the data provided by each country and the fact that under certain circumstances the organizations that compile them carry out statistical calculations (e.g. fill the gaps) that may alter the original values provided by the different countries.

    1. Can you agree with the results of the multiple pressure analysis in your country (in general, and in detail regarding specific areas)?

    However, we return to the aspect already mentioned in question 1: keeping available the information on individual pressures associated with the respective composite index, in order not to lose relevant data of support the planning of political actions (original pressures in each BEAR).

    Illustrating with examples for the Mediterranean (table below - extract information from table 2 of the study “Compound pressures in agricultural water”), we can observe different levels of “pressure” aggregated by similar “pressure indices”, which recommend that the information source of pressures (BEAR dimension and local dimension) and their multiple effect at the territorial level (BEAR dimension) are, at the same time, the basis of information for the analysis of priorities and subsequent policy actions.

    BEAR

    Pressure index

    Nitrogen pressure

    Pesticide pressure

    Agricultural land use in flood plain

    Water abstraction

    Total crop yield

    [t km-2 ]

    #FECs

    %FECs

    Area

     [ha]

    Mediterranean intensive cropland

    3.5

    low

    high

    high

    high

    60.5

    1,267

    4.1

    8,631,175

    Mediterranean large-scale permanent cropland

    3.0

    very low

    high

    medium

    high

    51.9

    1,958

    6.4

    10,299,484

    Mediterranean extensive cropland

    2.8

    very low

    high

    medium

    medium

    50.8

    1,463

    4.8

    7,176,737

    Mediterranean livestock farming

    2.3

    low

    -

    low

    medium

    -

    244

    0.8

    646,847

    Extensive grassland area and fallow farmland

    1.0

    very low

    very low

    very low

    -

    0.2

    7,689

    25.1

    33,737,274

    1. How could such an analysis be subject to regular updates, and at which frequency?

    In most situations, the basic data will be updated by Portugal in accordance with the provisions of the Water Framework Directive, that is, in the context of the preparation of each planning cycle of the River Basin District Management Plans, usually carried out every six years.

    1. What further information could be included to support or improve the analysis?

    It would be important to include phosphorus in the nutrient analysis as mentioned above.

    It would also be interesting to include in this analysis, and more clearly, another layer of information: the spatial distribution of water bodies, distinguishing them into surface and groundwater.

  • lycheann (Anne Lyche Solheim) 14 Sep 2021 21:05:04

    Brief summary of comments from NIVA as NRC for Norway:

    Good approach to show the total pressure from agriculture as a sector affecting Europe's waters. But the single pressures are perhaps more important to show, as that can be linked to status of waterbodies and to measures against each of the particular pressures.

    The major weaknesses is the absence of phosphorus, which is more important than nitrogen for eutrophication of lakes and equally important as nitrogen for rivers. Total phosphorus is available in WISE-6 and phosphorus conditions from WFD database. Both total phosphorus and phosphate can originate from agriculture, the total P from soil erosion and artificial fertilisers and phosphate from manure and dairy production. 

    The problem with siltation/sediment transport from agricultural areas after heavy rains or longlasting rain should be included, because this pressure causes poor water transparency and may have negative impacts on oxygen levels and on the whole food web, see details in the attached file. Data source for water transparency can be sought in WISE-6 and the the WFD database as well as from remote sensing. 

    Another potential weakness is the method used for assessing the impacts of pesticides, which must be very close to a worst case scenario, and most likely far away from the reality in situ. We miss a comparison with monitoring data, and suggest that such a comparison should be done before using this pesticide index. Our experience is that pesticides have mainly been demonstrated to have an impact on benthic algae and benthic fauna in small streams, but not in larger rivers, nor in lakes. The impression given by the current index seems largely exaggerated/overestimated. 

    For Norway, please see further comments in the attached file. 

    Download attached file
  • scheidand (Andreas Scheidleder) 15 Sep 2021 13:46:49

    Answers from Austria to the 5 questions:

    1. Do you think it is useful to establish a composite multi-pressure index of agriculture on water at pan-European scale?

    In general it is useful.

    Pros

    • It cobines quality and quantity pressures on waters
    • It provides a general picture with different hot spots all over Europe
    • It supports especially EU countries with restricted knowledge and data availability
    • It provides a basis for an integrated evaluation of measures and for further integrative programs of measures (no regret measures)
    • Individual views on single pressures can be misinterpreted by underestimating the negative effects of agriculture and the potential of adequate measures. Perhaps this combined view could avoid this problem.

    Cons

    • It can lead to an oversimplified picture
    • The solely agriculture oriented indicator is rather problem oriented than solution oriented. Not sure if this helps in water management. 
    • Used data may not be the best and most actual data available in the countries

     

    2. Do you think the methodology used is appropriate for this type of analysis (regarding broad European agricultural regions delineation and definition of the composite multi-pressure index)?

    Yes. It is based on one data set, which should make the results comparable. Nevertheless, the quality of data might not be the best the MS could provide.

    For the landscape classification some parameters are incorporated more than once. This may lead to an over or under estimation of the indicators.

    Data from 2006 may not represent the current status

    Nitrogen balances should be calculated for more than one year (e.g. 4 years average, cf. CAP indicator). Methodology used should be described to make it comparable for the MS with the national EUROSTAT/OECD nitrogen balance and with regional balances by the MS. Referring to table 1 (“Pressure indicators and levels ranked into four classes of pressure intensity”) it should be clarified, which nitrogen balance results are ranked (which method, inputs and outputs with which values – e.g. for legume nitrogen fixation a great bandwidth of literature values exists -, gross or net nitrogen balace, surface of farmgate balance) to evaluate the limit values of the different classes.

    FECs without dominant archetype not analyses: combination of all archetypes may have more effects on water

    3. Can you agree with the results of the multiple pressure analysis in your country (in general, and in detail regarding specific areas)?

    The results and the distribution seems to be reasonable.

    The information in Figure 5 is not optimal for Austria. Irrigation is only responsible for a very minor (4%) share of water abstraction and located in very small parts of the country. But Austria falls, inspite of the small-scale resolution, ino category >200-2000 m³/ha/year water abstraction for agricultural irrigation. This is not a big number, however it is not representative for the whole territory.

    4. How could such an analysis be subject to regular updates, and at which frequency?

    Having in mind the scale and the uncertainty of input data a regular update would be reasonable once in six years. Some aspects under evaluation (e.g. the share of floodplains under agricultural use) will not show very dynamic changes…here even a period of ten or twelve years could be sufficient.

    An update every six years would also contribute to the evaluation of measures within the RBMP-Circle

    If such an indicator is established, it should be based on already available data or models and no new reporting obligation should be initiated.

    5. What further information could be included to support or improve the analysis?

    Pressure indicators: Water abstraction for irrigation: is the indicator evaluated with respect to the climatic conditions (precipitation, evapotranspiration)?

    The pressure of soil erosion on surface waters (P-input) is missing (by use of models).

    The vulnerability of other existing ecosystems, e.g.:

    • Changes in biodiversity over years, which is highly reduced due to nutrients and pesticides.
    • The decline of moors due to nutrient inputs and drainage could also be an important factor, as these have a disproportionate effect on climate change.

    Maybe also the contribution to fine dust pollution by N emissions could be estimated.

  • vossenan (Nanna Granlie Vossen) 22 Sep 2021 13:29:52

    Comments on report from Danish EPA:

    Page 32:

    "Nitrogen fertiliser consumption per hectare of fertilised utilised agricultural area (UAA) is over 100kg/ha in Denmark"

    Comment: 

    Clarification on whether nitrogen fertiliser consumption is measured as a combination of commercial fertiliser and manure is needed.

    Our data shows that consumption in 2018 is 217,800 tons N of manure and 223,500 tons N of commercial fertiliser, totalling at 451,300 tons N on 2,632,453 ha. This shows an average consumption of 170kg N / ha. 

    Source: 

    DCE rapport nr. 352, 2019 ”Landovervågningsoplande 2018” (”LOOP-rapporten) https://dce2.au.dk/pub/SR352.pdf

     

    Danmarks statistik: https://www.dst.dk/da/Statistik/emner/erhvervslivets-sektorer/landbrug-gartneri-og-skovbrug/det-dyrkede-areal

    Page 32:

    "Since 2008, the consumption of phosphorus has increased by more than 100 % in Denmark"

    Comment:

    The Danish EPA does not agree with this statement. Our data shows that in 2008, 23.1 kg P was applied and 20.2 kg P was utilised, resulting in a balance of 4.2 kg P. In 2018, approximately 25 kg P was applied, but as precipitation was markedly low in 2018, on 15.3 kg P was utilised, resulting in a balance of 10 kg P.

    Source:

    Table 9.1, pg. 96

    DCE rapport nr. 352, 2019 ”Landovervågningsoplande 2018” (”LOOP-rapporten) https://dce2.au.dk/pub/SR352.pdf

    Page 32:

    "Countries with high livestock densities such as Malta, the Netherlands, Belgium, Denmark, Cyprus and Ireland, also show the highest rates of manure inputs in relation to their agricultural area (over 98 kg N/ha per year)"

    Comment:

    Our data shows that in 2018, approximately 86 kg N/ha of manure in 2018 was utilised in Denmark.

    Source:

    Appendix 1 in DCE rapport nr. 352, 2019 ”Landovervågningsoplande 2018” (”LOOP-rapporten) https://dce2.au.dk/pub/SR352.pdf

    Page 46:

    "According to Eurostat, between 2005 and 2016, irrigated areas increased in several countries, such as the Netherlands, Belgium and most Baltic countries, while a decrease was observed in Denmark, France and the United Kingdom (ESTAT, 2019b)."

    Comment:

    The EUROSTAT data referenced shows data for years 2005, 2007, 2010 and 2013. This data shows that total irrigated area in Denmark has remained relatively stable from 2005-2013: 448,950ha, 435,350ha, 480,440ha, 438,980ha.

     The Danish EPA thus finds this statement misleading.

    Answers to questions:

     

    1. Do you think it is useful to establish a composite multi-pressure index of agriculture on water at pan-European scale?

    The individual pressure indices are likely to be most useful for member states, particularly in terms of planning and implementation of mitigation measures. It is difficult to see the practical value of the composite map.

     

    1. Do you think the methodology used is appropriate for this type of analysis (regarding broad European agricultural regions delineation and definition of the composite multi-pressure index)?

     The precise methodology is difficult to understand without reading all the reference material. The document would benefit from a comprehensive, yet easy-to-read summary of the methods.

     

    1. Can you agree with the results of the multiple pressure analysis in your country (in general, and in detail regarding specific areas)?

     The Danish EPA finds that the figures presented in the report do not give a true picture of pressures from agriculture on water. For example, figure 3 shows that the field surplus is >100 kgN/ha in Denmark. However, our data shows that the field surplus is about 65 kg N/ha. 

    1. How could such an analysis be subject to regular updates, and at which frequency?

     Not applicable, as the Danish EPA finds that the composite index has little value.

     

    1. What further information could be included to support or improve the analysis?

     To improve the analysis, the Danish EPA suggests validating assumptions and modelled outputs by referring to Water Framework Directive waterbody level impact assessment data from WISE reporting.

     Furthermore, the Danish EPA suggests the national monitoring data “NOVANA” be applied.

submit comment