LPISQA Legacy/2012 Workshop Malta

From Wikicap - European Commission

Observations from the European Commission services

Wrap-up of issues by the JRC

LPIS QA seems to work as self-assessment and formally confirms what MS already knew about their system

During the workshop Member States expressed:

  • the desire for a clear and stable ETS 2012 methodology;
  1. without changes late in the assessment period,
  2. with more clarifications and illustrations.
  • a request to review of the LPIS update data flow,
  1. clarifying the 3 year refresh cycle
  2. including clarifications and accommodating boundary stability (provisionally phrased as boundary delineation margin by NL) based on survey and measurement limitations.
  • a call for clear and stable requirements for the post-2014 LPIS,
  1. esp. on greening (grasslands delineation, EFA elements esp. landscape features)
  2. including a strategy on the required changes to the reference parcel design (note: need to update LCM/ATS)
  • the perception that the LPIS QA should more accommodate for the specificities linked to the different reference parcel design options, in particular
  1. measurability issues for cadastral parcel designs
  2. multi-parcel defects for physical and topographical block designs

Comments by DGAgri J3

  • the MS concerns about results and their interpretation are noted
  • scoreboard and assessment will considered together with the scoreboard, still the EC wants also a tool for comparison between MS

Comments by DGAgri D1

  • the concerns of the Member States on the implementation of the 2014 reform are noted, the EC is aware of the time and resource limitations and the challenges for implementing all changes.
  • the EC will provide more guidance on the expectations for LPIS update and refresh (as promised in the 2010 LPIS workshop)

Issues mentioned during the session as recorded by JRC

MS ETS v5.2 methodology interpreting the LPIS scores and results post-2014 CAP consequences for LPIS Other
DK the 4 eye control seems cumbersome There are a lot of registrations requiered Hard to implement - new elements to measure and the 2013 resources are not yet budgeted LPIS QA is self-assessment tool - J3 should be aware that remedial actions need time to be implemented
HU provide open source tools for reading results, multi-parcel issue is clarified, not number of AP but units of management, not measurable block - specific issues for physical blocks, 4-eye control and strict QC is already implemented in HU, many ETS conforming blocks are not so for the internal quality control. foresees a big problem with what is permanent pasture (how to control 5-years back, start in 2004? , what images to use; the old orthophoto; what about small parcels?
EE no particular statements - ETS is improved. It would be nice to stay stable henceforth about quality assessment result - we would like to have faster feeback from EC on the performance, before the beginning of the following assessement a/ Art 31a - problems for orthophoto planning to ensure 3--years cycle update. b/ EFA-implementation: unique landscapes require not general rules but flexible ones; 40% of EE is forest; these lands and wetlands can be also considered as EFA
FI glad with changes in ETS, some issues on digitizing small area. Need for a fixed version. results - good tool for self-assessment; good to benchmark and see what other countries are doing worries about time very short time for implementation before end 2013. why not JRC to make one LPIS golden standard
LT 5 - meter buffer and area to zero - problems; many RPs affected, if unclear all areas to zero - do we need to make still field inspection we don't know if we report correctly critical defect more EC clarification needed; many thing left to MS to decide; we need notes from EC. permanent pasture is important to know what does it mean? clarification more quickly. how to explain new rules to farmers?
NL some clarification on measurability and inspection; more examples; more general level illustrations; less complexity stepping stone for improving LPIS quality; not to focus on scoreboard, but on remedial action plans, if not PA will perform it bias toward political way, but DG agri understood this what type of RP is best for CAP 2014? If EC and JRC cannot give guidance, everybody is on their own? If MS decide, will J3 disallow later? a/ NL proposes a boundary delineation margin for purpose of stability of LPIS and to account for the uncertainty of the interpretation; b/ clarification of min mapping scale
SE no major changes unless for more simplification good and useful tool to perform self-assessment; Results should not be against MS; J3 should look in the remedial action plan please EC, give us clear rules! a/ we want LPIS stability: supporting the NL proposal for boundary delineation margins if no ineligible feature inside the margin - then should be OK. b/ what about simplification?
IE multi-parcels is an issue in IE - topographic block (many field divided by hedgerows). Revise the guidance on digitizing area <0.1 ha: in separate layers,+ not reported. reporting XML files is complex if all LF has to be digitized -will take too long to complete! a/ provide templates for ArcMap b/
UK problems with 5-meter buffer - problems. presence of non-eligible features inside the buffer for TB should be OK. while there are weaknesses in mapping - farmers are aware of these and lodge correct applications big concern for the incoming complexity; lots of LFs; we can only see canopy to make better judgment; a/ supports NL proposal for boundary margin, b/ supports DK suggestion to categorize the Q&A in wikiCAP
SI a/ welcomes to change the QE 7 and integrate OTSC and LPIS QA random sample. linear object should be extracted? - is it part of the 0.1 ha rule? area boundary margin is good multi-polygon is concern still - we don't know whether it is right or not ? clarification of multi-polygon Grassland should be in scope of LF counting for EFA as do LF that border to agri land without direct contact (state land restriction), Else the 7% requirement will only cause more fallow land supports the NL proposal for boundary margin
LU lots have been done, only problem with field measurements - too much work When will we have the 2011 screening reports? nice to have Simplification is missing. For the 2014 deadline, there is no always budget and staff to implement and make the IT changes
IT pro-rata - lack of clarity how to implement causes erroneous processing Will cross-compliance be related to non-agriculture features / water bodies? 30/6/2012 deadline is not achievable; problems for PAs (they are in the middle of the OTSC). ETS should be performed later in campaign
DE ETS should be stable, clear as there are lot of problems in changing. farmer update - changes from other sources in tempore non suspecto - more clarification. With a simpified assessment report template QE6 - EC should clarify; make paper? a/ simplification is required; b/ streamline definitions. c/ clear guidance on EFA is needed
BE-FL critical defects were problems in 2011; after reading 2012 - problem solved; still some reviewing of critical defects and examples needed (like multi-polygon). boundary margin J3 explained how the result will be interpreted there is no simplification; LF will be quite a burden; not necessary to digitize them. a/ art 31a needs clarifications how it relates to ETS resultes; 3 b/ year update is not a good way; ETS is a better way
BE-WA EP- add the reason for the user-defined legend codes + it would be better to impose xml format + why entries that not exist? VHR ortho- can we be used only partly for the sample. Tutorial in wikiCAP on XML parsing what is the proof of a stable physical boundary for incomplete bock? How to split the block to prevent multi-parcel?
CZ Need for training and manual for photointerpretation; quick identification of RP at risks (automated) non-compliance criteria too strict: delineation from scratch implies virtual changes and causes deviation from human factor and image quality. A study found that 35% of non-compliant are due to such error LF - tension between the original intention of LF for biodiversity and the eligibility. problems - if limited to certain sizes, cannot always be restricted to the size or set it up in advance
ES ETS should accomodate the CP. QE7 is not representative as current. Why field observations for skipped parcels? please clarify boundarly delineation is problematic for CP - ETS result may not be representative for the quality of the Spanish LPIS worries about the impact of green payments, EFA and LF on LPIS. It should be adapted - huge economic impact. Deadline (pre-selection 30/6/2012) is very close
MT MS not using GNSS lack the means to vindicate critical defect.. GNSS measurement is not common. How to delineate unclear boundaries; boundary adjustment; aggregation of small parcels on an image? QE 7 is not very representative. Very few number of matched samples. why to put a threshold on LF: is it serving the environment purpose? habitats? a/ Will we have the screening reports before new campaign? if screening feedback arrive later in the year and there are problems - what to do? b/ provide support for GML and XML
RO mutli-parcel - the current 10 subunit criterion is not suitable for RO; there are many such parcels in RO worries about the workload to digitize permanent pasture and LFs
LV a/ 2010 was hard with GMLs - now no problems. b/ how to treat mutli-parcels is not always clear there should be no need for EFA for farmers having only permanent grassland - answer of D1 to IE supports EE and LT about clarity and complexity. do we need to update every RP with the new data; or risk analysis?

Formal positions by the MS sent to the JRC by email


We hope these observations are taken in a constructive spirit and we will contribute to fine tune an already proven and efficient tool for LPIS assessment.

LPIS QA methodology

Member States who are lacking GNSS infrastructure/equipment, presently have no tool to vindicate parcel boundary critical defect “invalid rp boundary or unmeasureable by capi”. Can JRC suggest any alternative to such method? Are the concerned MS obliged to setup such infrastructure to comply for QE3 of ETS? Does article 6(d)of 1122/2009 makes the use of such technologies obligatory? Answer from JRC or the Commission is welcome in order to build a business case in the event such infrastructure, equipment procurement and training services are needed for such a setup.

LPIS QA results

My comment on QE7 questioning that the sampling method may be based on a very restricted sample according to the current methodology, is in accordance with your mention that it is currently being considered, that the first two hundred parcels for ETS inspection would be selected from the on the spot check parcels. Would the two hundred parcels be inspected using GNSS equipment?

LPIS 2014

We do not fully support the idea that a threshold is made on the protection of landscape features. During two presentations at the workshop, it was demonstrated that both Malta and Germany have adopted such thresholds which result in confusion to the farmer. If a landscape feature is over the threshold by a few metres, this LF will no longer need protection according to this threshold and hence becomes non eligible for payment and is not protected (the farmer can remove it without sanctions applied). The Commission is requested to give some guidance on such methodologies especially in the context that in 2014, payment entitlements will have to be allocated on landscape features as these will form part of the eligible area and an important contribution to the greening element of the new CAP.

United Kingdom (for England, Wales, Scotland and Northern Ireland)

The Comments made from Northern Ireland, representing the whole of the UK were:

LPIS QA Methodology

  • We had many RP with an unclear LUI and an ineligible feature within the 5m buffer. These could therefore not be measured.
  • However, the ineligible feature was not within the unclear portion of the boundary and are therefore considered to represent a low risk to the fund.
  • Therefore we feel that for topographical or physical block systems in particular, the LUI should only be classed as unclear if the ineligible feature is within 5m of the unclear portion of the RP.

LPIS QA – Results and Interpretation

  • Northern Ireland is in the middle of a LPIS refresh and therefore did not perform as well as we would have liked in some of the LPIS QA Quality Elements.
  • However if we compared the observed eligible area found during the QA to the area that was used by the farmer to activate a claim, we would have performed well.
  • This means that our comprehensive media/training campaign has ensured that framers recognise the weaknesses within the maps and only claim for the correct eligible area.
  • We feel that some recognition should be given to this in the QA.

LPIS 2014

  • We have great concerns about having to map landscape features and feel that the effort will be disproportionate to any benefit received.
  • Mapping landscape features for us will be complex. We have great lengths of hedges in particular. These hedges can be shared between farmers and are not always symmetrical meaning that it is difficult to know which farmer owns which part of the hedge.
  • Also we can only see the top of the hedge canopy on the ortho image which is often many times wider that the base of the hedge.

Additional Statements

  1. We support the call from the Netherlands for an agreed on-screen tolerance, based on pixel size, when mapping boundaries from an ortho. Northern Ireland is in the middle of a LPIS refresh and has spent considerable time agreeing tolerances with our contractor. A standard figure would have streamlined this process and helped to ensure that we were compliant.
  2. We support the request by Denmark to group Member States Q&A on Wikicap according to subject matter as this will make it much easier to find relevant information.

Czech Republic

LPIS QA methodology

The requirement for independent delineation of LUI during LPIS QA leads to many cases when the parcel is classified as NON-COMPLIANT despite the fact that there is no change in the countryside and the RP boundary has been correctly photo interpreted. This is mainly caused by differences in data sources used and a human factor (this issue was also mentioned in the presentation by Jaap Kroon). We made an analysis on non-compliant RP’s and about 40% of them are falling to this category due to these reasons.

LPIS QA results and interpretation

Follow-up analysis to investigate the reasons for non-conformances found during the LPIS QA 2011 was implemented in the Czech Republic in the beginning of this year. In approx. 40% of cases no error was found, the non-conformity was caused just by small differences in photo interpretation. Remaining 60% represent real errors or problems. A set of specific measures to enhance the LPIS quality has been prepared (e.g. develop detailed photo interpretation manual and arrange for specific training of staff, identify and check RP in risk of non-conformity).

LPIS 2014

It seems from the presentations relating to the implementation of Landscape features (Malta, Germany) that there is certain tension between the environmental aspect (need to protect landscape biodiversity, protect soil from erosion, etc.) and the “eligibility” aspect (need to set up rules for the registration of landscape features in LPIS – size or percentage limitations). This issue should be addressed at the EU level in the near future.

The Netherlands

LPIS QA methodology

We still need more clarification on the measurability of the LUI. What are the (exact) criteria for measurability? More examples are needed, maybe specified for the different types of RP's.
More in general: Do not make the ETS more complex and detailed as it is now. I am afraid that MS will not be able to follow all intentions and instructions.

LPIS QA results and interpretation

If we want the LPIS QA to be the stepping stone to asses the Quality of the LPIS with the aim to improve the quality of the LPIS we should not focus only on the numbers in the Ets scoreboard. The assessment and the remedial actions taken are the important aspects of the LPIS QA. If the focus from the EC lies only on the numbers in the ETS scoreboard this will stimulate the PA's to perform the ETS in a political way with the aim to bend the outcomes into the thresholds set. The presentation by Daniel Amoros Pascual of DGAgri J3 gave some reassurance that this is understood bij DGAgri.

LPIS 2014

I intended to ask here the question of which type of RP is best suited to meet the requirements for LPIS coming form the new CAP (especially the Greening) from a technical point of view. But after the presentation of Wim Devos on the possible impact of the new CAP on LPIS I am slightly worried that we will have to do without clear guidance from DGAgri and JRC. As it seems now the MS's are on their own to decide in which way they will implement the new CAP in LPIS. All MS's will come with their own interpretation on what is the best solution. The time for implementation is also very short, if not too short. The EC/JRC will come with added requirements for the LPIS with regards to the new CAP in the coming years. This will lead to findings in EC audits and in non conforming LPIS QA's for many MS's. This is not in the interests of the EC, the Paying Agencies and the farmers.

Optional statements

Boundary Delineation Margin:

For the purpose of the stability in the LPIS, that is in the interest of both farmer and PA, there is a need for a Boundary Delineation Margin (BDM). This BDM is applicable in those situations in the LPIS where there is a given uncertainty on the exact location of the boundary based on CAPI. It does not apply for evident ineligible elements at the boundary of the RP or inside the RP. The BDM should cover for the uncertainty that is influenced by:
  • The quality of the orthoimages
  • The delineation by the operator
  • The interpretation by the operator
Write the BDM in the EU regulation, for example instead of the minimum mapping scale of 1:5.000.


LPIS QA Methodology

Please correct our statement into:
"Multi-parcel issue is clarified = not number of AP but units of reference parcels, not measurable block - specific issues for physical blocks"

LPIS QA results and interpretation

That ETS conforming blocks are not-conforming for the internal quality control is because there is still a difference between the ETS conformity and non-conformity from the procedure of the internal QC which stems from special landscape and LPIS management. Further studies and analysis should be done in such cases

LPIS 2014

We would like to add another idea:
Certain experiences of the LPIS QA studies and activities should give a strong support for the technical decisions of implementing the CAP reform.