LPISQA Legacy/ETS/Analysis

From Wikicap - European Commission

Connection between the measures and the quality element indicators (aka: Which non-conforming parcels are counted where?)

In order to account a non-conforming parcel in the right indicator the connection between the quality measures used for the assessment and the quality element indicators has to be known. The arrows in the figure below provide the links between the RP quality measures with conformity levels and the LPIS quality measures. The latter directly relate to specific LPIS quality elements.

  • RPs with un-waivered critical defects (10106_2), are considered non-conformant. They are reported in measure 10205_2 (Table 14.2 of Annex I).
  • Ps with “area-based” non-conformity - quality measures 10102_2 and 10102_3 - are reported in Measure 10202 (Table 11 of Annex I).
  • he DQ_Scope of Table 9 (10107) comprises all reference parcels that are considered non-conformant and for which the cause of non-conformity should be reported. This includes the “area-based” and “unwaivered critical defect” types of non-conformity.
  • The number of causes for non-conformity is reported in measure 10204 (Table 13 of Annex I).

Connection between quality elements and measures

For QE5, reference parcels having more than 10% difference between the observed eligible area (rounded to 0.01 ha) and the area declared, are considered non-conformant. However, when they are conformant with the rest of the quality elements at parcel level, they are not subject to quality measure 10107. In this case the cause of the difference is obvious and out of the direct control of the administration (i.e. farmer’s behavior).

Units of measurements and precision to be used

Area of the agriculture land cover polygons should be rounded to square meters without decimals (e.g. 25761 m2). Area reported in the XML and GML files should be rounded to square meter (e.g. 25761 m2).

The value for the Arec given in the LPIS Polygon zero state should be expressed in hectares with exactly 4 digits after the decimal. This will ensure than observations during the screening will be truly compatible with the ETS observations (e.g. 2.6159 ha).

The values for “observedToRecordedAreaPercentage” are expressed in percentage with 2 decimals after the decimal point (e.g. 98.48%).

Cumulative land changes – denominator of measure 10207

Although the scope of the measure relates to all parcels declared over the previous years, to calculate the rate of change for the year under assessment only the information relating to that particular year shall be used. This means that:

  1. The rate of change declared by farmer in year N = number of parcels with change notification by farmer in year N divided by the total number of parcels declared in year N
  2. The rate of change identified by inspectors in year N = number of parcels with change discovered by inspectors in year N (classical OTSC + CwRS) divided by the total number of parcels inspected in year N (classical OTSC + CwRS)

Linking QE6 to update needs

Update is the most important challenge for any GIS, including the LPIS. Several different processes may contribute to keeping the information up to date (see LPIS update) from partial updates till an "acute refresh", i.e. replacing the database by a newly produced one. From cost efficient point of view the latter is by far the worst option.

The purpose of QE6 is to monitor the update processes so that an acute refresh can be avoided. Like other other measures, QE6 should not deterministically trigger a reaction but lead to analysis and a sound remediating plan.

QE3 provides a categorization of non-conforming and defective parcels according to the reason of failure, which also comprises the reason of insufficient updates. However this category does not give information on the individual processes triggered by different actors (farmer/ inspector/ LPIS custodian/ national mapping agency). The knowledge of the relative abundance of these transactions, if managed effectively, gives input to the actual update needs.

How to implement measure 10208

Please apply the following clarifications: Area not found: the difference between area declared and area determined based only on eligibility conditions on the land. This means that for SPS the availability of entitlements is not considered. i.e. the second part of 2009R1122 art 2(23) and art. 57(2) are not applicable.

Although 2009R1122 art 84 (1)(d) calls for a bulk reporting on July 15th, the “area not found” for this measure should have been determined before late autumn of the preceding year. See 2.4 of the art 34 guidelines.

Restricting the difference to over-declarations according to eligibility criteria described in 2009R1122 art 58: add up the values of the positive differences on an aid application by aid application basis.

The current measure is not representative enough

At current stage a revision of this quality element is probably not needed but an alignment between the OTSC random sample and the LPIS QA sample should be prepared from 2013 onwards. The purpose of this action would be to achieve a common random sample OTSC/LPISQA large enough to draw valid conclusions on the effect of LPIS on irregular aid applications.

Since the OTSC sample is produced by the Paying Agency and the LPIS QA sample pre-selection by the JRC, alignment of both procedures will be necessary. JRC proposed such alignment on the Paphos conference and update the art 34 and art 6 guidance where necessary. However, this proposal is not effective for the 2013 assessment.

Causes for “area not found” in OTSC

A part of declared are may not be found in course of OTSC. The possible causes include:

  • The declared land is not in agricultural use
  • Parcel over-declaration
  • Parcel under the minimum size
  • Parcel not found at all
  • Eligible land but not used for farming

How to proceed with a non-conforming parcel?

The purpose of the ETS is to collect unbiased data on the LPIS as a whole. It is not intended to be the main (and surely not an exclusive) information source for updating individual parcels. However, the ETS provides a trigger (“anomaly”) that drives the update process. Therefore the ETS inspection results of an individual parcel should thus be considered as a potential anomaly. So the Member State should:

  1. Consider updating all inspected parcels where an anomaly was identified.
  2. Collect the required update information for the non-conforming parcels in compliance with the existing procedures. The ETS inspection observations may not necessarily be fit for this purpose.

Adding a precision component to QE1

To help correctly assess the value of the quantification of the maximum eligible area (QE1) the bias is complemented by a precision component. As a result, the existing bias measure will be addressed as QE1a, while the introduced precision component as QE1b.

Introduction of the QE1b affects:

  • Analysis of observed data - calculation of QE1 - lower and upper interval boundaries
  • LPIS quality assessment report - QE1b added
  • LPIS scoreboard schema - LIB and UIB added