Transition Facility Statistical Cooperation Programme 2005
LOT 2 –Pesticide Indicators
MISSION REPORT HUNGARY
Event:Monitoring mission Hungary
Location:Budapest
Date: 26-27 Une 2007
Expert:Catherine Kesy and Peter Pauli (Short-term experts)
People met:
Name / Position / InstituteZsuzsanna Szabo / Statistician / KSH, Environmental Section
Pal Aujeszky / Head of Section / KSH, Environmental Section
Zoltan Ocsko / Senior counsellor / Ministry of Agriculture and Rural Development
Not able to participate: Mrs Laszlone Pecsi Paula (Plant Protection Service)
Mission rationale
As mentioned in the Terms of Reference, support will be provided for most of the countries on the design and implementation of a new survey on the use of plant protection on wheat, particularly at the stage of the preparation of the sampling strategy.
The items to be taken particularly into consideration during the direct assistance were:
- The sampling frame
- The definition of the sampling unit
- The usage of stratification
- The production of national estimates and the precision of final results
Introduction
The mission has mainly followed the agenda prepared by KSH (see Annex 1).
The mission started with an introduction of the role of each of the participants in this project:
- Hungarian Central Statistical Office (HCSO)
- Agricultural Office Deputy Directorate Responsible for Authorization of Plant Protection Products/ Plant Protection Department
- Expert from Ministry of Agriculture and Rural Development (MARD).
The Project’s daily management is in the hands of the people met during the mission who were also the main players in the previous PHARE (ADAS) project. The practical operations are concentrated in the Statistical office (Environmental section).
The main bottleneck is that it has not been allocated a well defined part of KHS’s regular work to this project. In Hungary that means in general that not much regular working time is made available for the team until the activity is backed with a legal EU basis and automatically promoted to the work programme. As a consequence some work is made in overtime.
General information on pesticide statistics in Hungary
In the past, Hungary noted around 1500 State farms, now it has 707 000 private owned farms and 7 900 enterprises. The enterprises however have an average size of 487 ha (i.e. 50.5% of the total Agricultural land) against 3.5 ha for the private farms.
It means that, covering the enterprises by a robust and decent sample survey plus a restricted sample from the private farms, the objectives of the PUS can easily be reached.
In parallel to Pesticide use estimates, which are already being produced since the early 50ties, Hungary also produces the PPP sales on an annual basis for more than 40 years. These data are currently collected by the Research Institute for Agricultural Economics. Country pesticide use in the past was based on domestic production but now around 90% of products on the market are imported (See Annex 2 and 3).
Presentation of the “Wheat” questionnaire (See Annex 4)
Hungary has chosen to drastically reduce the questionnaire compared to the one used under the ADAS project, which was far too detailed. It increased resistance and, hence, increased non-response and did not generate extra output. In addition, in the ADAS project all farms were visited while in this project the bigger enterprises will be enumerated by e-mail only. Keeping the questionnaire simple avoids non-response in this group. The private farms however will be enumerated by pesticide inspector though here too non-response must be kept to the minimum.
All questionnaires will be filled out electronically. The pesticide inspector will have the application in its lap-top and fill in all data on the spot. The enterprises receive the same application by e-mail and fill in themselves.
The questionnaire contains a general part (covering farm variables) and a specific part. In particular, the latter is interesting since provided with pull-down menus for choosing the product used, type of spraying, unit used, etc. In this way, error margins remain limited and cross checks and validation can be made on the spot. In fact, if mistakes are being made, a warning is given and corrections must be made before the questionnaire is entered in the database.
This approach may serve as an example for those countries that have not decided on the format and approach yet. It is practical, reliable and bypasses the process of data entry, which is a source of error risks and error checking.
Sample design
Hungary prepared a paper in which the sample methodology is presented (see Annex 2) and, hence, only some additional remarks are being made in this report.
The basis for the sample frame is the Farm register (FR), which originates from the Agricultural Census 2000. As in most New Member States, the structure in agriculture is rapidly changing and the FR has no updating mechanism as the Business register and, hence, it is rapidly deteriorating.
After the Census, two FSSs were carried out (2003 and 2005) with a sample fraction of around 15% each, which contributed to the partial update of the register.
This register is still the basis for all sample surveys in agriculture: crop, animal, etc. and also for the PUS survey. All farms (2 858 enterprises and 9 546 private farms) with wheat production were selected and classified in 60 strata; 3 size classes for each of the 20 counties (NUTS 3). From this sample frame a random sample of 780 units was drawn: 430 enterprises and 350 private farms.
As discussed during the workshop in Vilnius, the number of 780 is the result of a calculation of the maximum number of units that can be surveyed within the project budget with the questionnaire that stands now (see also previous point). It means that 780 units are not the results of the sample but of the optimisation method.
A substitute list is produced for non-response though efforts will be made to keep this to a minimum. In this way, imputation and assumptions will be negligible.
The farm is the sample unit and all data will be calculated on farm level, not plot level.
Survey organisation anddata processing
The survey was planned to start in June, but organisational and re-organisational issues of some of the project’s partners have created delays. The current plan is to start the fieldwork during the summer and have preliminary results in early September.
Prior to the survey, an announcement of the upcoming event will be distributed to all selected farms. Then, dates will be agreed for the actual visit. Because the PPP inspectors are the enumerators, it may be expected that no significant refusals will be noted. If so, the substitute list will be used to achieve the 100% response.
All data collection is carried out by the electronic questionnaire (see above and Annex 4); by e-mail for the enterprises and face-to-face by the PPP inspectors who will use their lap-tops.
Training of the 25 inspectors involved will be relatively easy and short since they have already been involved in the ADAS project, which was much more extensive. However, the collection method and software is new and needs to be explained.
Data check (the validation) will mainly be carried out on the spot by the software after the enumerator entered the data. It means no other data entry stage and that the first stage of data processing is immediately finished once the last questionnaire is entered in the system.
Because of the procedure, as soon as the fieldwork is over, the statistical manipulations can start: data cleaning, raising the sample results, correction factors, etc. This process may reveal systematic irregularities, which have to be resolved on central level, either mechanically or by phone. Only in extreme cases a farm will be revisited.
In addition, first results will be ready shortly after the former activity is finished and final results are expected to be ready for publication early 2008.
Discussion of questions related to the harmonization of data collection
The issue of crop selection and periodicity was discussed. Under the assumption that resources would be available, KSH would prefer to carry out this survey on a regular basis and for a key group of products, e.g. every two years for all selected products. This generates time series and, hence, risk indicators, not only pesticide use data for one product once every 5 years. Such an approach would also enable cross-country analysis, which is now less feasible.
The choice of wheat in this Pilot project is understandable since no surprises are to be expected: only the more professional farmers are involved, there is an average of 2 sprayings and no real distinction between the use of products on different plots on one farm. It can even be argued that this could also be estimated via other (cheaper) methods. In this sense it has only delayed the solution for the real problematic products (fruits and vegetables), which require really different observation methods due to the wide variety and number of treatments of PPPs used (because of weather conditions, climate zones, etc).
Conclusions andNext Steps
- So far, the preparation of the PUS for wheat in Hungary seems to be on track, apart from a delay of around 6 weeks in the start of the data collection.
- Hungary has chosen to use an electronic questionnaire only. The “private” farms are visited by inspectors who will use tablet PCs, entering the data directly into the system. The “big” enterprises will receive the questionnaire by e-mail, uploading the data directly into the system.
- We expect that the data quality of the private farms will be high and the non-response low. For the enterprises, there is the risk that it will be the opposite and, hence, KSH needs to keep a tight control on the process.
- The sample size is (too) small simply because of a lack of financial resources. As long as this survey is not part of the statistical work programme (i.e. official EC regulation), only project money (extra) determines the size of the survey. In addition, work connected to such surveys is also not planned in the work programme and must mainly be done in overtime.
- Results can be made available very quickly after the survey period due to the e-format of the survey. If this approach generates satisfying results, it may be given into consideration to other countries.
- The Regulation should be more adequate than the one currently proposed. The time span of 5 years is too wide; it should be 2 in which all relevant crops are being surveyed.
- In this project, only wheat is covered, but this masked the solution of the real problem: the production in “small” farms for own consumption and their use of PPPs. Wheat is not or not significantly produced for own consumption. In contrast, products like potatoes, tomatoes, onions and cucumber are in huge quantities and PPPs most definitely used (even in relatively high doses). Eurostat’s proposals for defining the threshold for being a farm are in this respect found inappropriate.
- Hungary has showed great interest in risk calculation. In that respect, the forthcoming October meeting is welcome, as it will discuss the HAIR indicators. Information about this project is available on a website maintained by RIVM at (links provided by Eurostat):
and
- The calculations will be made at farm level, not at parcel level; nonetheless, as the questionnaire is collecting data about PPP received on each parcel, the Contractor is suggesting that each parcel should receive a specific plot n°, to enable calculations of local risk or other types of analysis.
- Some concrete questions to be answered by Eurostat include:
-the definition of average use: total PPP use / total area cultivated
-rate of application: total PPP use / area treated
-area treated: area receiving PPP * the number of applications on this area
or area receiving PPP
1
Annex 1 - AgendaHungaryMission
AGENDA
ICON Experts’ Visit to Hungary
Transition Facility 2005 – Pesticide Indicators
Budapest - Fényes Elek st. 14-18., 2nd floor
26-27 June 2007
Day 1, Tuesday 26 June
9.15 / 1 / Opening of the visit / HCSO9.30 / 2 / Introduction of ICON and ASA’s roles in the project / ICON
9.45 / 3 / Introduction of the participants of the project
Overview of their roles and tasks
General information on pesticide statistics in Hungary / HCSO – Plant Protection Inspectorate – Expert
10.45 / Coffee break
11.00 / 4 / Overview of data requirements of the new Regulation (see Annex 1A) / ICON
11.15 / Questions, comments, discussion / HCSO–ICON
11.30 / 5 / Presentation of wheat questionnaire:
- further additional information to be included
- analysis of the additional information
11.45 / Questions, comments, discussion / HCSO–ICON
12.00 / Lunch break
13.30 / 6 / Presentation of the imputation and estimation methods of experts / ICON
14.00 / 7 / Presentation of the utilized sampling strategy
- sampling unit
- national estimation
14.20 / Questions, comments, discussion / HCSO–ICON
14.35 / Coffee break
14.50 / 8 / Data processing – presentation of the conversional table / HCSO
15.05 / Questions, comments, discussion / HCSO–ICON
15.20 / 9 / Highlight of crucial issues concerning the survey / ICON
15.40 / Questions, comments, discussion / HCSO–ICON
16.00 / End of session
Day 2, Wednesday 27 June
9.30 / 10 / Discussion of questions related to the harmonization of data collection:- calculation of average use
- crop selection
- reference period
11.00 / 11 / Conclusion, remarks / HCSO–ICON
11.30 / End of visit
Annex 1A - Comments on issues highlighted by Hungary (ICON)
Overview of data requirements of the new Regulation - Issue discussed on Tuesday, 26th of June at 11.00
Requirements set to Member States :
- Provision of data on PPP use
- Provision of data quality report
FOR MEMORY
Proposal for a Regulation of the EP and of the Council concerning statistics on PPPs.
The proposed Regulation creates a legal framework and lays down harmonised rules for the collection and dissemination of data concerning the placing on the market and use of plant protection products. In particular, it instructs the Member States:
–to collect data regularly (annually as regards placing on the market - every five years as regards use);
–on how to collect data, whether by representative surveys, statistical estimation procedures on the basis of expert judgements or models, reporting obligations imposed on the distribution chain for plant protection products, reporting obligations imposed on professional users, from administrative sources or by a combination of these means;
–on how to transmit data to the Commission.
------
- Provision of data on PPP use
EXPLANATORY MEMORANDUM
1)CONTEXT OF THE PROPOSAL
1.1.Grounds for and objectives of the proposal
By requesting mandatory data collection, the main objective of this Regulation is to ensure that comparable data are collected in all the Member States, making it possible to calculate harmonised risk indicators and to measure the progress made towards more sustainable use of plant protection products throughout the Community.
1.2.General context
Since the effects of the relatively new legislation on biocides[1] will not become apparent until well after 2006, when the first evaluation of active substances for use in biocidal products will be finalised, neither the Commission nor most Member States currently have sufficient knowledge or experience to propose further measures regarding biocides. The scope of the Thematic Strategy and the scope of this proposal have thus been limited to plant protection products. However, both scopes may be expanded in the future, if similar measures were considered necessary for biocides.
(5):in view of the aim of calculating accurate risk indicators according to the objectives of the Thematic Strategy on the Sustainable Use of Pesticides, statistics need to be detailed up to the level of the active substances.
Article 1
Subject matter and scope
1.This Regulation establishes a framework for the production of Community statistics on the placing on the market and use of plant protection products.
2.The statistics shall apply to:
–the annual amounts of plant protection products placed on the market according to Annex I;
–the annual agricultural use amounts of plant protection products according to Annex II.
Article 4
Implementation measures
- The following measures necessary for implementation of this Regulation, including measures to take account of economic and technical developments, shall be adopted in accordance with the regulatory procedure referred to in Article 5(2):
(a)adoption of the appropriate technical format for the transmission of data (Article 3(3));
(b)definition of the format and content of the quality reports to be supplied by the Member States (Section 6 of Annexes I and II);
ANNEX II
Statistics on agricultural use of plant protection products
SECTION 1
Coverage
1.Statistics shall cover the agricultural use of plant protection products in each MemberState.
2.Each Member State shall select a set of crops listed in categories D, F, G and I of the characteristics defined in Annex I of Council Regulation (EEC) No 571/88 and shall compile statistics for these crops. Statistics shall cover at least 75% of the total quantity of substances placed on the market annually for agricultural use as estimated in the quality report on the second reference year referred to in Section 6 of Annex I.
3.Statistics shall cover all substances listed in Annex III consisting of active substances, safeners or synergists contained in plant protection products used on the selected crops during the reference period.
SECTION 2
Variables
1.The quantity of each substance listed in Annex III contained in plant protection products used on each selected crop shall be compiled with the total area cultivated and the 'area of the crop treated' with each substance.
2.The definition of the 'area of crop treated' shall be determined according to the procedure laid down in Article 5(3).
SECTION 3
Reporting measures
1.Quantities of substances used shall be expressed in kilograms.
2.Areas cultivated and areas treated shall be expressed in hectares.
- Provision of data quality report
Article 4