52003DC0181

REPORT FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT on the implementation of Decision 1445/2000/EC on the application of areal-survey and remote-sensing techniques to the agricultural statistics for 1999 to 2003

CONTENT

1. Introduction

2. Implementation of the areal-survey and remote-sensing projects

2.1. The LUCAS project

2.2. The MARS-STAT project

3. Evaluation of the results of the two projects

3.1. The LUCAS project

3.2. The MARS-STAT project

4. Conclusions

5. Annex

1. Introduction

This report has been drawn up pursuant to Article 6 of Decision 1445/2000/EC of the European Parliament and of the Council of 22 May 2000 on the application of areal-survey and remote-sensing techniques to the agricultural statistics for 1999 to 2003 [1]. This Article states that "by 31 July 2003 at the latest, the Commission shall present a report to the European Parliament and the Council on the implementation of these measures and on the use of the resources made available, accompanied, where appropriate, by any proposals on how areal-survey and remote-sensing techniques may continue to be used for agricultural statistics".

[1] OJ L 163 of 4.7.2000

This general report follows the annual reports presented by the Commission to the Member States in accordance with the procedure provided for in Article 5(2) of the Decision. Those reports concern the implementation, the methods used, the use of appropriations, the evaluation of the results obtained, and the work programme for the following year.

The aim of this report is to review the two principal measures carried out pursuant to Decision 1445/2000/EC, namely: an areal survey at European level (LUCAS) [2] and the operationalisation of a system of meteorological monitoring of crops and yield forecasts (MARS) [3].

[2] Land Use/Cover Area Frame Statistical Survey

[3] Monitoring Agriculture with Remote Sensing

Experience acquired since the agro-meteorological system was developed and implemented is sufficient for reasonably firm conclusions to be drawn on the results of this measure, whereas, in the case of LUCAS, the trial period is too short to draw any definitive conclusions.

The main factors making it impossible at this stage to draw any final conclusions on the implementation of the LUCAS areal survey are the following:

* Firstly, although Decision 1445/2000/EC was intended to cover the period from 1 January 1999 to 31 December 2003, the protracted adoption procedures meant that it did not come into force until 5 July 2000. It was therefore not possible to carry out the first survey until 2001.

* Secondly, although the original plan was to carry out an annual survey in the European Union, the unexpectedly high cost of the surveys led the Commission to reduce the survey frequency to once every two years, in order to keep within budget. This means that only one survey has been completed to date.

* Lastly, the results of the second pilot survey, which will be carried out in 2003, will not be available until the end of 2003. They cannot, therefore, be taken into account in this report, nor could they have been taken into account if presentation of the report had been delayed until 31 July 2003.

In these circumstances, the report is confined to a very brief description of the two measures mentioned above and a preliminary evaluation of the results obtained thus far, which nevertheless justify the continuation of these measures for an additional four-year period, at the end of which a more complete and solidly based evaluation can be made.

2. Implementation of the areal-survey and remote-sensing projects

2.1. The LUCAS project

2.1.1. Purpose of the project

The LUCAS pilot project satisfies the need for information on land use and land cover. That need is felt most acutely in the context of the new common agricultural policy and from the point of view of enlargement.

It also answers the need for an analysis of how agriculture, the environment and rural areas interact. Greater integration of the environment in the various Community policies is regularly mentioned by the European Council (e.g. in the conclusions of the Gothenburg European Council [4]).

[4] Conclusions of the Presidency, Gothenburg European Council, 15 and 16 June 2001

The main purpose of the LUCAS pilot project is to test the feasibility of an areal survey as a means of obtaining harmonised information on the use and cover of land, especially agricultural land, within an acceptable period and with satisfactory precision at EU and (if possible) national level, but also to collect territorial information on the environment.

The first areal survey took place in 13 Member States in 2001. Because of the outbreak of foot-and-mouth disease in the United Kingdom and Ireland in 2001, those countries were not able to conduct the survey until 2002. A second comprehensive survey will be conducted in 2003. This second pilot survey will also allow a time analysis of changes in land use and land cover and the landscape structure.

2.1.2. Methodology

LUCAS is a statistical areal sample survey (i.e. a survey based on the visual observation of a sample of geo-referenced points, without the involvement of farmers). As LUCAS is intended to provide information on the entire territory, not just agricultural land, areal sampling (i.e. per section of territory) was preferred to a sampling of agricultural holdings.

The LUCAS pilot project has hitherto comprised two stages: a field survey, in spring, which is the areal survey proper, and which collects data on land cover, land use and the environment, and a survey in the autumn, consisting of interviews of a sample of farmers, which provides additional information, particularly on yields, agricultural techniques, sowing intentions, etc. It should be stressed, however, that only the field survey is really essential to the LUCAS project.

A two-tier sample design was adopted for Phase 1 of the pilot survey, comprising primary sampling units (PSUs), which are 18 km-square cells in a regular grid, and secondary sampling units (SSUs) which are rectangles (1 500 m x 600 m) located in each PSU. Each rectangle contains 10 sampling points. The sample contains approximately 10 000 PSUs, covering the entire territory of the EU. The LUCAS observation is carried out at the exact spots where the SSUs are located.

In addition to the collection of data on land use and land cover, the enumerator systematically observes the "environment" of the PSUs. Moreover, variations in the linear characteristics of land and land cover in space (landscape structure) are collected along the straight line connecting the five first SSUs.

Phase 2 comprises the collection of additional information on yields, sowing intentions and farming practices. It is carried out among a sample of 5 000 farmers selected at random from the survey base used in phase 1.

Table 1 (see annex) shows the cost breakdown for the 2001-2002 LUCAS survey for the various lots into which the work was divided. Table 2 shows the expenditure incurred by the Commission (Eurostat) in 2002 for a number of methodological analyses concerning the LUCAS survey.

2.2. The MARS-STAT project

2.2.1. Purpose of the project

The purpose of the agro-meteorological system of monitoring crops and forecasting yields, developed by the Joint Research Centre (JRC) as part of the MARS project, is to provide the elements necessary for an understanding of how climatic events impact on harvests and to forecast the yields of the main crops.

In accordance with the implementation methods provided for in Decision 1445/2000/EC, the activities are aimed at activating both the agro-meteorological model and the processing chain of NOAA and VEGETATION low-resolution satellite data. These satellite images are used to analyse, monitor and forecast European field crop yields. The JRC has undertaken to maintain the software and provide overall project supervision.

2.2.2. Methodology

The MARS System of monitoring crops and forecasting yields comprises four activities:

* the purchase of weather data and the processing, calculation, interpolation and storage of derived weather data;

* use of the weather data in the CGMS (Crop Growth Monitoring System) agro-meteorological model. The results consist of a series of indicators for simulating the growth of the main crops. These indicators are stored in a geo-referenced database which allows the production of thematic maps of weather statistics and indicators for each crop;

* acquisition of NOAA-AVHRR and SPOT-VEGETATION satellite images and the processing of these data by the SPACE system. Indicators of the state of vegetation are calculated by incorporating CORINE Land Cover data;

* the creation, at the JRC, of a group of experts dedicated to the statistical analysis of the data and quantitative forecasts using previous indicators.

The activities mentioned in points 1, 2 and 3, being operational in nature, have been privatised. They were awarded to a contractor who was selected following a call for tender (published in OJ S228 of 24.11.99). The contract concluded for this purpose (Mars-OP) comprises four lots and three phases. Lots 1, 2 and 3 correspond to the activities of items 1, 2 and 3 described above. Lot 4 comprises research activities which were funded from JRC research funds. The first phase covered the period 2000-2001, the second phase 2002 and the third, 2003.

The activity referred to in point 4 has been managed directly by the Commission at the JRC, in coordination with DG Agriculture. The most notable result of this activity is the MARS Bulletin, which appeared on six occasions in 1999, six in 2000, seven in 2001 and seven in 2002, with six planned for 2003 (the final publication being at the end of November 2003). The Bulletin is sent to the appropriate Commission departments, the Member States, the statistical services and the various Ministries of Agriculture. It is also published on the Internet at http://mars.jrc.it/stats/ bulletin as a service to professionals and the scientific community.

Table 3 (see annex) shows the use of Community funding for the MARS-STAT project (agro-meteorological system) in the period 2000-2003.

3. Evaluation of the results of the two projects

It should be borne in mind that the late adoption of Decision 1445/2000/EC, which covered the period 1999-2003, meant that it was not possible to carry out the first LUCAS pilot survey until 2001. It was possible to continue the agro-meteorology-related activities during the transitional period (1999-2000) thanks to funds temporarily available to the JRC for other research and development activities (MARS Project).

3.1. The LUCAS project

The pilot survey carried out in 13 Member States in 2001 and in the United Kingdom and Ireland in 2002 demonstrated the feasibility of this project on a Community scale. During this first phase, it was shown that the project would:

* provide a common methodology and classification for data collection and estimates of land use and land cover;

* provide harmonised information for the entire territory of the European Union;

* yield early information on the areas under crops;

* extend the system (which, being dedicated entirely to information on land use and land cover, was originally intended for users of agricultural statistics) to environmental experts, who will thus have at their disposal a unified database covering, inter alia, soil erosion, the landscape, natural disasters, noise, etc.

* provide statistical data necessary for implementing indicators measuring the integration of environmental concerns in the common agricultural policy, as described in Commission Communications COM(2000) 20 final and COM(2001) 144 final.

Before the project was implemented, it had been estimated that the sample design described above would be accurate to within about 2% for the main classes of land use and land cover at EU level. This was confirmed by the 2001 and 2002 pilot surveys. A number of Member States are now thinking of adapting this tool to their own needs to replace the collection of certain data.

Unfortunately, areal surveys proved more expensive than initially anticipated, which is why the survey can be conducted only once every two years. It also explains why the number of observation points has had to be reduced, with the result that representativeness is satisfactory at EU-15 level only, and not at Member-State level. Under these conditions, it would be advisable to study how representativeness can be improved, either by increasing the number of observations in certain Member States or by optimising the sampling plan.

After just one Community-wide survey, and without its full potential yet having been analysed and used, it is impossible to make a comprehensive evaluation of this pilot project. It will not be possible to make such an evaluation before the results of the 2003 survey are analysed. These will not be available until the end of 2003. Nevertheless, in order to make the evaluation as authoritative as possible, it should be based on the results of a sufficient number of surveys, incorporating all the technical changes which are found to be necessary in the light of experience. The Commission believes that this should be possible once the third survey has been implemented and the results analysed: hence the need for a limited but sufficient extension to the basic Decision.

In the meantime, a feasibility study could investigate the possible contribution of photo-interpretation to this project. If the results are positive, photo-interpretation could be used to identify observation points located at an altitude of more than 2 000 metres or at places with poor accessibility. Under certain conditions, it could even replace ground observation for certain years.

Given appropriate technical changes (e.g. the training of enumerators), an extension to the pilot phase would also make it possible to explore the options for improving the accuracy of the results at Community and national level and to ascertain the extent to which the Member States could be more closely involved, on a voluntary basis, in implementing this project, e.g. by taking on responsibility for additional observation points with a view to increasing the density of the points grid. Lastly, interrupting the pilot phase after just two surveys in order to conduct a preliminary review would probably result in a loss of know-how which would be difficult to make good in the short term.

3.2. The MARS-STAT project

The MARS System has made it possible to evaluate the impact of the climate on harvests at EU-15 level and in the applicant countries, in an independent and homogeneous manner throughout the European continent.

* for the 1999/2000 marketing year, the system made it possible to evaluate:

- the problems brought about by heavy rains during sowing in France, Germany and Central Europe;

- the drought in Spain.

* for the 2000/2001 marketing year, the monitoring demonstrated:

- good sowing conditions;

- the winter drought in Spain, followed by good conditions;

- the early-summer drought in Central Europe;

- excessively wet harvest conditions in several parts of Northern Europe.

* for the 2001/2002 marketing year, the system made it possible to evaluate:

- the problems caused by heavy rains during sowing in France, Germany and Central Europe;

- the favourable conditions for winter crops in Spain;

- the good harvest in the applicant countries.

* for the 2002/2003 marketing year, the monitoring demonstrated:

- good overall sowing conditions;

- the winter and spring drought in Spain and southern Italy;

- generally improved straw cereal yields in France;

- heavy summer rain in several European countries. These conditions affected the harvests which were under way in August, especially in Germany, and caused localised damage to crops (particularly the floods in Germany and the Czech Republic).

The MARS Bulletin is made available to the relevant Directorates of DG Agriculture three or four times a year. The yield forecasts resulting from this activity are used by DG AGRI as input data to compile the estimated balances for field crops for the European Union and for the applicant countries.

With the help of the contractor for the Mars-OP activities, the JRC also set up the website http://www.marsop.info, where users can consult the data resulting from the Mars-OP contract. This site, which was the object of a presentation to the Working Party on Crop Products at Eurostat in 2002, allows authorised users to download information, indicators and meteorological, agro-meteorological and satellite maps. The statistics on the MARS site show a hundred independent hits on the bulletin site (http://mars.jrc.it/stats/ bulletin) (including short-term analysts and international organisations) in the course of the year. There are about a hundred registered users of the site http://www.marsop.info, including about fifty whose use of the site is not subject to time limits.

Evaluating the results is a permanent activity at the JRC's MARS unit. The a posteriori errors in the quantitative yield forecasts are calculated from the final official data. For 1999, 2000 and 2001, the forecasting errors are equivalent to or less than the average errors of the MARS System published in previous reports. For 2002 and 2003, comparisons will be made once the official data become available. A study known as QUAMP (Quantitative Analysis of MARS Predictions) was launched by the JRC to detect and evaluate the sources of forecasting error. The results will become available in 2003.

4. Conclusions

In the light of the partial evaluation set out in this report and of experience to date, the Commission believes that a four-year extension (from 2004 to 2007) to the basic Decision would enable it to carry out an additional areal survey and a much less random and more authoritative evaluation than one based solely on the results of two, and possibly only one, survey. In addition, a similar extension for the MARS project (the agro-meteorological aspect) would be fully compatible with the positive evaluation of this measure in recent years.

5. Annex

Table 1

Expenditure incurred for the 2001-2002 LUCAS survey (in EUR)

>TABLE POSITION>

Table 2

Expenditure incurred for the LUCAS methodological analyses in 2002 (in EUR)

Description // COST (EUR)

Methodological analysis of the results of the 2001 LUCAS survey // 90,200

Study of the role of photo-interpretation in the LUCAS survey // 49,421

Study of the use of data from the Community's LUCAS survey // 182,074

Lucas-Phare project // 447,500

TOTAL // 877,624

Table 3

1.1.1.1. Expenditure incurred for the agro-meteorological system (MARS-STAT) (in EUR)

>TABLE POSITION>