4. Basic data availability and quality

As discussed in a previous section (section 2), data needed to estimate tourism pressure indicators have to be sought within a much wider area than that delimited by the NACE groups presented in Annex 1. This is so because the impact of tourism on the environment is quite complicated and cannot be directly described by existing statistics or covered by easily organised additional data collection within the identified NACE sectors.

For some areas such new data collection might be possible, e.g. if questions were asked directly from managers of hotels and similar establishments about the volume of water used, the amounts of waste produced or the number of refrigerators or air conditioners within the selected establishments. However, this information is not collected today.

On the other hand, a substantial part of the most important environmental information obtained cannot be obtained through questionnaires or similar types of data collection methods. As is the case for a major part of environmental studies, data will have to be assembled through field measurements, satellite pictures, air photography, chemical analyses etc. Statistics concerning the influence of tourism where special types of data collection may be useful could be, e.g., information on actual land use and changes of land use for tourism purposes, pollution of air and water caused by tourism, loss of biological diversity etc.

Therefore, in order to put together a list of available data, necessary for the production of indicators on the impacts of tourism, a review of the existing tourism statistics is not enough. Additional efforts to identify data useful for the proposed methods of calculation, as described elsewhere in this report, are required. That is why the present overview of accessible statistics include both variables directly related to the subject areas (tourism and environment) and variables which are potentially useful for more indirect estimations of the relevant indicators.

The research has included contacts with international data bases, e.g., at Eurostat, at UNECE in Geneva and at The European Environment Agency (EEA) in Copenhagen. As the contents of these data bases were far from fulfilling the needs of the present project, the work was carried on looking for possible useful data in publications belonging both to international organisations and to single countries. The results of these studies are presented in Annex 1, where all the considered basic data are listed and referred to the various indicators previously identified, as discussed in section 3.

The data have been organised with headings according to the type of statistics where they are generally published in national and international reports. Letters (B, DK, D, GR, E, F, IRL, I, L, NL, A, P, FIN, S, UK) designating the countries for which data can be had are given for each variable. The first row under each heading refers to the existence of data, without subdivisions, for the countries indicated. Subdivisions of the statistics which may be of interest e.g. by month, region (generally NUTS III) or other specifications are listed separately under the same heading. If data are missing for some countries, the corresponding letters are replaced with the sign “-”. Information about the data sources are given in a separate paragraph (7.2.).

For some countries where no statistics for a specific variable have been found, data may still exist. As examples can be mentioned statistics from national data bases which have never been published, data from special investigations or a statistical publication not covered by the present study, etc., can be mentioned. However, the overview in Annex 1 gives an idea about the general availability of data. More detailed studies on data needed for calculations of priority indicators should preferably involve direct inquiries to statistical offices.

The data presented in Annex 1 correspond to different ways to arrive at a given indicator. This is so because the same data may not be available for all countries. Therefore different models of calculation may have to be applied for different regions. Furthermore, on the basis of the information available, it was not possible to select the most reliable way to make the calculations. Such evaluations should be done in the future, in connection with practical tests of the various methods proposed.

The quality of the data is difficult to evaluate without access to good descriptions of the methods used for each investigation in the various countries. The reliability and comparability of the data related to areas where concepts and definitions have been agreed upon in international fora may be regarded as satisfactory for the present project. Among the different types of statistics of interest for SIP tourism, some are produced within sectors with a long tradition of international co-operation such as energy or population statistics.

As far as tourism statistics are concerned, the new EU directive 95/57/EG of November 1995, should increase the comparability and quality of the data within the nearest few years. The situation for transport and energy statistics have been evaluated within the SIP transport and SIP energy projects. For both areas, the quality of existing data has been judged as being fairly good and reliable. However, problems may appear if data are needed on different or more detailed regional levels. This will happen quite often within the SIP tourism project, especially when evaluating the impacts of an increased number of visitors to specific tourist intensive areas.

For environment statistics, the situation is particularly complicated. There are some areas where the quality of available data are generally regarded as satisfactory. That is the case for air pollution, water abstraction or supply for household purposes and amounts of municipal wastes. As in the case of the energy and transport sectors, problems may appear if data are available only at national level, but not at regional levels.

Concerning water pollution and land use or land cover, the lack of data and data availability present problems. Land use data are especially important for the evaluation of impacts on biodiversity. In the future data from the CORINE land cover system should be available. Unfortunately these data are only available on a level of aggregation where effects of tourist activities or areas for construction of tourism establishments cannot be distinguished. Changes in land use, which is the most relevant measure, can also not be obtained from existing data.

4.2. Comparison between the proposed indicators and existing data

To distinguish between the effects on the environment caused by tourism and the impacts of other activities is not a simple matter. For the present project it has been suggested that indirect methods of calculations should be used. Thus, to separate the impacts of tourists from those caused by residents within an area, the number of overnight stays compared with the total number of the residential population could be used. In the case of air pollution caused by transport, a coefficient is needed which relates the number of tourists, by type of transport, to the total number of travellers using the same means of transport (expressed as, e.g., the no. of passenger-km).

Some of these measures, called “general indicators“, have been presented in paragraph 3.3. In addition to their use as coefficients, they can also be regarded as indicators of the total tourism pressure, as required by the SAG for the topic area “Marine environment and coastal zones“. However, in this case, a regional division by coastal areas or by NUTS regions on more detailed levels than the country area should be looked for. Also when used as coefficients, a regional division of these measures is generally required.

In Table 2 presented at the end of this paragraph an overview of “core indicators“, together with existing data and some comments about how the data could be used for the calculations of the proposed measures, is given. It is apparent that one indicator can be of interest for several environment policy fields: therefore the same information may appear more than once. On the other hand, the identified data are not always available for each Member State or on a detailed regional level. More precise information about data availability has already been presented in Annex 1, and discussed in the previous paragraph 4.1.

First it should be observed, that there are few indicators for which data have been found for all countries and, in this respect, the general indicators are an exception. Other data which can be directly used as measures describing the impacts of tourism on the environment, have not been found or have been found only for one single country. On the other hand, it seems possible to calculate several of the proposed indicators, using statistics already available (as can be seen in the table presented in Annex 2).

2) useful statistics have been identified but for a restricted number of Member States (sometimes very interesting data have been found in one single country). As example, it could be mentioned the no. of visitors to protected areas (data 5.3.5 in Table 1) or area used for tourism establishments (data 5.3.3 in the same Table);

3) statistics of special interest from the point of view of tourism have been found at country level, but without any subdivisions at regional level. In this case there is a possibility that data are available in the different statistical offices or other data collecting institutions at regional levels, although they have not been found in the databases and publications studied in the course of the present project. In order to identify such data, more detailed investigations, including questionnaires to relevant institutions, are necessary. A more comprehensive study of this kind should be carried out for a selected number of indicators.

In the case of pressures generated by tourism transport, it seems that data from SIP transport, integrated with additional information, can be used for calculating some of the proposed indicators, i.e. the emissions of air pollutants of different types and the change of land use for transport purposes.

The situation is similar for air pollution caused by other types of energy use. In this case emission data are expected from the SIP energy project. For the calculations of the part due to tourism, a coefficient or “general indicator“ has been suggested. The original proposal (no. 2 of the general indicators presented in paragraph 3.3.) gives the ratio between tourist overnight stays divided by no. of residents times (365-k). The expression within parenthesis is meant to give the total number of days which the residential population has spent within the area. The number of days which the residents have spent for travelling or visiting places outside the region is indicated by the letter k.

However, no data for estimating the value of k for areas smaller than the country was found. Therefore, for regional presentation it is proposed to use the total number of days of the year (365). Thus, in the presentation of the following tables, the parenthesis (365-k) has been replaced by 365.

As well as for many other areas of environment statistics, information about changes is often the most important aspect. The material in the CORINE land cover database may therefore be of special interest in the future, provided that a future updating takes into account the need to make comparisons between different time periods. The lack of data on land use and water pollution is a serious problem, as already mentioned in the previous § 4.1.

4.3. Identification of data gaps: recommendations

Some ideas about how to handle data gaps have already been mentioned above. As a first step, it is suggested that efforts should be made to investigate whether more detailed data are available from sources not included in the present study. This means that contacts should be taken with national statistical offices and similar institutions.

Such enquiries should concentrate on a few clearly specified types of data, needed for the calculation of selected high priority indicators. The description of the required data within the proposed study should be very precise. Methods used for data collection, and other information necessary to evaluate the comparability of the statistics from different countries, should also be clarified.

For the purpose of the present indicator project, a pilot study concerning the use of data from existing water pollution registers, within the EU Member States, is proposed. An inventory of these registers was presented during the workshop in Helsinki and it was envisaged that data from these registers could be used to produce some of the core indicators suggested. A special study, with the aim to evaluate the comparability of these data and, if the result of this evaluation permits, calculate some of the core indicators suggested is therefore recommended.

A second step to improve the data situation could be to initiate co-operation with institutions responsible for the tourist sector and tourism statistics, with the aim to identify specific high intensity tourism areas. One possibility to approach this problem could be to calculate a factor describing the relation between the number of tourists and the resident population within the identified areas. Such factors have been proposed for the calculation of e.g. the impacts of tourism on energy and water use or the production of wastes.

Another area of co-operation with tourism statistics concerns the contents of existing surveys. It could be of interest to discuss the possibilities to include a few more questions in travel habit surveys, about, e.g., distance travelled during holidays, visits to national parks, hunting or fishing activities etc. Additional questions to existing surveys on accommodation facilities concerning, e.g., no. of air conditioners or quantities of water used could also be considered. Such data are not necessarily needed on a yearly basis. To evaluate the possibilities to develop this type of statistics, discussions should be initiated with representatives of the sector tourism at Eurostat and other responsible organisations within the sector, as, e.g., the WTO.

Finally it would be of great importance for the future statistics on the impacts of tourism on the environment if an updating of the CORINE land cover data base could take into account the data needs of the pressure indicator project. The identification of changes of land use is generally not as costly as mapping of the total land area. Different methods, in addition to the use of satellite pictures, could be used.

If tourism intensive regions have been identified, as proposed above, a special study of land use changes could be restricted to such areas. On the other hand, sectors, such as industry, forestry or transport, also have effects on land use and therefore on, e.g., loss of biodiversity or water use and pollution. As it should be of interest to compare the relative influence of tourism with that of other sectors, a study of land use changes should rather include all types of such changes.

The best methods to identify areas where changes due to tourist activities have occurred are however not well known at present. Statistics about areas used for tourism establishments are available on a regional level for Austria and Sweden. Such statistics may be had from land use registers, from enquiries to local administrations, from maps and aerial photographs. Satellite pictures generally seem to be more difficult to use for the present purpose, as areas used for specific tourism purposes are difficult to distinguish from other types of built up land areas.

A study of existing methods to arrive at comparable data on land use changes for different countries is therefore proposed. Also the use of GIS both for data collection and for analyses needs to be further investigated. Co-operation with CORINE and the GISCO projects might give useful inputs to the proposed study, even if the necessary raw data will probably have to be collected through several different sources.

4.4. Processes and coefficients

Unlike some other sectors, e.g. energy and transport, most of the core indicators proposed for the sector tourism cannot be estimated directly from existing variables, via a multiplication with specific emission coefficients. However, impacts caused by tourism are often already covered by other sectors. To be able to calculate the part of these pressures caused by tourism, some indirect methods have already been proposed in the present study. For this purpose quotients on, e.g., tourist overnight stays to residential overnight stays or on number of holiday travellers to the total number of travellers are needed.

The general indicators, presented earlier in this report, represent measures of the over all pressures of tourism within specified regions (usually a country). At the same time, these general indicators can be used as means to estimate that part of a certain type of pressure, covered by other sectors, which is due to tourism (as exemplified by the quotients mentioned above). For this purpose, the general indicators can also be used as coefficients. These methods have already been described in the text and in the table contained in paragraph 4.2 resp. Annex 1 .

The various ways of calculation, using data coming from other sectors, have been presented in Annex 5 . Here the variables needed for the estimation of “core indicators“ are shown together with notes on methods for calculation and on the additional measures or indicators needed from other SIP sectors. Thus, the identification of processes within the transport sector will in principle be made by the SIP transport project and the identification of processes related to use of energy by tourism within the SIP energy project etc. At the end, the same data will also be the basis for the calculation of some of the pressure indicators related to the sector tourism.

The principal difference, between the tables presented in Section 4.2. and in Annex 5, is that the variables (not the indicators) are presented in the first column of Annex 5 because the information is meant as input to the EPIS project (Thomas, 1997). Even if the concept of “processes“ is difficult to apply for important parts of the area covered by SIP tourism, e.g., for statistics on land cover and for most types of environment statistics, the variables of interest have thus been specified.

Some of these variables are meant as a basis for the calculation of coefficients (or quotients) already mentioned. In the next step these coefficients will be used, together with indicators from other SIP projects, to calculate some of the proposed indicators on the impacts of tourism.

In other cases, the coefficients calculated will be used together with other variables listed in the annex, to arrive at some of the other pressure indicators proposed for the sector tourism. As examples could be mentioned the extra water abstraction caused by the presence of tourists or the percentage of land area occupied by tourist establishments.

It has already been observed that the basic data availability needed for these calculations varies between countries. This is true especially in the case of environmental data for smaller regional areas. However it should be possible to test the proposed methods for at least some of the core indicators. A practical test to produce coefficients-useful for the estimation of the pressures caused by tourism in relation to the pressures caused by the residential population-for indicators on the impacts of tourists on waste production, water abstraction and water pollution is recommended as a first step.