![Sample our Environment & Agriculture journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5934/TOC-Subject-Sample-2021-Environment-Agriculture.jpg"})
Abstract
This analysis provides an example of how biodiversity can be measured by means of different indicators, and how the latter can be used to assess the influence of the biodiversity profile of a region on the tourism flows towards it. Domestic Irish tourist flows have been chosen as a case study. Results show that most of the biodiversity and landscape indicators included in the analysis turn out to be statistically significant in determining tourists’ choices regarding the duration of their trip. As a result, policies pursuing biodiversity conservation appear to have a positive impact on the regional tourism.
Acknowledgements
The authors wish to thank the editor and two anonymous reviewers for constructive comments. This study has been conducted within the framework of the project CIRCE – Climate Change and Impact Research: the Mediterranean Environment, contract N. GOCE 036961, funded by the European Commission within the Sixth Framework Programme. The content and opinions presented in this article exclusively reflect the views of the authors.
Notes
1. The survey is one of several Central Statistics Office (CSO) tourism surveys conducted to comply with the requirements of the Council Directive 95/57/EC of 23 November 1995 concerning the collection of statistical information in the field of tourism.
2. These thresholds are provided by the Natura 2000 database and are taken as a starting point for the computation of the values of each indicator. Narrower intervals would be useful in order to provide a more precise measure of biodiversity; however, considering the extreme difficulty in achieving reliable data, the information contained in this database is deemed to be sufficiently detailed.
3. The chosen procedure for constructing the indicators may cause some concern, due to the width of the intervals associated to the qualitative rankings. However, it seems important to notice that a precise assessment of species and habitat diversity is extremely difficult to achieve, at least at policy relevant scales. In addition, the focus should be placed on the comparisons of the biodiversity profiles of different areas. From this perspective, as long as the data are collected following a standard format and the indicators are constructed in a consistent manner, the lack of precision of the measurement does not affect the quality of the results.
4. The Poisson model has been formally tested against negative binomial models as can be seen from . The chi-squared value associated to the Likelihood-ratio test of alpha = 0 is 85.89, therefore suggesting that in this specific case the Poisson model better fits the data. The Poisson model has been run with and without correction for random effects, using the county of origin as the group variable. The two models did not differ substantially. The use of pooled panel estimation would have been appropriate; however, it is not compatible with the data structure, since the sample of households included in the survey is not constant over the period 2000–2003.
5. In order to test for multicollinearity among the independent variables, a linear regression model has been run, using the same dependent and explanatory variables, and the variance inflation factor (VIF) has been computed. On average, the VIF turned out to be 3.89. A correlation matrix among biodiversity and landscape variables has been computed and variables whose coefficient was higher than 0.30 have been eliminated from the analysis.