Biases in International Environmental Datasets: Evidence from Water Quality Monitoring in Europe
Water Resources Research, Vol. 46, 2010.
28 Pages Posted: 29 Dec 2008 Last revised: 18 Nov 2013
Date Written: December 27, 2008
This paper is now published in heavily revised form as:
Beck, Lucas, Bernauer, Thomas, Kalbhenn, Anna. 2010. Environmental, Political, and Economic Determinants of Water Quality Monitoring in Europe. Water Resources Research 46, W11543, doi:10.1029/2009WR009065, 2010.
Please read and cite the published version.
Most international environmental datasets rely heavily on information that governments and their agents decide to collect and make available. Does this mean they are systematically influenced (i.e. biased) by political, economic, and other non-environmental factors? To find out we study data from the European Environment Agency's (EEA) water quality monitoring network in 1965-2005. If we can detect systematic biases in the data generated by this network, which is operated by relatively rich countries in a coordinated manner, biases in datasets covering more heterogeneous sets of countries in less coordinated settings are likely to be even stronger. We find that, ceteris paribus, data reporting is more extensive in international upstream-downstream river settings, EU membership has a negative effect on reporting, reported monitoring activity is less extensive in river basins where environmental pressure is higher, and richer, more democratic and economically more open countries report more. These results suggest that the EEA data is systematically affected by non-environmental factors. Our findings are likely to be relevant to most areas of environmental monitoring, except those few where data can be generated independently of governments.
Keywords: monitoring, water quality, international river, environmental data
Suggested Citation: Suggested Citation