Identifying and closing gaps in environmental monitoring by means of metadata, ecological regionalization and geostatistics using the UNESCO biosphere reserve rhoen (Germany) as an example

Abstract

In Germany, environmental monitoring is intended to provide a holistic view of the environmental condition. To this end the monitoring operated by the federal states must use harmonized, resp., standardized methods. In addition, the monitoring sites should cover the ecoregions without any geographical gaps, the monitoring design should have no gaps in terms of ecologically relevant measurement parameters, and the sample data should be spatially without any gaps. This article outlines the extent to which the Rhoen Biosphere Reserve, occupying a part of the German federal states of Bavaria, Hesse and Thuringia, fulfills the listed requirements. The investigation considered collection, data banking and analysis of monitoring data and metadata, ecological regionalization and geostatistics. Metadata on the monitoring networks were collected by questionnaires and provided a complete inventory and description of the monitoring activities in the reserve and its surroundings. The analysis of these metadata reveals that most of the monitoring methods are harmonized across the boundaries of the three federal states the Rhoen is part of. The monitoring networks that measure precipitation, surface water levels, and groundwater quality are particularly overrepresented in the central ecoregions of the biosphere reserve. Soil monitoring sites are more equally distributed within the ecoregions of the Rhoen. The number of sites for the monitoring of air pollutants is not sufficient to draw spatially valid conclusions. To fill these spatial gaps, additional data on the annual average values of the concentrations of air pollutants from monitoring sites outside of the biosphere reserve had therefore been subject to geostatistical analysis and estimation. This yields valid information on the spatial patterns and temporal trends of air quality. The approach illustrated is applicable to similar cases, as, for example, the harmonization of international monitoring networks.

Using generalised dissimilarity modelling and targeted field surveys to gap‐fill an ecosystem surveillance network

“…Because we aimed for representation and inclusion rather than even sampling across ecosystems (Schröder et al., 2006), we used a ‘nearest‐neighbour’ approach (Faith & Norris, 1989), in which each ecosystem is ideally ‘serviced’ by a representative site, with a preference to add new sites in habitats that are currently the least well represented (Hargrove & Hoffman, 2004). We did not aim to make sampling even across all environments, not least because part of the original rationale behind Ausplots was that the Australian rangelands were more poorly monitored than mesic coastal areas (Guerin et al., 2017).…”

A geostatistical approach to data harmonization – Application to radioactivity exposure data

“…It is well known that different protocols complicate the joint use of disparate datasets in statistical models (Gego et al, 2005;Herold et al, 2006), and that preprocessing is required before different datasets are merged. Standardization is the prevailing approach, as concluded in inter-comparison studies that examined field protocols in various environmental risk assessment fields (Wagner et al, 2001;Parr et al, 2002;Schröder et al, 2006). Standardization implies that all parties involved agree on the methods used for data collection.…”

“…Both components must be addressed. Furthermore, detailed information on data collection and measurement specifications such as meta-data in a GIS (Schröder et al, 2006) is of beneficial use in the development of a harmonization procedure.…”