The High Nature Value farmland indicator in Germany
Farmland covers about 50% of the total area of Germany. Agriculture therefore has a considerable influence on the biodiversity of the open landscape. Progressive agricultural intensification has led to a dramatic decrease of low-intensity farmland and agricultural biodiversity after the mid-20th century. High nature value (HNV) farmland consists of e.g. species-rich grassland, extensively managed arable land or vineyards and fallow land. Often it not only provides a high biodiversity but also affords habitats for rare or endangered plant and animal species, which cannot survive within the intensively used agricultural landscape. Also landscape elements, which provide habitats for further species, count as HNV elements. Implemented agri-environmental schemes aim to preserve agricultural biodiversity and to preserve and improve HNV farmland. In order to indicate increase or decrease in quantity and quality of HNV farmland it is necessary to monitor the state of HNV farmland over time. Hence it is possible to evaluate success and failture respectively of implemented agri-environmental policy measures.
The high nature value (HNV) farmland indicator is one of 35 indicators that incorporate environmental concerns into the EU Common Agricultural Policy. It is an ‘objective-related’ baseline indicator under the EAFRD Implementing Regulation ( Regulation No 1974/2006/EC, Annex VIII), where it is defined as one of three biodiversity indicators in Axis 2 (Improving the Environment and the Countryside).
The indicator is also included in the indicator set for the German National Strategy on Biological Diversity and in the German Länder’s core indicator set (LIKI).
As an objective-related baseline indicator, the HNV farmland indicator must be reported to the EU by all member states under Regulation No 1698/2005/EC (the EAFRD Regulation). At national level in Germany, it is one of the reporting requirements for the National Strategy on Biological Diversity. By agreement between the Federal Ministry of Food, Agriculture and Consumer Protection (BMELV), the Federal Environment Ministry (BMU) and the German Länder, a conceptual design for HNV monitoring was developed by the Federal Agency for Nature Conservation (BfN).
Development of the HNV farmland indicator for Germany started with a research and development (R+D) project in which nationally available data sets relevant to the HNV indicator (data on habitats and habitat types etc.) were identified and compared. The available data sets proved too disparate and incomplete for use in compiling the indicator for Germany. The data had various gaps – failing to cover relevant habitat types such as species-rich arable land and traditionally used orchards – and are gathered too infrequently for regular updating of the index. The R+D project therefore went on to develop a new approach for identifying and monitoring HNV farmland.
Concept and Implementation
The German government and Länder agreed on a new sample-based approach for a nationally uniform HNV monitoring. The sample design is the same as that already used for many years by the Federation of German Avifaunists (DDA) in monitoring common species of breeding birds. Data for the HNV farmland indicator are gathered in field surveys, using some 900 sample plots throughout Germany each covering one kilometre square.
The BfN developed the survey design for an initial nationwide survey of HNV farmland in consultation with the German Länder. Survey instructions (pdf-file, 1,2 MB), define the open countryside structures used in extrapolating the indicator, such as grassland, vineyards, arable land and landscape elements. The nature value of grassland, arable land, fallow land and vineyards is measured against lists of indicator species. In the case of grassland, different lists are used in different regions to reflect geographical variation in species distribution. Evaluation is based on measuring the count of indicator species present in a 30 x 2 m transect within each land parcel. Areas covered by Habitats Directive habitat types and habitats protected by law are per se included if they constitute open countryside habitats on farmland and are assessed according to the standing evaluation proceedure under the Habitat Directive.
The identified HNV farmland structures were assigned nature values on a scale:
HNV I: Exceptionally high nature value
HNV II: Very high nature value
HNV III: Moderately high nature value
So besides quantitative information HNV farmland changes an evaluation of the state of quality and potential quality changes in HNV farmland over time is possible.
The initial survey was carried out jointly by the German government and the Länder in 2009 and coordinated by the BfN ( R+D project FKZ 3508 89 0400, pdf-file, 6,6 MB). Since then the German Länder contract out the surveying of sample plots to experienced field ecologists. Some Länder survey a quarter of their total sampling plots each year while others cover half of their sampling plots every two years. Thus, full coverage of the sample design is achieved after four years. Meanwhile, the respective current value may be calculated as “mean on a gliding scale”. For this, the complete plot setting is being extrapolated, with taking into account the recently mapped plots with the latest data, whereas for all other plots the formerly recorded data is included. This method allows complying with the biennal reporting obligation. BfN collates the Länder data and extrapolates the national indicator value every second year.
To ensure homogenous conditions for the nationwide field mapping, various measures for a quality management are undertaken. BfN arranges training courses on a regular basis to educate the field ecologists. Furthermore a number of recently mapped plots are reviewed within each season. Additionally, all field data is subject to exhausted plausibility checks.
Extrapolation of the survey results is effected both on national level and on Länder level. All ongoing activities are supported by a Federal-Länder committee that meets once a year to exchange experience, discuss methodological problems as they arise and decides on organisazional matters. Coordination remains with BfN. The conceptual design allows implementing methodolocical changes also retrospectively, and so revising initial miscalculations in the further course of mapping. Hence time series can be calculated consistently even if subsequently modifications of the methodology or assessment criteria have been applied. Various modifications of the mapping methods and clarifications of the mapping instructions were applied based on the experiences following the initial survey. Methodological adaptation was conducted without limiting the comparability within the time series. The methodology for extrapolation and calculation of the sampling errors was subject to major alterations. Results of an expert assessment (pdf-file, 190 KB) on the extrapolation methodology that had been rendered by Prof. Dr. Saborowski from the Department Ecoinformatics, Biometrics and Forest Growth of the Georg-August University of Göttingen suggested an improvement of the extrapolation algorhithm and the consideration of occuring covariances when calculating the sampling error. These modifications led to alterations compared to formerly reported values.
As at March 2012 data for the total survey of 2009 and the subsequent mappings of 2010 and 2011 are complete and available. The subsequent value was calculated on basis of the “mean on a gliding scale” (see above). On national level this adds up to two data points with the following values (HNV farmland percentage of agricultural area and absolute amount):
|Year||HNV rel.||Sampling error||HNV abs. (ha)||Sampling error|
|2009||13,2 %||± 0,5%||2.593.461||± 12.967|
|2011||12,1 %||± 0,4%||2.380.387||± 9.522|
The proportion of HNV is unequally distributed with regard to the quality levels:
Thus, the indicator value tends to decrease with the highest decline in the lowest level. The highest quality level remains static with a low value. A more differentiating view on the gained data illustrates that the decrease is mainly caused by loss of HNV grassland, while in e.g. landscape elements, no noteworthy changes occur.
It should be noted that as to half of the sampling plot areas that have been surveyed for the second time in 2012 or will be in 2013, results of the subsequent survey as well as retrospective corrections of the 2009 values have not been taken into account yet. Thus, indicator values of 2009 still might slightly change after the second survey will be complete at the end of the year 2013.
The practised methodology turned out to deliver statistically sound results even exercised biennially. Following integration of survey data of 2013 a series of two nationwide complete and reliable indicator values will be available at the end of the programming period. Thereby, in an economical manner the HNV indicator supplies solid data on status and development of biological diversity in the agricultural landscape and contributes essentially to the evaluation of national and European agricultural policy measures.
Furthermore, the conceptual design offers an additional potential for analysis, as the data allows a more differentiated view. Qualitative changes in the HNV setting can be assessed in time by addressing the different quality levels separately in their temporal dynamics as well as trends for qualitative development of HNV types as e.g. grassland, fallow land or arable land. Merely in cases of rare HNV types that cannot be assessed in sufficient numbers to guarantee statistical soundness this potential reaches its limit.
In this context it is to appreciate, that some Länder have already enlarged their subsample in line with the sampling concept. This provides a benefit on both, the national and the Länder level, whilst a deeper analysis is possible as well as the sampling error could be minimized in order to identify trends as soon as possible.
Additionally, correlation of HNV data and other biodiversity data sets is possible and is recently tested in several research projects.
HNV monitoring therefore offers a new, valuable data basis with a high potential for various advanced research approaches and queries on biological diversity within the agricultural landscape.
Benzler, A. (2009): The implementation of the HNV farmland indicator in Germany. - Rural Evaluation News 2: 4-5.
Benzler, A. (2012): Measuring extent and quality of HNV farmland in Germany. - In: Oppermann, R.; Beaufoy, G. & Jones, G. (Eds.): High Nature Value Farming in Europe. S. 507-510. - Ubstadt-Weiher (Verlag Regionalkultur).