From April 2020 till October 2021, we ran a thorough data revision. We have made many improvements to our tools and dataset in this period, which led to improved supranational species indices and indicators. We are happy to announce that the process of data gathering, data control and calculations improved enormously.
- Two new countries, Croatia and Flanders, joined the PECBMS. Hence, we produced indices and trends for 170 common European species based on data from 29 countries. In addition, we newly include data from all three Belgian regions (Brussels, Flanders, Wallonia).
- The number of species published in 2021 remained the same as in the last years.
- Indices of Belgium, Ireland, Luxembourg, Portugal, and Romania, which were not updated last year, are actualised to 2019.
- We could not update indices of France, so we are using data up to the year 2017. We expect to use updated French data in the 2022 update.
- We publish the indices in the downloadable tables with standard errors next to the confidence limits. The reason is to provide standard errors for the researchers who prefer to use them. Of course, confidence limits must be corrected if their value should be negative, but standard errors need no corrections.
- We finished a complete data revision of PECBMS data. Therefore, some countries’ data for particular species has been considered suitable for PECBMS or excluded to improve the index.
- We implemented a new tool (in an R-script) gathering criteria used to select national data. This tool points out the problematic data on a national level to assess its use for supranational index calculation.
- We unified the terminology and assessed the new subspecies or species within 170 PECBMS species. It was the case of Phylloscopus bonelli, Sylvia hortensis and Phylloscopus collybita. There was no need to split one species into two because the data we receive always fall under the original species’ name.
- The indices might have changed because of revised data selection or entering Croatia and Flanders to PECBMS. There has also been thorough data revision of national data in several countries, which caused changes on both – national and supranational levels.
- As usual, two species (Oenanthe cypriaca and Sylvia melanothorax) haven’t been included in the common bird indicators for Europe and the EU, as they are endemic species for Cyprus. Consequently, only 168 species are included in common bird indicators. The number of species included in common farmland and common forest bird indicators for Europe and EU remained unchanged (39 farmland species and 34 forest species).
- We provided national coordinators to use an updated version of a tool to calculate national indices (developed in Statistics Netherlands) called RTRIM-shell. The tool uses RTRIMpackage in R to produce the same outputs as in the previous updates of the indicators. We needed a new tool because we started to use a new tool for supranational index calculation. Moreover, this new tool replaced the old TRIM/BirdStats programme, which became outdated in some processes. Therefore, we plan to use explicitly the RTRIM-shell soon.
- We used an updated version of the MSI tool to calculate supranational species indices (developed in Statistics Netherlands). However, the tool produces the same outputs as in the previous updates.
- We improved the online tool for national and supranational data checks so that national coordinators can use it also for their own data checks. Moreover, we updated the online tool, which compares the consistency of supranational indices and trends.
- We created a new database. Since all the tools for national indices, supranational indices and indicator calculation are R-scripts, there was a need to create a database in R as well. This way, we reached simplification and speeding up of database creation.
The thorough data revision led to:
-
- data improvement on the national level due to comprehensive data assessment and time to discuss with national coordinators
- improved data used for index and indicator calculation
- update of tools used for data processing and calculation
- a thorough revision of the PECBMS candidates; we have already involved Croatia in PECBMS, and we assess data from Russia and Andorra for implementation in the future updates
- simplification and speeding up the process from national index calculation to publishing on PECBMS web
Besides this, computation procedure, data quality control and presented indices, trends and indicators are generally consistent with the update in 2019. All the inconsistencies are justified by data selection and new countries joining PECBMS.