The online portal www.ornitho.ch served as the data collection centre for the 2013–2016 atlas, enabling users to view the current state of progress for every atlas square and species. The species-specific inclusion criteria involved a minimum Atlas Code and a reference date, and in many cases matched the criteria used in the 1993–1996 atlas. Priorities were defined for fieldwork in each field season.
The 2013–2016 atlas builds on data collected in four field seasons from 2013 to 2016. Additional data gathered in 56 kilometre squares in 2012 for the federal scheme «Biodiversity Monitoring Switzerland (BDM)» were also incorporated. One BDM survey round lasts five years (with one fifth of the total number of squares surveyed each year), so we decided to integrate an entire period from this monitoring project.
Compared to the 1993–1996 atlas, collecting records was much easier in 2013–2016. During 1993–1996, volunteers had to record observations on special paper forms or A6-size recording cards, look up the geographic coordinates, and send the documents to the Swiss Ornithological Institute by post. Observers also had access to a programme called «IDEXT», made available back in 1989 to facilitate the recording process. For the 2013–2016 atlas, records were entered on www.ornitho.ch, often including information on the exact location. Looking up coordinates was no longer necessary thanks to the map background available in the online system. «NaturaList» became available in 2015, an app that allowed georeferenced records to be entered directly in the field (only for Android smartphones at the time).
The 2013–2016 atlas was the first atlas to be developed «live». To make this possible, a special atlas tool was added to www.ornitho.ch. The new addition allowed users to check the progress of fieldwork anytime. Thanks to real-time updates of maps and atlas-square overviews, users could check, for example, which species had already been detected by other observers. This saved hours of unnecessary search effort. Moreover, the standardised entry procedure eliminated many sources of error. For example, required data fields on the online entry form made sure that important information (e.g. Atlas Codes) was not forgotten. To ensure that the real-time maps and atlas-square overviews were reliable at all times, additional effort was required on the part of the Swiss Ornithological Institute to check entries on an ongoing basis (and not only upon completion of fieldwork). Protected data (e.g. observations of sensitive species) remained hidden in the new atlas tool and could not be viewed publicly. The restriction also applied to the designated observers responsible for an atlas square and their collaborators. The only information publicly available on these species was whether they had already been recorded in the atlas square or not.
Regular email updates and the regional atlas meetings helped to elicit a strong response, and recording activity was high from the beginning, including in peripheral regions.
Special atlas features on www.ornitho.ch
The atlas tool on www.ornitho.ch allowed users to consult the map showing the number of recorded species per atlas square as well as the atlas square maps for individual species for the 2013–2016 atlas and the earlier atlases (1972–1976 and 1993–1996). The records used to generate the maps could also be accessed with ease. Comparison maps at atlas-square resolution illustrated the change in a species’ distribution between two survey periods. A list of contact persons and collaborators for each atlas square was provided, as well as a direct link to «Terrimap online».
Species-specific inclusion criteria
A minimum Atlas Code (AC) and a reference date were defined as inclusion criteria for each species. For most species, they were the same as in 1993–1996. Defining a reference date had proven useful in 1993–1996, as it provided a clear rule that helped to exclude many controversial records, most of them relating to late passage migrants. Observations made after the reference date in suitable habitat were included. We also accepted indirect evidence such as droppings or feathers, if they were attributed to a species with certainty by specialists. Observations of displays in autumn, however, were not accepted.
For most «widespread species», 15 April was determined as the reference date (except in 2013: 13 April). This was the earliest possible date for territory mapping surveys. An earlier date was determined for resident species (e.g. woodpeckers, grouse) and a later one for late migrants. All observations had to occur in suitable habitat (AC of at least 2). Observations that strongly indicated breeding (≥ AC 7) were included even if they were made before the reference date. On the other hand, observations that occurred after the reference date but that probably concerned late migrants were not included. Additional criteria had to be met for Eurasian Siskins below 1000 m; only records that strongly indicated breeding (≥ AC 7) were included.
Criteria for the categories «rare species» and «rare species on the Central Plateau and in the Jura» varied. For many species, an observation during the breeding season in suitable habitat (≥ AC 2) after the reference date was sufficient. For some species (e.g. Western Yellow Wagtail), the criteria of probable breeding (≥ AC 4) had to be met. For rare breeders (e.g. Black-crowned Night-Heron) and some other species (e.g. Goosander), observations that strongly indicated breeding (≥ AC 7) were required. The same applied to Northern Wheatear and Redpoll below 1000 m (at higher altitudes, ≥ AC 2 was sufficient). For most waterbirds and for the White Stork, confirmation of breeding was required (≥ AC 11), as they often occur during the breeding season without necessarily attempting to breed. For example, valid observations included a nest containing eggs, an incubating adult, or a family with young not yet capable of flying. No reference date was fixed for species requiring an Atlas Code of 7 or 11.
In the case of «colonial species», there had to be strong indication of breeding (≥ AC 7, e.g. Northern Lapwing, Alpine Swift, Eurasian Jackdaw) or confirmed breeding (≥ AC 11, e.g. Great Cormorant, Collared Sand Martin). «Colonial species in settlements» (Northern House Martin and Common Swift) were subject to the reference date of 15 April and an Atlas Code of at least 2.
Species list and «mini-atlas» per atlas square
One aim of the 2013–2016 atlas was to record all breeding bird species present in a given atlas square. The species list from the 1993–1996 atlas served as a reference, indicating which species were to be expected. After every field season, the designated observers responsible for an atlas square and their collaborators received an email with the updated species list for their square and a comparison with 1993–1996. They therefore knew prior to the next breeding season which species were still missing, for which species further observations were desirable, and which species had been confirmed or recorded for the first time.
Leading up to the 2015 field season, the first «mini-atlas» in PDF-format was produced for the observers in charge of an atlas square. The mini-atlas plotted the recent records of each species as well as those from 1993–1996 on a map of the atlas square. In squares with low coverage, the records from 1997–2012 were included as an additional reference, enabling a targeted search for species that had not yet been confirmed. A further three maps displaying the number of breeding species, the number of records, and the number of observation days per kilometre square illustrated the progress that had been made so far in the atlas square. Further mini-atlases in updated form were provided in early 2016 and upon completion of fieldwork in early 2017. Together with the special atlas features on www.ornitho.ch and the annual species list, the mini-atlas provided valuable support for subsequent fieldwork. By showing the records from 1993–1996 as well as the current ones, it motivated observers to continue their search for missing species. Thus, the mini-atlas almost certainly contributed to the excellent level of coverage in many atlas squares.
Weather conditions in the four field seasons
Of the four field seasons, only the spring of 2014 was mostly warm. The other three seasons were cooler and/or more humid than the long-term average, with severe cold spells in April 2013 and 2016.
In 2013, many lowland areas experienced the coldest March since 1987. From mid-March to early April, the weather was cold with snow in the lowlands. Temperatures plummeted again in the second half of April, with snowfall below 1000 m, especially in western Switzerland. May was the coolest since 1991. It was also wet in many places, with a storm bringing record amounts of rain towards the end of the month. Summer weather gradually arrived in June.
The second year saw early warm and sunny weather in March. April was also milder than average throughout Switzerland, with above-average sunshine especially on the Central Plateau and in Ticino. In May, the weather was cooler and changeable, with just a few warm and sunny days towards the end of the month. June was again warmer than average and very dry in Valais and northern Switzerland.
In 2015, breeding activity got off to a fast start thanks to mild conditions in April, despite a brief resurgence of winter weather in late March. In early May, heavy rainfall caused water levels on lakes and rivers to rise to record levels. Further precipitation followed, making it the wettest May since records began in 1864. June, in contrast, was the fourth warmest on record in many places. The warm weather continued into July, and some places experienced the hottest July on record. Especially the first week of July saw record temperatures.
The weather was at its most contrary in 2016: A very mild winter was followed by overly cool temperatures in March. Temperatures in April were average, despite several severe cold snaps, but there was more rain than usual, both in April and in May. Some places experienced the wettest May since records began in 1864. Snow fell to below 1000 m in late May, and 20–30 cm of fresh snow accumulated at higher altitudes. June was also dull and rainy. Some locations recorded one of the wettest months of June. Water levels were high on lakes and rivers, with some flooding. Only a few days in the second half of the month reached summer temperatures.
For some species, the minimum criteria were insufficient to obtain a satisfactory picture of breeding distribution. The reason, in most cases, was the presence of passage migrants, wandering summer visitors or non-breeders, especially in areas at the edge of the known breeding range. Additional criteria were defined for these species (e.g. Red Kite, Common Kingfisher, Eurasian Golden Oriole, Willow Warbler, European Pied Flycatcher), involving, for example, a higher Atlas Code (≥ AC 7), a later reference date, the prolonged presence of a singing male, or presence at the same site across several years. The additional criteria ensured that these observations could be assessed consistently by the atlas team.
Priorities in the four field seasons
Across all four years, the survey effort centred around the species search in the atlas squares and the territory mapping in the kilometre squares. Starting with the second field season (2014), we undertook additional targeted surveys.
In 2014, the aim was to survey colonial species as comprehensively as possible. A special focus was given to Grey Heron, Alpine Swift, Collared Sand Martin, Eurasian Jackdaw and Rook.
In 2015, Mute Swan, Goosander, Great Crested Grebe, Yellow-legged Gull and Common Kingfisher were surveyed on large lakes and rivers. In many places, these counts were carried out in mid-May, mostly by ornithological working groups or by fieldworkers contracted by the Swiss Ornithological Institute (some stretches of river were not surveyed until 2016). However, early and exceptional flooding interfered with the surveys. A further focus in 2015 was on counting displaying Eurasian Woodcocks, a species for which considerable gaps in coverage existed in the cantons of Valais, Uri, Grisons and Ticino. A spatial expert model based on habitat variables was developed for this purpose. The model identified areas of potentially suitable habitat for the Woodcock, which could then be targeted in the surveys.
Finally, in 2016, efforts focussed on surveying Common Swift and Northern House Martin colonies with more than ten breeding pairs, provided this had not already been accomplished in previous years. Also, contracted fieldworkers supported the territory mapping of Woodlarks in the Jura and conducted a systematic survey of Western Yellow Wagtail populations in northeastern Switzerland, western Switzerland and Ticino.
In all four field seasons, contracted fieldworkers, interns and persons on civilian service duty were enlisted besides the volunteer observers to survey remote kilometre squares, visit underrepresented atlas squares, search for nocturnal birds or other elusive species, and fill in any other gaps.