Power analysis for a potential white-backed woodpecker monitoring design in Switzerland

The white-backed woodpecker (Dendrocops leucotos) is the rarest and most endangered woodpecker species in Switzerland and Europe (Aangelstam & Mikusinski, 1994; Frank, 2002). Its global range spans the Palearctic region from the Baltic states to the southeastern-most part of Russia, Kamchatka, and Japan. However, in central and western Europe, the distribution becomes very fragmented with patchy populations in southern Scandinavia, the Alps, the Balkans, Italy, and the Pyrenees. Within the Alpine range, the species is found from Austria to Liechtenstein and very locally in the easternmost parts of Switzerland.
This woodpecker is a food specialist, and its presence is strongly conditioned by high abundance of xylobiont insects, and more particularly the larvae of longhorn beetles and jewel beetles (Cerambycidae, Buprestidae), which feed on dead wood (Bauer et al., 2005). In the Alps, this woodpecker primarily inhabits mixed spruce-beech forests at altitudes between 500 and 1500 meters above sea level. Intensive forestry for centuries resulted in the loss of such mature habitats. Effectively, this led to the displacement of the species from Switzerland and a large part of the Alps back to remote and inaccessible areas (Carlson, 2000; Czeszczewik & Walankiewicz, 2006). At present, in most of Europe, the species finds adequate habitat in protected areas exempt from timber harvesting. In recent decades, in the Alpine region, the species has been gradually expanding its range westwards by colonizing sustainably managed commercial forests (Mollet et al., 2009). The species was rediscovered in the Principality of Liechtenstein in 1981, and has been recorded in easternmost Switzerland since 1996, with 526 records in 236 different 1×1 squares according to the ornitho database (Knaus, 1997; Figure 1).
Due to its rarity, specialized ecological needs and low population density, the white-backed woodpecker is highly sensitive to disturbance and eventually extinction, which justifies the monitoring of the population trends in this species. However, its rarity presents challenges in gathering sufficient monitoring data. Statistical models based on limited or low-quality data may produce inaccurate population estimates, thus limiting the ability to reliably assess population trends or spatial dynamics (Wisz et al., 2008). Therefore, selecting effective monitoring strategies that ensure the collection of large enough, high-quality datasets is essential for the conservation of this species. In this respect, simulation-based approaches are valuable tools for planning and optimizing monitoring designs, as they allow the evaluation of different scenarios without the need for their implementation in real life, thus avoiding expending large amounts of economic and human resources for years without any guarantee of success (Nielsen et al., 2009).
Here, we analyzed the potential effectiveness of a proposed structured monitoring design for the white-backed woodpecker in Switzerland in capturing the trends of the study population through time and across space. To do so, we simulated site occupancy (or presence/absence) data of the species at a 1 km x 1 km resolution and modelled an observation process comprising non-detection errors under the proposed monitoring design. Then, we analyzed the resulting detection/nondetection data using logistic regression models to evaluate how powerful the analytical methods and the proposed monitoring design are at retrieving the trends of the Swiss population over time and space. The main aim of the study is to provide evidence to support or refine the proposed monitoring design. In addition, we provide specific advice on data formatting and analysis settings.