Artificial nest sites in conservation: reproductive use of bricks and plastics by a critically endangered spring goby

The Italian spring goby Orsinigobius punctatissimus (Canestrini, 1864) is a critically endangered freshwater fish endemic to the Padanian plain in northern Italy. This small species (<60 mm) inhabits spring-fed environments such as resurgences and oxbow lakes and reproduces mainly in spring and summer. Spawning occurs under suitable substrates like stones, branches, and leaves, where females attach 100–300 eggs (2 – 3 mm diameter) to the nest ceiling while positioned upside down, followed by fertilization and prolonged parental care by the male until hatching. Embryonic development lasts 10–12 days at 18–20 °C and up to 20–25 days at 13–14 °C (Gandolfi et al., 1985; Lugli et al., 1997; Vanni et al., 2019; Zanetti et al., 2019). Over recent decades, the species has experienced a >70% reduction in its range within Lombardy (LIFE Gestire 2020), driven by habitat loss, pollution, spring droughts, and climate change. These pressures have fragmented habitats and reduced the availability of reproductive substrates, thereby impairing reproduction and accelerating population decline. Moreover, springs and resurgences are increasingly exposed to plastic contamination, especially near agricultural and urbanized areas. While most studies on macroplastic (>1 cm) impacts in freshwater ecosystems emphasize ingestion, especially in fishes (Azevedo-Santos et al., 2021), recent evidence indicates that plastic debris can also be used as refuge or nesting substrate (Jagiello et al., 2024). During a monitoring campaign (2022–2025), we documented plastic debris being used as nest sites by O. punctatissimus, suggesting an opportunistic exploitation of any available cavities and emphasizing the critical importance of nest availability and microhabitat diversity for this species. Based on these observations, we hypothesized that improving the availability of suitable artificial substrates could enhance reproductive success and support population recovery. The first aim of this study was to test whether O. punctatissimus shows a preference for specific types of macroplastic as nesting sites. To do so, we systematically collected all macroplastic items from ten reproductive sites across the Adda, Oglio, and Po river basins. Sampling was conducted during the peak breeding season (April–May 2025) in known spawning areas, mainly springheads. Each item was photographed, categorized by type and color, and examined for the presence of eggs. We classified samples into two groups: hunting shotshells (with a small cavity) and fragments/objects (without cavities). A binomial generalized linear model (GLM) with logit link was used to model the probability of egg presence, and a Fisher's Exact Test was performed to validate results given the limited sample size. Odds ratios with 95% confidence intervals quantified effect size. The second aim was to evaluate the effectiveness of artificial reproductive substrates. We employed 134 hollow bricks (8–20 per site) at eight springs hosting populations. Bricks had different hole sizes (small, medium, large) and were arranged within 4 × 5 m cells: empty cells, low-density cells (1 brick), and high-density cells (3 bricks), with an average density of 1 brick/20 m². Sites were monitored monthly to record reproductive use of bricks, collect environmental data, and document usage by other species, notably the invasive red swamp crayfish (Procambarus clarkii Girard, 1852) and the Padanian goby (Padogobius bonelli Bonaparte, 1846). A binomial generalized additive model (GAM) with logit link tested brick preference, and an Analysis of variance (ANOVA) assessed changes in fish density before versus after brick placement. Egg presence was negatively correlated with the number of P. clarkii, showed a nonlinear relationship with P. bonelli density and varied significantly by season and basin. Multiple oviposition occurred in the same brick, with a marked preference for small holes. A significant interaction emerged between hole size and P. clarkii abundance: small holes were preferentially used when crayfish were present, likely providing better protection. Mean fish density increased by 0.68 ± 0.47 ind/m² after substrate placement, whereas controls showed no change. These findings demonstrate that P. clarkii exerts negative effects on nest-site selection via spatial competition, while P. bonelli may influence reproduction in a density-dependent manner, potentially shifting from competitor to commensal at higher densities. Hole size preferences suggest a trade-off between clutch size and protection from predators and competitors. The plastic survey yielded 188 items from eight sites, including shotshells (n=65, 34.6%), non-rigid fragments (n=60, 31.2%), rigid fragments (n=33, 17.6%), threads (n=9, 4.8%), and other objects (n=21, 11.2%); 5.9% of items contained eggs, with shotshells used most often (72.7%). Statistical analysis confirmed a strong association between shotshells and egg presence (GLM: OR = 9.75, 95% CI = 2.64–46.69, p = 0.001; Fisher’s test: OR = 9.55, 95% CI = 2.12–59.43, p < 0.001). This preference is likely driven by their small, enclosed cavity that provides protection from crayfish and gobies. Overall, our results highlight that nest-site availability is a critical limiting factor for O. punctatissimus, and that habitat enhancement via provision of suitable artificial substrates can significantly increase local density and support conservation of this endangered species.