Geographic risk assessment of Batrachochytrium salamandrivorans invasion in Costa Rica as a means of informing emergence management and mitigation

Abstract

Remotely-sensed risk assessments of emerging, invasive pathogens are key to targeted surveillance and outbreak responses. The recent emergence and spread of the fungal pathogen, Batrachochytrium salamandrivorans (Bsal), in Europe has negatively impacted multiple salamander species. Scholars and practitioners are increasingly concerned about the potential consequences of this lethal pathogen in the Americas, where salamander biodiversity is higher than anywhere else in the world. Although Bsal has not yet been detected in the Americas, certain countries have already proactively implemented monitoring and detection plans in order to identify areas of greatest concern and enable efficient contingency planning in the event of pathogen detection. To predict areas in Costa Rica with a high Bsal transmission risk, we employed ecological niche modeling combined with biodiversity and tourist visitation data to ascertain the specific risk to a country with world renowned biodiversity. Our findings indicate that approximately 23% of Costa Rica's landmass provides suitable conditions for Bsal, posing a threat to 37 salamander species. The Central and Talamanca mountain ranges, in particular, have habitats predicted to be highly suitable for the pathogen. To facilitate monitoring and mitigation efforts, we identified eight specific protected areas that we believe are at the greatest risk due to a combination of high biodiversity, tourist visitation, and suitable habitat for Bsal. We advise regular monitoring utilizing remotely-sensed data and ecological niche modeling to effectively target in-situ surveillance and as places begin implementing educational efforts.

Fig 1. Process of identifying high priority monitoring areas.

Conceptual diagram showing the data, tools, and general processes used to identify areas highly suitable for Bsal that also have high levels of amphibian diversity. Circles indicate input data, outputs are outlined by squares, and final outputs are outlined by square dotted lines. Components of biodiversity modeling using amphibian data are shown with green gradient and components of habitat suitability mapping of Bsal are shown with blue gradient.

Fig 2. 2018 Human visitation in Costa Rica protected areas.

Protected areas include wildlife refuges, biological reserves, national parks, and world heritage sites. Protected areas are outlined in black. Where available, visitation numbers within protected areas for 2018 are indicated by color, with darker blue indicating more visitors. Insert map shows the geographic location of Costa Rica as a land bridge between North and South America.

Fig 3. Costa Rican salamander alpha diversity.

These maps denote alpha diversity of salamanders in Costa Rica. Darker shades indicate higher alpha diversity. Alpha diversity was calculated by summing the total number of salamander species in each ~1 km by 1 km pixel.

Fig 4. Predicted Bsal suitability in Costa Rica.

Higher suitability is indicated by reds and oranges and can be observed in the Cordillera Central and Cordillera de Talamanca mountain ranges, as well as along the northeastern Caribbean slope. Lower suitability is indicated by blues and greens.

Fig 5. Predicted Bsal suitability categorized into high, medium, and low.

Protected areas are indicated by black diagonal lines. Insert map shows all of Costa Rica.

Fig 6. Areas of overlap between predicted moderate and high Bsal suitability and high and very high salamander diversity.

Areas of both Bsal moderate suitability and high/very high salamander diversity are shown in yellow. Areas highly suitable for Bsal with high/very high salamander diversity are shown in red. Protected areas are outlined and the number of visitors in 2018 is indicated by the pattern in the legend.

Fig 7. Areas considered high priority for monitoring for Bsal introduction.

Protected priority areas outlined in black and identified by name overlaid on Ecological Niche Model results for Bsal suitability. Areas of warmer colors (reds and oranges) indicate high Bsal suitability, whereas cooler colors indicate lower suitability.