1. Ophiocordyceps fungi manipulate the behavior of their ant hosts to produce a summit disease phenotype, thereby establishing infected ant cadavers onto vegetation at elevated positions suitable for fungal growth and transmission. Multiple environmental and ecological factors have been proposed to shape the timing, positioning, and outcome of these manipulations.

2. We conducted a long-term field study of Ophiocordyceps camponoti-floridani infections of Camponotus floridanus ants – the Florida zombie ants. We propose and refine hypotheses on the factors that shape infection outcomes by tracking the occurrence of fungal growth from hundreds of ant cadavers. We modeled and report these data in relation to weather, light, vegetation, and attack by mycoparasites.

3. We investigated environmental factors that could affect the occurrence and location of newly manipulated ant cadavers. New cadavers were positively correlated with epiphytic Tillandsia bromeliads, canopy openness, and weather conditions (an interactive effect of temperature, humidity, and precipitation) with an increased occurrence during the sub-tropical summer. We further suggest that incident light at the individual cadaver level may reflect microhabitat choice by manipulated ants or selective pressure on cadaver maintenance for conditions improving fungal survival.

4. We also sought to connect fungal fitness to environmental conditions. Continued fungal development of reproductive structures and putative transmission increased with moist weather conditions (interaction of humidity and precipitation) and canopy openness, while being reduced by attack by mycoparasites. Moreover, under the most open canopy conditions, we found an atypical Ophiocordyceps growth morphology that could represent a plastic response to conditions influenced by high light levels.

5. Taken together, we explore general trends and the effects of various ecological conditions on host and parasite disease outcomes in the Florida zombie ant system. These insights from the field can be used to inform experimental laboratory setups that directly test the effects of biotic and abiotic factors on fungus-ant interactions or aim to uncover underlying molecular mechanisms.

We collected data on O. camponoti-floridani-infected C. floridanus from three wilderness areas in Seminole County, Florida, USA between October 2018 and November 2019. We scouted each wilderness area to find Ophiocordyceps graveyards with few to no observable ant cadavers in the forest between them.

We collected a diverse range of data over time to identify potential drivers of cadaver occurrence and placement, and, subsequent fungal development and transmission. We monitored graveyards over the course of 411 days with a median interval of 60 days between visits to the same graveyard. We characterized cadavers by their biting vegetation, ant caste, height above the forest floor, and, for a subset, incident light levels on the cadaver.

To collect vegetation data for each graveyard, we conducted three line-intersect surveys spaced at even intervals perpendicular to the adjacent trail and spanning the 9 m length of the plot. Our line-intersect transects were two-dimensional as we observed variation in vegetation types at different heights. Along the 9 m length of each transect we measured vegetation between the 25^th quartile (81 cm) and the 75^th quartile value (169 cm) of all cadaver occurrence data. Pieces of vegetation of the same category were measured as continuous if the linear distance in the plane of the transect was ≤ 5 cm.

We obtained weather data from the National Centers for Environmental Information (NCEI), Land-Based Station WBAN:12815.

Microsoft Excel.