This README_file.txt was generated on 2022-01-027 by Dumas G‡lvez 1. Title of dataset: Data from: Ecology of fear and its effect on seed dispersal by a neotropical rodent by Dumas G‡lvez and Marisol Hern‡ndez 2. Author information Corresponding investigator Name: Dumas G‡lvez Institutions: Programa Centroamericano de Maestr’a en Entomolog’a, Universidad de Panam‡. Estafeta universitaria, Avenida Sim—n Bol’var, 0824, Panama Smithsonian Tropical Research Institute, Panama COIBA AIP, Panama Sistema Nacional de Investigaci—n, Panama Email: dumas.galvezs@up.ac.pa Name: Marisol Hern‡ndez Institution: Escuela de Biolog’a, Universidad de Panam‡. Email: cmary0419@gmail.com 3. Date of data collection: 2019 - 2021 4. Geographic location of data collection: Gamboa, Parque Natural Metropolitano (Panama) 5. Funding source that supported the collection of the data: Vicerrector’a de Investigaci—n y Postgrado (Universidad de Panam‡). Grant: CUFI-2018-CNET-P-017 6. Recommended citation for this dataset: G‡lvez, Dumas; Hern‡ndez, Marisol (2022), Ecology of fear and its effect on seed dispersal by a neotropical rodent by Dumas G‡lvez and Marisol Hern‡ndez, Dryad, Dataset, https://doi.org/10.5061/dryad.9p8cz8wj2 DATA & FILE OVERVIEW 1. Description of dataset These data set were generated to investigate the effect of ocelot's cues on the seed dispersal and cache pilferage behavior by the Central American agouti in Panama. 2. File List: File 1 Name: Galvez&Hernandez2022a.xlsx File 1 Description: Survival data (seed removal) of seeds removed by agouties in sites with low and high ocelot density. Seeds were identified by group of seeds (set) and individually (seed.ID) File 2 Name: Galvez&Hernandez2022b.xlsx File 2 Description: Survival data (cache removal) of experimental caches removed by agouties in sites with low and high ocelot density. Caches were identified by group of caches (set). File 3 Name: Galvez&Hernandez2022c.xlsx File 3 Description: Dispersal distances of seeds removed by agouties (from File 1) in sites with low and high ocelot density. Besides, together with the angle of dispersion, cache densities were calculated by two methods (cache.density and nnd.mean). File 4 Name: Galvez&Hernandez2022d.xlsx File 4 Description: Survival data (seed removal) of seeds removed by agouties that were placed next to ocelot's cues (O) or under a control condition (C). METHODOLOGICAL INFORMATION FOR Galvez&Hernandez2022a.xlsx AND Galvez&Hernandez2022b.xlsx We placed tagged seeds in order to track them in sites with low or high ocelot density. This allowed us to monitor dispersal rates, dispersal distances and cache spacing. We collected Attalea butyracea fruits directly from the ground or on a tree by using a pole saw. We manually removed the exocarp and kept them in water during 48 hours in order to remove the mesocarp. Then we let the seeds dry under the sun for 48 hours. We then assembled batches of 20 seeds of similar size and each seed was drilled a 1 mm hole in the distal side, to which we attached a 50 cm green nylon thread with a 15 x 2.5 cm pink-colored flagging tape attached to the other end, a standard protocol for quantifying seed dispersal (Jansen et al. 2014). Each seed was identified with a unique identity number which was written on the flagging tape. The batches were placed in grids of 2 x 10 seeds haphazardly and we placed 5 batches per site (low vs high ocelot density) during the rainy (3 October - November 2019) and late rainy season (hereafter: transition, 27 November Đ 26 January 2020, Figure S1). That represents ten replicates per ocelot density treatment (200 seeds per treatment). We monitored the seeds at 1, 2, 3, 4, 7, 9, 18, 28, 38 and 49 days after placement during the rainy season and at 1, 2, 7, 16, 23, 31, 43, 51, 57 and 61 days during the transition. We did not include the dry season since this period is normally out of the fruiting phenology of the species (Adler and Lambert 2008). We searched the surrounding area of the batches for the flagging tape tag up to a distance of 45 m, which revealed the location of the cache and we mapped each cache by using an ultrasonic rangefinder (Haglšf DME 201, Haglšf Inc., Madison, WI, U.S.A.) and a precision compass (Suunto KB-14, Suunto Oy, Vantaa, Finland). With the locations of the caches, we estimated cache spacing by calculating the minimum spanning tree (MST), which is the structure that connects all the points (caches) with the minimum total distances without any cycles. We computed MSTs with the function ComputeMST (library emstreeR in R) and we obtained an averaged nearest-neighbor distance (NND), which was used to compare between sites. Smaller NNDs indicate a higher cache density. As a second estimation of cache spacing, we use a minimum convex polygon (MCP), which encompasses all the mapped caches. Cache density is expressed as the total number of caches divided by the area of the MCP. We and others have used both methods before to estimate caching areas (G‡lvez et al. 2009 and references therein). DATA-SPECIFIC INFORMATION FOR: Galvez&Hernandez2022a.xlsx 1. Number of variables: 6 2. Number of cases/rows: 401 3. Variable List: season: season when seeds were monitored set: set ID of the group of seeds seed.id: ID of individual seeds treatment: level of ocelot density time: time in days status: seed removed (1) or not removed (0) 4. Missing data codes: None DATA-SPECIFIC INFORMATION FOR: Galvez&Hernandez2022b.xlsx 1. Number of variables: 10 2. Number of cases/rows: 400 3. Variable List: season: season when seeds were monitored Plot:evel of ocelot density set: set ID of the group of seeds seed: ID of individual seeds distance: distance in meters to the place where the seed was cached by an agouti azimuth: angle of the direction of the place where a seed was cached by an agouti x: x coordinate of location of a particular cached seed, the (0,0) coordinate is the place where the set of seeds was initially placed. y: y coordinate of location of a particular cached seed cache.density: cache density obtained as the number of caches divided by the area on minimum convex polygon that contain all the caches. nnd.mean: mean nearest neighbor distance between caches made by agouties. 4. Missing data codes: NA ETHODOLOGICAL INFORMATION FOR Galvez&Hernandez2022c.xlsx In order to quantify pilferage rates in sites with low or high ocelot density, we mimicked agouti caches by burying seeds approximately 5 cm below the ground surface. Caches in this experiment were placed five meters from batches in experiment 1. We created one 2 x 5 grid of caches per site separated by 2.5 m, similar to previous protocols (G‡lvez et al. 2009; Jansen et al. 2014). We attached a 60 cm green nylon thread to the seed as described before but the free end was attached to the nearest sapling or tree. This allowed us to slightly pull the cache to check whether the seed was still present. We monitored the caches during the same times that we checked the seeds in the rainy and transition period (experiment 1, see above); besides we checked them at 1, 2, 3, 8, 17, 23, 32, 45, 52 and 63 days during the dry season (29 January Đ 13 March 2019), a period of food scarcity when agoutis rely more on caches (G‡lvez et al. 2009; Emsens et al. 2013). We placed 5 batches per season per ocelot density site, which represents 15 replicates per ocelot density treatment (300 seeds per treatment). DATA-SPECIFIC INFORMATION FOR: Galvez&Hernandez2022c.xlsx 1. Number of variables: 5 2. Number of cases/rows: 300 3. Variable List: season: season when seeds were monitored set: set ID of the group of seeds treatment: level of ocelot density time: time in days status: seed removed (1) or not removed (0) 4. Missing data codes: None ETHODOLOGICAL INFORMATION FOR Galvez&Hernandez2022d.xlsx To corroborate our seed dispersal and cache pilferage experiments, we carried out a third experiment (9 February until 24 February 2021) in which we placed urine and feces of ocelots near the batch of seeds, to monitor dispersal rates. We collected urine samples from one female ocelot in the Parque Municipal Summit by placing 2 x 25 cm strips of fabric placed inside a flat cage (0.5 x 5 x 30 cm) in the corner of the enclosure that the ocelot used for urination and defecation. The feces samples were collected manually from the same corner. We used urine and feces samples for the experiment the same day of collection. We marked the seeds as described before and placed the batches haphazardly in grids of 2 x 10 seeds. For the batches with the ocelot cue, we rolled one strip of fabric impregnated with urine around a 15 cm stick that was placed vertically at 5 cm from the seeds. We also placed next to the stick approximately 25 cm3 of ocelot feces. For the control, we only placed the stick without an odor cue at the same distance from the seeds. We added the same amount of feces at day 1, 2 and 3; and urine at day 6 after placement of the seeds. We placed a batch every 150 meters, intercalating the type of batch (Figure S2). Moreover, since rodents can be more likely to move seeds away from risky patches before eating them (Lichti et al. 2017), we monitored the seed handling time of agoutis by placing camera traps in front of the seeds in seven of the replicates from each treatment. For this, we recorded the time at which an agouti took a seed until the time that it took one last seed in a feeding bout. We considered recordings as independent if more than ten minutes passed after the last seed was taken and we averaged those handling times per camera. We did not include spatial analysis of this experiment since agoutis and leafcutter ants removed the thread or flagging tape from a large number of seeds, making it difficult to map them after removal. We only found 36 out of 142 and 35 out 155 removed seeds for the ocelotsŐ cue and controls, respectively. DATA-SPECIFIC INFORMATION FOR: Galvez&Hernandez2022d.xlsx 1. Number of variables: 4 2. Number of cases/rows: 398 3. Variable List: treatment: seeds placed to ocelot's cue (O) or control (C) set: set ID of the group of seeds time: time in days status: seed removed (1) or not removed (0) 4. Missing data codes: None