Data for: Modelling the surprising recolonisation of an understudied aquatic mammal in a highly urbanised area: Fortune favored the Smooth-coated otter in Singapore
Cite this dataset
Hughes, Kilian (2023). Data for: Modelling the surprising recolonisation of an understudied aquatic mammal in a highly urbanised area: Fortune favored the Smooth-coated otter in Singapore [Dataset]. Dryad. https://doi.org/10.5061/dryad.j3tx95xm6
Abstract
Ever-growing human activities present an active and continuing threat to many species throughout the world. Nevertheless, concerted conservation efforts in some regions have balanced these threats and allowed endangered species to recolonise former parts of their original ranges and reverse their decline. This is notably the case of the smooth-coated otter (Lutrogale perspicillata). In 1998, individuals returned to Singapore after more than a 20-year absence. In 2017, 79 otters were counted throughout the heavily urbanized city. Despite this comeback, the future of the species in Singapore is unclear. By collating information on the species’ life history traits, we implemented a spatially explicit individual-based model. The model demonstrated that successful establishment of Singapore's population from the initial immigrants was highly uncertain. In 43% of cases, stochastic extinction occurred. From the 9% of model replicates that closely reproduced the observed colonisation history, projections showed that the population would reach close to 200 individuals in 50 years. This study successfully demonstrates the use of individual-based modelling to simulate the inherently stochastic recolonisation dynamics of an endangered species and predict its longer-term future. We discuss emerging issues that may arise from increasing negative interactions between otters and humans and the general challenges associated with rewilding highly urbanized environments. We stress the importance of long-term monitoring surveys and education campaigns to mitigate human-wildlife conflicts. With species and natural habitats increasingly threatened by our ever-growing human expansion, understanding the factors that allow human-dominated landscapes to be compatible with biodiversity is of the utmost importance.
README: Data for: Modelling the surprising recolonisation of an understudied aquatic mammal in a highly urbanised area: Fortune favored the Smooth-coated otter in Singapore
Data for manuscript
List of data sets used for the data analysis for the manuscript:
"Modelling the surprising recolonisation of an understudied aquatic mammal in a highly urbanised area: Fortune favored the Smooth-coated otter in Singapore"
List of files and description:
- Sim50_Parameters
This file details the different species and model parameters that were used as input into the RangeShifter software.
- Sim50_Occupancy
This file contains a list of all the cells in the landscape (x and y coordinates) or of all the patches (PatchID). The remaining columns give the occupancy probability of the cell / patch at defined time steps. The occupancy probability is obtained by dividing the number of times (replicates) that the cell / patch has been occupied in a given year by the total number of replicates.
- Sim50_Occupancy_Stats
This file represents the mean ratio between occupied and suitable cells (Mean_OccupSuit) and its standard error (Std_error) at the set time interval.
- Sim50_Pop
This file represents the population output. It is produced by checking the box Populations in RangeShifter. It contains statistics regarding each population present in the landscape at a given time interval. As for the species’ range output, data are collected before reproduction at each reproductive season at the specified yearly interval and at the end of the simulation.
This output file contains the following information:
1. Replicate number (Rep)
2. Year (Year)
3. Reproductive season within the year (RepSeason)
4. Cell location (x and y coordinates ) or patch ID (PatchID)
5. Species number (Species); not yet used\, always zero
6. Number of individuals in the population (NInd)
7. In the case of a stage-structured population\, the number of individuals in each stage (NInd_stageX). If the reproduction is sexual\, these columns will be replaced by the number of females (Nfemales_stageX) and of males (Nmales_stageX) in each stage. In the case of sexual model without stage structure\, two columns will indicate the number of females (Nfemales) and of males (Nmales) in the population.
8. In the case of a stage-structured population\, the number of juveniles born (NJuvs). If the reproduction is sexual\, these columns will be replaced by the number of females juveniles (NJuvFemales) and males (NJuvMales).
- Sim50_Range
This file represents the species "Range" output. is produced by checking the box Range. Data are written before reproduction at each reproductive season at the specified yearly interval. An extra line is written at the end of the simulation. The file contains the following general information regarding the species’ range:
1. Replicate number (Rep)
2. Year (Year)
3. Reproductive season within the year (RepSeason)
4. Total number of individuals (NInds)
5. Total number of individuals in each stage (NInd_stageX); these columns will be present only in case of stage-structured models
6. Total number of juveniles born (NJuvs); only in case of stage-structured models
7. Total number of cells (NOccupCells) or total number of patches (NOccupPatches) occupied by a population capable of breeding: for a stage-structured population\, individuals of the breeding stage(s) must be present\, and for a sexual model\, both sexes must be present
8. Ratio between occupied and suitable cells or patches (OccupSuit)
9. Species’ range\, in term of maximum and minimum coordinates (min_X\, max_X\, min_Y\, max_Y) of cells / patches occupied by breeding populations (as above)
10. If there are any variable traits\, summary trait phenotypes (mean and standard deviation) across the whole population range will also be included
- Sim50_Patches_Individuals
This file was created to collate the number of individuals in each patch at the end of the simulation accounting for both sub-adults and adults.
Methods
The data was created through the use of an individual-based model named: "RangeShifter". The different data sets show the outputs of the model for the smooth-coated otter species in Singapore. The data shows the increase in population from colonization to 70 years after. Data sets include population growth every 10 years and also habitat patches occupied with number of individuals per patches.
This data was then fed into RStudio for analyses.
Usage notes
The data was was obtained using the program RangeShifter version 2.0 and RStudio version 4.1 was used to open the data files and analyse them.