A stochastic hydroecological metapopulation model for Litoria raniformis which explores the role of environmental water provision in the probability of persistence over a 60 year time step

These files underpin the stochastic hydroecological metapopulation model presented in the article "Stochastic metapopulation dynamics of a threatened amphibian to guide optimal water delivery"

AUTHORS: Rupert Mathwin, Matt S. Gibbs and Corey J. A. Bradshaw

CONTACT: rupert.mathwin.ecology@gmail.com

URL: http://GlobalEcologyFlinders.com

INSTITUTION: Flinders University

INSTITUTION: Rupert.Mathwin.Ecology

RELEASE DATE: May 2022

R code accompanies article:

Mathwin, R, Wassens, S, Gibbs, MS, Young, J, Ye, Q, Saltré, F, and Bradshaw, CJA Stochastic metapopulation dynamics of a threatened amphibian to guide optimal water delivery in review

AIM: This stochastic hydroecological metapopulation model examines the population viability of Litoria raniformis across a metapopulation of 23 wetlands between Locks 3 and 2 in southern Australia's Murray-Darling Basin. We test five competing conservation treatments which manipulate hydroperiod. We then reexamine these treatments under catastrophic drought conditions.

https://github.com/RupertLovesEcology/Lraniformis-Metapopulation-model/blob/main/OrderedAnnualWetnessDrought2.csv

Repository includes the following files:

• 'Lraniformis_Metapopulation_Model_V12.R' — R code to run the metapopulation model.
• 'Lraniformis_Metapopulation_globalSensitivity_V12.R' — R code to run the global sensitivity analysis on the metapopulation model (uses latin hypercube resampling to populate a boosted regression tree).
• 'OrderedAnnualWetness.csv' — .csv containing the summed spring and winter river heights as a proxy for wetness. Determines if the wetland remains full between years (and accumulates aquatic predators).
• 'OrderedAnnualWetnessDrought2.csv' — .csv containing the summed spring and winter river heights as a proxy for wetness. Determines if the wetland remains full between years (and accumulates aquatic predators). This treatment includes a second severe drought.
• 'OrderedHydroDrought2.csv' — .csv containing the 10th highest daily river height which determines wetland filling throughout the reach. This treatment includes a second severe drought.
• 'OrderedHydroForecast.csv.csv' — .csv containing the 10th highest daily river height which determines wetland filling throughout the reach.
• 'PostDrought.csv' — .csv containing the 10000 starting scenarios for the reach (23 populations with age demography).
• 'matrixOperators.R' — functions to manipulate matrix models
• 'movementType.csv' — .csv which has classified every possible single dispersal between 2 wetlands into one of five journey types. This is used to assign landscape resistance.
• 'stayWet.csv' — .csv containing the wetness threshold that each wetland requires to remian wet between years (verified from historic satelite data).
• 'wetlandMetadataV2.csv' — column 1 is the size category (which determines the local popualtion capacity) and sill height is the river height required to start filling the wetland
• 'wetlandMovement.csv' — .csv which measures every possible single dispersal between 2 wetlands following several movement rules (e.g. can only cross the river once).