Simulate population dynamics and survey data that respond to stream hydrology.
The R package does the following steps:
- Acquire hydrology time series data from a user-defined stream with a USGS stream gauge
- Calculate four hydrologic metrics describing mean, variance, high and low flow characteristics.
- Allow user to define correlations between hydrologic metrics and population parameters: carrying capacity and intrinsic growth rate.
- Simulate a population time series in the selected stream using a multi-population viability analysis model (MPVA; Leasure et al 2019 Ecology).
- Simulate field survey data from this population using the sampling and observation models from MPVA.
The user can try to recover the known flow-ecology relationships from the simulated survey data using any analysis method.
Simulations can be repeated many times to calculate type I, II, and III error rates.
See ./demo/
for example workflows, starting with ./demo/eflowsim_example.R
.