Associate Scientist (Academic Research),

Civitello Lab,

Department of Biology, Emory University,

Atlanta GA

Email:

Website: https://avbelsare.netlify.app


OvCWD Modeling Framework comprises of three agent-based models, OvPOP (Odocoileus virginianus POPulation simulation model), OvPOPsurveillance, and OvCWDdy (Odocoileus virginianus Chronic Wasting Disease dynamics model). The models are coded in NetLogo, a Java based modeling environment. The following models have been adapted for simulating CWD spread in Indiana white-tailed deer populations.

INOvPOP: https://www.comses.net/codebases/21af8266-f25f-4567-a3e4-cff04be0d806/releases/1.0.0/

Recommended citation for INOvPOP: Aniruddha Belsare (2022, June 01). “INOvPOP” (Version 1.0.0). CoMSES Computational Model Library. Retrieved from: https://doi.org/10.25937/qy95-2d14

INOvCWD: https://www.comses.net/codebases/5b8ed3c8-d34e-44c0-b373-84ec43984279/releases/1.0.0/

Recommended citation for INOvCWD: Aniruddha Belsare (2022, June 01). “INOvCWD” (Version 1.0.0). CoMSES Computational Model Library. Retrieved from: https://doi.org/10.25937/wx4k-wv30



OvPOP model is parameterized using current best information available for the region of interest, like demographic parameters and harvest data. This step is followed by model calibration: an iterative process of fine tuning model parameters so as to adapt the results of simulation models to actual data. The objective of OvPOP calibration is to generate a realistic, in silico white-tailed deer population for the region of interest.

For each calibration iteration, run 10 iterations of OvPOP.

Calibration run 1:

OvPOP output analysis: Use Shiny App ‘ShinyOvCWDAnalysis’ https://anyadoc.shinyapps.io/ShinyOvCWDAnalysis/ The ‘results’ folder will have two files titled ‘sa[CountyName].csv’ and ‘deerpopdy[CountyName].csv’ respectively. Upload these files to the Shiny App using the buttons provided to generate following plots and tables.

  1. Estimate model deer population growth rate. Use sa[CountyName].csv for this analysis.
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'

  1. Compare model deer population abundance with field estimates.Use sa[CountyName].csv for this analysis.Also, update INDNR estimate (mean N and standard deviation) for the selected region.

  1. Age-sex composition snapshot (for a selected iteration). Use deerpopdy[CountyName].csv for this analysis. Update INDNR estimates for age-sex proportion in the code.


The same data is used to create barplots for assessing age-sex composition of the model deer population.

NOTE: Model calibration for LaPorte County is not complete. Also, the following tables need to be updated with appropriate INDNR estimates.

  1. Assessing model deer population parameters (year 25, age-class ratio and doe: buck ratio). Use sa[CountyName].csv for this analysis.Also, update INDNR estimates for the selected region in the code.
Population parameters INDNR estimates INOvPOP
Fawn proportion 0.30 0.34
Yearling proportion 0.22 0.24
Adult proportion 0.48 0.42
Female: male ratio 1.56 1.18
  1. Comparing model deer population harvest (age-sex classwise) with MDC harvest data. Use deerpopdy[CountyName].csv file already in the results folder.Update INDNR estimates for the selected region.
Age-sex class harvest INDNR data INOvPOP
male fawn harvest 379 34
male yearling harvest 313 119
male adult harvest 954 273
female fawn harvest 292 13
female yearling harvest 654 78
female adult harvest 520 201

INOvCWD: Investigating CWD dynamics


NOTE: INOvPOP should be calibrated so as to generate a realisitc and representative Kankakee white-tailed deer population. The following analysis is done without any calibration as the purpose is to illustrate the process.


  • INOvCWD was simulated for the Kankakee region for 25 years, one CWD infected yearling was introduced in the 1st year.

  • One hundred iterations of this baseline scenario were undertaken.

  • The output file ‘cwdinfdyKankakee_seedinf_1.csv’ was copied in the data folder and renamed ‘cwdinfdyKankakee_bl100.csv’

Spread of CWD in Kankakee white-tailed deer population over 25 years.


  • Analyzing INOvCWD outputs. Use cwdinfdy[CountyName]_bl100.csv for this analysis.

  • Each line represents CWD trajectory for one iteration.

  • Reset the horizontal red, dashed line that represents 1% prevalence (calculate 1% prevalence from the data in cwdinfdy[CountyName]_bl100.csv file.)

  • CWD outbreak did not persist for some iterations.

  • Demographic competence is defined as the ability of a host population to sustain the transmission of an infectious disease - was calculated as the proportion of 100 iterations where CWD outbreak persisted beyond 10 years.Higher the demographic competence, higher the likelihood of CWD becoming established in the deer population. For typical midwest white-tailed deer populations, the model-estimated demographic competence for CWD ranged between 0.35-0.40.
## [1] "Kankakee Deer Population \nDemographic competence: 0.57"

Baseline CWD trajectory for the Kankakee white-tailed deer population

## Warning: Ignoring unknown aesthetics: frame
  • Generating a statistical portrait using CWD prevalence data from 100 iterations for Kankakee white-tailed deer population. We will compare alternate scenario trajectories with this baseline trajectory.