We’re going to use the Northern Bobwhite (hereafter ‘quail’) “whistle count” surveys alongside remotely-sensed tree cover data for our first generalized additive models (GAMs). Broadly, we’ll be comparing quail abundance (measured by whistle counts) to tree cover while accounting for spatiotemporal autocorrelation. Then we’re going to practice writing results for the model outputs.
Download this R script. We’ll use it to work through the challenges today.
| Column Name | Description |
|---|---|
| Year | years 1965 - 2020 |
| County | Nebraska county name |
| Route.Name | Route name |
| Route.Number | Route number |
| UTMXN | UTM northing zone 14 |
| UTMYE | UTM easting zone 14 |
| WhistleCount | Mean whistle count per route per year |
| Bob_Region | Quail management region code |
Let’s load the data and tidy it:
#===============================================================================
## Preparations
# # Clear environment?
# rm(list=ls())
# List of packages necessary to run this script:
require(librarian)
shelf(tidyverse, here, sf, brms, maps, terra, tidybayes,
lib = tempdir())
# Set path
repo_url <- "https://raw.githubusercontent.com/LivingLandscapes/Course_EcologicalModeling/master/data/"
# Set seed
set.seed(2252)
# Load data
dat_list <-
list(RouteDictionary = read_csv(paste0(repo_url, "QuailWhistle_Nebraska_RouteDictionary.csv")),
RouteUTMs = read_csv(paste0(repo_url, "QuailWhistle_Nebraska_RouteUTMs.csv")),
RouteWhistleData = read_csv(paste0(repo_url, "QuailWhistle_Nebraska_RouteWhistleData.csv")))
#=============================================================================
## Tidying data
# Tidy whistle data
whistle <-
dat_list$RouteWhistleData %>%
pivot_longer(cols = Cass:Lancaster_YH,
names_to = "Route.Name",
values_to = "WhistleCount") %>%
mutate(Route.Name = case_when(Route.Name == "Cass" ~ "Cass North",
Route.Name == "GageN" ~ "Gage North",
Route.Name == "GageS" ~ "Gage South",
Route.Name == "Jeffersn" ~ "Jefferson",
Route.Name == "JohnsonN" ~ "Johnson North",
Route.Name == "JohnsonS" ~ "Johnson South",
Route.Name == "Lancaster.East" ~ "Lancaster East",
Route.Name == "Lancaster.North" ~ "Lancaster North",
Route.Name == "Lancaster.Branched.Oak" ~ "Lancaster-Branched Oak",
Route.Name == "OtoeN" ~ "Otoe North",
Route.Name == "OtoeS" ~ "Otoe South",
Route.Name == "Richardn" ~ "Richardson",
.default = Route.Name)) %>%
left_join(dat_list$RouteDictionary %>% mutate(Route.Name = `Route Name`)) %>%
mutate(Route.Number = `Route Number`) %>%
filter(Route.Name != "Lancaster_YH") %>%
left_join(dat_list$RouteUTMs %>%
mutate(Route.Number = Route) %>%
group_by(Route.Number) %>%
filter(row_number() == 1)) %>%
mutate(YearNum = Year - 1964,
YearNum2 = (Year - 1964)^2,
Easting_scaled = scale(UTMYE),
Northing_scaled = scale(UTMXN))