coreytcallaghan / vagrant-birds-in-NA

assessing the role of vagrant birds in birdwatching
0 stars 0 forks source link

Do birders contribute more data when visiting an area? #2

Open coreytcallaghan opened 7 years ago

coreytcallaghan commented 7 years ago

Hey @wcornwell and @mitchest . This would be for objective 4. Essentially, I want to ask the question of whether birders who travel from a distance away, are they more likely to submit more checklists within the region the vagrant is hanging out. So, in other words, is there a benefit (in terms of citizen science data collection) of vagrant birds?

I think the approach is pretty straightforward, but I'm not entirely sure how to code it. It might be something that one of you have done before? So, before I dug in I wanted to throw it up as an issue.

I think the best way to go about it would be to select observers based on distance, and then for each observer count the number of checklists submitted within a 'bounding box' of the birds coordinates. Could even do 2-3 spatial levels of distance surrounding the bird. Maybe they only submit a couple checklists directly near the vagrant, but not more than 20 km away?

This code (also in the r script "do_birders_contribute_more_data.R"):

# packages
library(dplyr)
library(readr)
library(ggplot2)
library(tidyr)
library(data.table)
library(geosphere)

# load data
load("Data/observer_data.RData")
load("Data/ABA_analysis_results.RData")

# read in dataframe of rarities
vagrant_birds_2016 <- read_csv("Data/vagrant_birds_2016.csv")
ABA_vagrants <- vagrant_birds_2016 %>%
  filter(scale=='ABA') %>%
  rename(COMMON_NAME=species) %>%
  rename(STATE_PROVINCE=state_prov)

# add the bird lat/long to the dataframe of vagrants
ABA_vagrants <- ABA_analysis %>%
  select(COMMON_NAME, STATE_PROVINCE, bird_lat, bird_long) %>%
  filter(complete.cases(bird_lat)) %>%
  distinct(.keep_all=TRUE) %>%
  inner_join(., ABA_vagrants, by=c("COMMON_NAME", "STATE_PROVINCE"))

Should provide you with a dataframe (ABA_vagrants) which has the bird's lat/long. Need to group by COMMON_NAME and STATE_PROVINCE for a unique combination. It also has all the observer data in ABA_observers. And ABA_analysis shows more information.

Let me know what you think the best approach would be to tackling this question, and if you would know how to code it...

Cheers, Corey

coreytcallaghan commented 7 years ago

Because I'm not sure if I was clear enough. A final way to phrase it would be to ask the question:

What percentage of birders travel to a vagrant bird and then submit 1 checklists with the vagrant bird on it?