site-level covariates must be constant

Question

I am trying to prepare a dataframe of species observation for abundance and occurrence data using habitat covariates.

When using this code:

library(auk)
occ_wide <- format_unmarked_occu(occ, 
                                 site_id = "site", 
                                 response = "species_observed",
                                 site_covs = c("n_observations", 
                                               "latitude", "longitude", 
                                               "pland_00_water",
                                               "pland_11_wetland",
                                               "pland_12_cropland",
                                               "pland_13_urban"),
                                              
                                 obs_covs = c("time_observations_started", 
                                              "duration_minutes", 
                                              "effort_distance_km", 
                                              "number_observers", 
                                              "protocol_type",
                                             "pland_11_wetland"))

I get this error:

Error in format_unmarked_occu(occu, site_id = "site", response = "species_observed", : Site-level covariates must be constant across sites

What does this mean and how can I overcome this error?

a reproducible code can be found here: https://github.com/lime-n/ebird_data/blob/master/occ.csv

I have found a way of bypassing this by using

library( data.table )
setDT(mydata)[ !duplicated( site, fromlast = TRUE ), ]

However, when I run the occu model, it shows mainly negative values:

dggs <- dgconstruct(spacing = 5)
# get hexagonal cell id for each site
occ_wide_cell <- occ_wide %>% 
  mutate(cell = dgGEO_to_SEQNUM(dggs, longitude, latitude)$seqnum)
# sample one site per grid cell
occ_ss <- occ_wide_cell %>% 
  group_by(cell) %>% 
  sample_n(size = 1) %>% 
  ungroup() %>% 
  select(-cell)
# calculate the percent decrease in the number of sites
1 - nrow(occ_ss) / nrow(occ_wide)

occ_um <- formatWide(occ_ss, type = "unmarkedFrameOccu")
summary(occ_um)

occ_model <- occu(~ time_observations_started + 
                    duration_minutes + 
                    effort_distance_km + 
                    number_observers + 
                    protocol_type +
                    pland_04_deciduous_broadleaf + 
                    pland_05_mixed_forest
                  ~ pland_04_deciduous_broadleaf + 
                    pland_05_mixed_forest + 
                    pland_12_cropland + 
                    pland_13_urban, 
                  data = occ_um)
# look at the regression coefficients from the model
summary(occ_model)

which means I cannot run this code as it requires postive values:

occ_gof <- mb.gof.test(occ_model, nsim = 10, plot.hist = FALSE)

Meaning that I did not approach it correctly, because the majority of valuable sites are removed.

EDIT: problem is within longitudinal coordinates

site-level covariates must be constant

0 Answers