All plots are trapped approximately monthly (1977 - present). Sampling is ongoing and data will be added over time.
Portal_rodent contains the rodent trapping data. Each individual of a target species is PIT tagged and data on species, location caught (plot and stake), sex, reproductive condition, weight, and hindfoot length are recorded.
In almost all cases, you will want to remove negative period codes before using this data. Negative period codes indicate data collected outside of the normal census protocols, and should absolutely not be used as regular trapping data.
The id column can be used to track individuals over time. It is unique to individual, across the entire time series. The pit_tag column indicates which of these unique ids are based on PIT tags, which are more reliable than other tagging methods.
Portal_rodent_datanotes lists the data flags used in the Note2 column of Portal_rodent.
Portal_rodent_species contains the species codes used in Portal_rodent. Any animal found in a trap is recorded. Non-rodent species are occasionally trapped and so are given species codes. Several columns are also included in the species table to restrict the species list to only rodents, only target species, or only granivores.
Occasionally, trapping sessions will be missed either partially or completely due to dangerous weather or other extenuating circumstances. Portal_rodent_trapping is a complete list of when trapping was done by year, month, period and plot. This can be used to differentiate real zeros from missing data in Portal_rodent, and to account for differences in trapping effort when summing abundances. sampled = 1 indicates a plot was sampled that period. effort indicates how many traps were set in that plot that period (usually 49). qcflag
= 1 indicates data that have undergone at least a year of quality control (the recommended amount, though more recent data are still appropriate for certain uses).
Trapping is conducted as close to the new moon as possible. This can ocasionally lead to more than one census in a month and it means that censuses and months do not line up conveniently. Months that are entirely missed are not noted in the database. We have generated moon_dates to provide a time series of new moon occurences, and align the trapping data to them. While period is only the census number and does not account for missing censuses, newmoonnumber assigns a number to each new moon, regardless if the census happened or not. Then period can be used to connect the data from Portal_rodent to the more consistent newmoonnumber.
Portal_suspect_stakes keeps a record of suspect stake values. In normal censuses, the stake column is restricted to the 49 on-plot stake locations. So suspect values (values that have been misrecorded, or where only row or column is known) are NA in the main table. The suspect-stakes table retains a record with the bad stake values.
(Non-standard stake values are normal in the off-plot trapping data (more info about this in methods). However, those trapping sessions are also assigned negative period codes and should be removed from the data before its use.)
Also recall that Portal_plots can be used to correctly assign treatment by year and month.
Please refer to Methods.md for a complete description of how these data were collected.