Portuguese Moth Recording Scheme

Study Extent

The geographic and temporal scope of this dataset is mainland Portugal and the Madeira Archipelago, between 2021 and 2022

Sampling

Sampling procedures were based on light trapping. There is some variation in the sampling protocol between participants (sampling stations) that can be attributed to differences in the design of moth traps (e.g. Heath trap, Skinner trap, Robinson trap), the light source (e.g. actinic light, mixed light, LED) and its power. Moreover, the time effort is also variable between samplings. These differences are considered in the columns samplingEffort and eventTime. The right method used for each record is stated in the columns samplingProtocol (e.g. Skinner trap with Mercury vapor light of 125 W).

Quality Control

To ensure the quality of the data, all the records that consisted of a significant increase in the known distribution of the species, i.e., a first record of a given species for Portugal or for a given province, were verified by the authors through the analysis of photos of the specimen(s) that support that record. In addition, records of uncommon species were also consistently verified by the authors. In the preparation of the dataset for GBIF publication, data consistency was verified using cluster algorithms using OpenRefine.

Method steps

1. The present data were obtained from the database of the project "Rede de Estações de Borboletas Noturnas", from which the records where column “Por confirmar” (to be confirmed) contained “Yes” were filtered in Open Refine – in order to select only the records that the participants considered confirmed. Records without assigned species were also removed.
2. To select only the records of macro-moths, just the families Cossidae, Drepanidae, Erebidae, Euteliidae, Geometridae, Hepialidae, Lasiocampidae, Limacodidae, Noctuidae, Nolidae, Notodontidae, Saturniidae, Sphingidae were considered for this dataset.
3. All the columns were placed in DwC format and mandatory columns were added: type, modified, license, collectionCode, datasetName, basisOfRecord, occurrenceStatus.
4. The identifications in species groups where safe identification is only possible through dissection were reduced to genus level. The original identifications from the participants were maintained in the column verbatimIdentification. The column taxonRank present the taxonomic rank of the most specific name in the scientificName column.
5. Some participants use the iNaturalist social network to get some help with moth identification (either by users, or through the AI algorithm). To signal the records identified using this network, a column named identificationRemarks was created and filled with "iNaturalist". The column identifiedBy is empty for these records. The column identificationRemarks also mentions whether the identification is based on the analysis of the genitalia of at least one specimen.
6. The sampling effort of each sampling event was determined from the start and end times (e.g. “07 hours and 30 minutes of sampling time” in samplingEffort column from “22:00:00Z/05:30:00Z” in eventTime column).

This dataset includes 23694 entries referring to records of 13 different families, 298 genera and 573 species/subspecies.

1. Animalia
   rank: kingdom
2. Arthropoda
   rank: phylum
3. Insecta
   rank: class
4. Lepidoptera
   rank: order
5. Cossidae
   rank: family
6. Drepanidae
   rank: family
7. Erebidae
   rank: family
8. Euteliidae
   rank: family
9. Geometridae
   rank: family
10. Hepialidae
    rank: family
11. Lasiocampidae
    rank: family
12. Limacodidae
    rank: family
13. Noctuidae
    rank: family
14. Nolidae
    rank: family
15. Notodontidae
    rank: family
16. Saturniidae
    rank: family
17. Sphingidae
    rank: family

The occurrences are in mainland Portugal and Madeira Archipelago.