Observation-guide
How to record observation effort without confusing absence with non-detection
A zero can describe a search with no results, but it does not prove that a species does not live there. Interpretation requires knowing what was sought and how.
Record the work, not just the result
A brief visit and a prolonged search do not represent the same effort. The GBIF guide explains that greater effort generally increases the probability of detecting an organism and that documenting effort helps compare records. Note the method, time spent and participants; when using sensors, describe their number, arrangement and operation. “I observed quite a lot” is insufficient: retain **values and units**, along with the procedure used to obtain them. [5446]
A record with values and units
This fictional example helps organise a notebook: “Visual observation of butterflies; two people; 30 minutes; route described in the record; search targeting the stated group”. Separate visit duration from participant numbers rather than merging them into an ambiguous figure. If you followed a protocol, retain its name, description or reference and note differences in the actual visit. GBIF proposes separate fields for effort value and unit, effort protocol and participants; the appropriate unit depends on the method used. [5446]
What seeing nothing means
If you do not detect the organism, record **not detected during this visit**, not “does not exist here”. The GBIF guide understands absences as non-detection within an explicitly stated taxonomic scope. A zero therefore needs context about which organisms were targeted and how that scope was documented. An incomplete list does not justify treating every unlisted species as absent. Keep result, scope and effort linked: this context allows reassessment without attributing a conclusion broader than the observation supports. [5446]
Sources and method
This story synthesises the references below and separates possible observations from conclusions that would need more evidence. It does not present original FaunaGuide fieldwork or replace expert identification.
- Guide for publishing biological survey and monitoring data — 5.5.2. Sampling protocol Four protocol terms exist; however, only 1 term is currently required to publish an Event dataset in GBIF: dwc:samplingProtocol. This requirement is because the initial Darwin Core Event classification only included the one term.