Definition of an Observation In general an observation

Definition of an Observation In general, an observation represents the measurement of some attribute, of some thing, at a particular time and place. Observations can be gathered using sensors or sensor arrays, direct human observations, or biological specimen collections. Specifically, an observation characterizes the evidence for the presence or absence of an organism or set of organisms through a data collection event at a location. Observations are not necessarily independent and could be linked via characteristics such as time, place, protocol, and co-occurring organisms.

TDWG OSR Interest Group Focus: Primary biodiversity data from observer-derived measurements and specimens stored within natural history collections. Concept: There is sufficient overlap in observational data and specimen records to join these data resources. Primary Issue: Unique requirements for individual projects (e. g. protocols, taxonomy, spatial). Goal: To create an observational data description that fully integrates specimen and observational data.

Metadata Standards Darwin Core and Access to Biological Collections Data (ABCD) Darwin Core: set of basic data element definitions that can be extended. ABCD: comprehensive structured hierarchical schema. Darwin Core and ABCD were developed to facilitate the exchange biological collections data at an international level.

TDWG OSR will focus on model extensibility and schema integration. Priority Issues: • Look for synergies between specimen and observations • Taxonomic unity • Variety of collection methods • Spatial representation • Define core data items for data discovery • Compare data attributes and identify overlap and look for synergies with other TDWG Interest Groups. • Support and provide input to the TDWG core ontology.

Key Elements of the Darwin Core • GUID- Global Unique Identifier • Record Information- e. g. where, date modified, collection or catalog number • Taxonomic Organization • General geographic Information • Time and Date • Specimen

Key Extensions to Darwin Core for the Bird-monitoring Data Exchange • Project code • Protocol description • Survey area description • Effort information duration of observation number of observers • Number counted • All species reported • Data access rights

GBIF ORNIS Avian Knowledge Network

Figure 2. The observational data matrix showing that as the type of analysis (rows) becomes more detailed the information (columns) necessary to be gathered increases. (http: //avianknowledge. net).

The occurrence of Red-breasted Nuthatch reported during the irruption of November 1998. While these data are useful for showing the extent of a species occurrence at the continental scale, they provide little information at greater spatial resolution. (Source: The Irruptive Bird Survey http: //www. birdsource. org/ibs).

Seasonal changes in Common Yellowthroat distribution in North America during 2006. The frequency of checklists reported is shown. These maps provide information both on coverage (e. g. where data were collected) as well as some indication of how common the species was. A total of 174, 197 checklists were submitted to e. Bird in 2006 and 10, 447 of them included Common Yellowthroat. The data were collected in e. Bird (http: //www. ebird. org).

Thirty year trends in Northern Flicker occurrence. This map was generated from observations made over a 30 year period. Individual Locations (@ 3000) were sampled repeatedly The data were collected in Breeding Bird Survey (http: //www. mbr-pwrc. usgs. gov/bbs. html).

Cross-project Data Integration

Use Data Mining models to predict the occurrence of grassland species 30, 000 observations • • • Rocky Mountain Bird Observatory 2001 – 2005 10, 000 locations 138 Predictors • • NLCD Habitat – 21 classes – 6 scales Climate (EPA) – Average Precipitation – Average Snowfall 30 grassland species

• Data federation is crucial for ecological analysis at appropriate scales. • Current techniques for organization have been an excellent first step. • Creation of a new observational ontology must build off these early successes.