Define deployments and outliers
In Movebank, all data are stored in user-owned studies. Data owners are responsible for choosing what data to upload to each study and for defining which uploaded locations and measurements represent animals. The only data that are viewable by default from the Tracking Data Map are those portions of data that the data owners have defined as animal tracks (to view all data, see here). All event data imported into Movebank are required to be associated at minimum with a Tag ID and Sensor Type. Beyond this requirement, Movebank assumes events are valid tag measurements and does not assume that tags are equivalent to animals. There are two primary reasons why event records should not be associated with an animal:
- The data were collected when the tag was not attached to the animal or the animal had not yet been released.
- The location is an outlier and doesn't reflect the animal's movement.
Movebank generally does not allow individual records to be deleted from a dataset. Instead,
- The beginning and end of the period during which a tag was attached to an animal is defined using Deployments.
- Outliers within the deployment period are marked in the Event Editor or mapped to an outlier attribute when the data are imported.
Define Deployments
The start and end of a deployment—the period during which a tag (Tag ID) was attached to an animal (Animal ID)—are defined in Movebank using the reference data terms deploy on timestamp and deploy off timestamp. Defining deployments allows the data owner to exclude recorded data that were collected before or after the period when the animal was actually being tracked, for example when the tag was being tested or when it was attached to an animal during relocation prior to release. Defining deployments is especially important when a tag has been deployed on multiple animals. If the uploaded data include only the deployment period, defining a deployment period may not be necessary for proper visualization and analysis; however, it is still good practice to explicitly define this period so that it is clear that, for example, there are no missing data at the beginning or end of the deployment.
There are several ways to define deployments in Movebank:
- in the Deployment Manager
- by uploading a reference data file
- from the Studies page
Mark outliers
There are sometimes records within a deployment that should not be considered part of an animal's track. These records are defined in Movebank using the data terms import marked outlier manually marked outlier, algorithm marked outlier (outliers marked by Argos data filters or general purpose filters), and visible. Outliers can include incorrect locations caused by equipment or data processing problems, or records for which the error is considered too large to be properly analyzed, for example with Argos Doppler or geolocator data. In the latter case, defining what points are outliers can be subjective, depending on how the data will be analyzed; for example, greater error might be acceptable when evaluating large-scale migration than when analyzing local foraging behaviour.
To mark outliers, data owners can include the attribute manually marked outlier in the data files that they upload to Movebank, or they can mark outliers within Movebank after upload using the Event Editor.
For Argos data, outliers can be automatically marked in Movebank using Argos data filters.
When outliers are marked using either of these methods (with the value "TRUE" indicating an outlier), the field visible is automatically added to the dataset. In the visible attribute, records that have been marked as outliers have the value "FALSE" (the record will not be shown on the Tracking Data Map), and records that remain part of the animal track have the value "TRUE" (the record will be visible).
View undeployed data
To view all imported data, including outliers or data that are not part of a deployment (for example, if data have been uploaded but have not yet been associated with animals),
- Go to the Studies page and log in to Movebank.
- Search or select the study from the list on the left to view the Study details.
- To view all data imported to the study, select View > Show in map (include undeployed).
- To view all data imported for a tag,
- Select Tags from the menu directly below your study name on the left.
- Select the Tag ID from the list that appears.
- Select View > Show in map.
- To view all data imported from a file,
- Select File from the menu directly below your study name on the left.
- Select the file name from the list that appears.
- Select View > Show in map.
Quality control of uploaded data
Define deployments and outliers