Query based on labels
Labels are used to organize all the data in the graph database. Here you can filter data based on
labels denoting different physical entities,
dataset types, habitats, and select
sites. For most data-related queries, you probably want to use one of
the dataset type labels.
In the query output tables, each row represents a node, and each column (except for the first "node labels"
column, which lists all the labels on each node) represents a property. Properties ending in "__" have been
mostly standardized throughout the graph database, and are therefore very useful for harmonizing data from
A full list of label and standardized property definitions can be downloaded
For a more comprehensive overview of the graph database structure, which may aid in understanding the
query output, see the more detailed overview
Filtering the results
If you select one label, you'll get all the nodes associated with that one
label. When running a live query (in development), if you select multiple labels, you'll get nodes that
have all of those labels (which may not be many). When retrieving cached queries, which are built on
a semi-regular schedule but give instant results, then you can only select one label at a time when running
queries. The live querying will remove that restriction, but for now, you can still further refine your
results by filtering the output table. For smaller output files, this can be done within the browser as
- Choose a label and run the query by clicking "Cached query."
- Click "Display table" (bottom of page).
- Search for specific keywords either within the full table (top right search box) or within a specific
column (search boxes under each column). This will filter the table to show only the rows with that
- Under "Download the table below as" (above the displayed table), choose either "Excel" or "CSV." Even
though the browser will only display a few rows at a time, the download will include the entire table,
minus any rows you eliminated by keyword filtering.