GLOBE News
Changes to Biometry Measurement Protocols
On 2 August 2023, three changes were made to biometry measurements in relation to the way data observations are counted and displayed. These changes affected biometry - trees, biometry - graminoid biomasses, and biometry - vegetation covers.
Biometry - Trees
Two new fields have been added to the ADAT output for this protocol: “dominant_common_name” and “codominant_common_name”. These have been added to help distinguish between tree types for different measurements. These are required fields when entering data in the GLOBE Observer app, so the decision was made to make these fields available to see when retrieving data measurements for this protocol. About 3% of all measurements (~2k out of ~78k biometry - trees measurements) have been updated to return the common names provided at the time of data entry. All common names for future measurements will be made available in the data output.
Biometry - Vegetation Covers
A rounding error has been corrected for the fields within this protocol. Previously, all percentages in the output were being rounded to either 0% or 100%. With the new update, the percentages are now being returned correctly, rounded to the nearest 1/10th of 1 percent. This update does not affect the number of rows/measurements being returned in the output, it just gives more accurate percentages for fields where applicable for this protocol. About 23% of all measurements (~18k out of ~78k vegetation covers measurements) have been impacted and updated to contain proper percentage calculations.
Biometry - Graminoid Biomasses
There was an issue with the output of the data for this protocol aggregating the number of samples for a single measurement into multiple measurements. This caused the number of measurements for this protocol to be inflated to a larger count than the actual number, and it also means that the averages for each measurement were incorrect. This update fixes that issue so that now each sample is aggregated into a single measurement as it should be, and it correctly calculates the averages for fields where that applies. About 10% of all measurements (~600 out of ~6k graminoid biomasses measurements) have been impacted and updated to properly report the averages of all samples of the measurement into a single row of data (per measurement). This also reduced the number of rows returned from the data output by ~5% (300 out of 6k), which now represents the accurate number of measurements.