So many of us will be dealing with observations that apply to a time interval rather than a single instance in time. It's worth confirming how we handle these.
An example would be precipitation accumulation from a rain gauge. It's not enough to say 3 mm of rain was recorded at 10:40 pm, we need to know whether that 3 mm was collected over the last minute, or hour, or day, etc...
This article also has some nice meteorological examples showing which observedProperty's would be instant and which would be interval.
This SSN Example also makes use of a Time Interval. Using this as a basis, an observation of precipitation accumulation could therefore be structured as:
So many of us will be dealing with observations that apply to a time interval rather than a single instance in time. It's worth confirming how we handle these.
An example would be precipitation accumulation from a rain gauge. It's not enough to say 3 mm of rain was recorded at 10:40 pm, we need to know whether that 3 mm was collected over the last minute, or hour, or day, etc...
The Time Ontology makes the distinction between a Time Instant and a Time Interval.
This article also has some nice meteorological examples showing which observedProperty's would be instant and which would be interval.
This SSN Example also makes use of a Time Interval. Using this as a basis, an observation of precipitation accumulation could therefore be structured as:
N.b. I haven't made this example JSON-LD compliant, but I should have.