Imagery capture is a very complex operation. What Nearmap refers to as a “survey capture” is essentially a reasonably stable flightmap, flown on one day. The boundary may change occasionally, but in most cases they are stable.
In some cases, due to weather conditions or air traffic control (and a host of other environmental factors), we may stop the survey halfway through and resume later, sometime weeks later. On these rare occasions part of an image from a particular date may disappear at a later time. If this happens, the individual survey footprint may get reduced. If the footprint is reduced, there is always a neighbouring survey with a similar date.
Cases where imagery footprint changes after publishing:
1 - Partial survey – delayed re-flight
This is an exception and only happens where there are quality issues with the first flight. Some areas have difficult weather conditions, for example, newly formed clouds or a spring leaf-off capture that is subject to limited sun and patchy weather. Our pilots follow strict instructions on how to minimise these issues, but sometimes they are unavoidable.
In these conditions, the original survey may not be captured in one day and parts of the flightmap need to be re-flown at a later time. As the surveys are reprocessed, sometimes the boundaries are cropped to keep the imagery consistent and to match against the neighbouring survey. In this case you may even encounter a property that has been cropped out, because it is located at the border of a survey. The example below illustrates a possible scenario.
|Two surveys for neighbouring regions. Address of interest is on the border of survey.||Subsequent survey with updated boundaries. Surveys were cropped to match better. Address of interest now falls in area that was cropped.|
2 - Partial survey – immediate re-flight
This is a variant of the weather related re-flight. It could happen the same day or a couple days later. In some cases you will see two publish events, once for the first part of the flightmap, and another after a re-flight. Other times these will be done at the same time. If the survey is republished, you will see the area increase.
3 - HC2 inter-survey merge
In these cases the vertical boundary does not change. This happens for all HC2 surveys (i.e. ones that product 3D) once the whole season (e.g. New York Fall 2019, which is comprised of multiple flightmaps) is completed. This will result in no boundary change in most cases (unless a flightmap was only partially completed for reasons outside our control).
What can I do if I suspect I'm missing part of the imagery due to survey boundary anomalies?
You can use the Coverage API to retrieve the survey footprints and
onlineTime, and persist them in your system. Check the new boundary and make sure it didn’t change. Checking for area change is typically enough to filter out insignificant republish events (such as HC2 region merge).