Waterbody types or data errors


Some test samples don’t have/show a water surface for a given date (and nearby dates). Example: uid = aabn, date = 2016-08-31, latitude: 36.5597, longitude: -121.51, no water for -30 days.
Now, these samples could be from wells and ponds that are too small to capture on satellite images and we are to overcome this as part of a competition. But also, it is a possible date error since google maps do show some water surface (probably artificial accumulations and controlled) for those coordinates. Maybe there is a need to recheck the competition data (at least the test part of the data).

Any feedback on this will be fine. Thank you.


Hi @neurogenesis ! Thanks for the question.

It is possible that there is noise in the underlying data. The data was collected from a large number of public health and water quality managers, and may be subject to human error. However, these cases are rare and there are no underlying systematic issues with the competition data.

This mirrors most real-world problems where data is not perfect :slight_smile: It is a good note and worth considering if there is a better way to identify and handle these cases.

Good luck!