My assumption is this:
Nicola sets off on the walk with the dog and once she got to the gate she would take the dogs harness and lead off. Now she is hands free from the dog she can send her email and start the zoom call, she is spotted by somebody that knows her and her dog and they confirm her phone being on show. She is later spotted in the higher field with her dog off lead. The time from 9.10am ( last sighting in top field) to 9.20am (phone tech confirmed at the bench area) is accurate for the 10 mins it would take to get back to the bench area.
Now Nicola has to get the dog back on the lead and the bench is a good area if you are still pre occupied with a zoom call, hence putting the phone on the bench while she sorts the dog out with harness etc, because she can still hear the phone whilst tending to the dog. However i believe she has some accident at the slope near the bench and ends up in the water, probably due to gettng the harness on and wet/slippy steep slope or due to a medical condition possible heart attack? This is why she never ends the zoomcall.
People are assuming the dog will just bindly jump into the water, but im not so sure about this assumption. Dogs can be very scared and if its a steep slope it could put the dog in doubt as to what to do. I think the dog just stays very close to that area because it is in a panicked state. My only problem with my take on events is that i find it highly unlikely that the weight of her body in wet clothing has been able to pass the weir, then pass all of the island areas, pass over all silt and not get caught up anywhere to be found before getting to sea. But then i am assuming that the search underwater is flawless.