Hi,
We are currently using the SWIR band from S2 to check the irrigation in rice crops. We need to see if the surface is flooded with water, under the crop canopy.
The SWIR band allows us to se through the crop canopy.
We are trying to have the same data in a daily frecuency with Planet Scope. I have tested the NDWI, but the results show the hydric state of the crop, not the prescense of water under it.
I wonder if you know a way or index to see water prescence under the leaves with PlanetScope bands.
As the water is under the crops this is very difficult to detect without the SWIR band which is not available with Planetscope data. You can use NDWI derived from the Green and NIR channels as documented here but this is probably not useful in your case.
My suggestion would be to try utilising Data Fusion methods using Planetscope in conjunction with Sentinel-2. You can find out more about Data Fusion in Sentinel Hub APIs here where there are some great examples too.
We didn’t find a specific solution for this. Unfortunately, Planet doesn’t have SWIR band so as far as we know, we cannot have a soil moisture/irrigation index for crops using this satellite.
We we starting to use radar satellites to have other indices semi-related to this since radar can also go through the plants and give you information of the soil, but again, it’s not exactly what SWIR band provides, we just built other indices to accompany the irrigation index we created with the SWIR band with Sentinel 2.
Another option we are exploring is the new Planetary Variables from Planet: Planetary Variables. We are taking a look at this as well.
I’m working on a project to ground truth the usefulness of different indices for similar hyper-specific use cases. I can deliver products from all bands, multiple sensors with weather stats on demand if you would be keen to share your findings in the dataset.