Running into some inconsistencies between my MicaSense RedEdge-MX NDVI readings and what I'm seeing on the ground during crop walks. Getting NDVI values around 0.6-0.7 in areas where the corn clearly shows nitrogen stress (yellowing, stunted growth). Has anyone experienced similar calibration issues with multispectral sensors?

Using Pix4Dfields for processing and have confirmed camera settings are correct. Wondering if it's an atmospheric correction issue or if there's something with the sensor calibration itself. The data looks clean otherwise - no obvious artifacts or processing errors.

Flight parameters: 120m AGL, 80% overlap, sunny conditions around noon. Any insights on improving NDVI accuracy for precision N applications would be much appreciated.

Check your calibration panel readings - I've seen similar issues when the panel gets dirty or the lighting conditions change between panel capture and flight. Also, try flying at different times of day to see if solar angle affects your readings.

Are you using the downwelling light sensor (DLS)? Without proper irradiance correction, NDVI values can be way off, especially on partly cloudy days. The RedEdge-MX should have one built in.

Had this exact problem last season. Turned out my ground control points weren't positioned accurately enough - even small GPS errors can throw off the correlation between aerial data and field observations. RTK makes a huge difference for this type of validation work.