Closed sjgraves closed 6 years ago
Aim: produce individual trees predictions of traits syndromes, including tree position and crown size/shape. (1) does working on crown level provide better predictions of traits and distribution? (2) why do we care: what can actually be done with individuals that couldn't with pixels [we show some applications]?
potential 3rd questions (but probably for another paper!): how do trait distributions differ within and among habitat types/land cover types and within and among NEON sites.
BUT one of the results will be mapped traits at different sites. So while we can't address the 3rd question, we can have some maps as figures and include some discussion points
Changing perspective in Remote sensing: from pixel to individual based predictions