The equation predicts annual evapotranspiration as a function of potential evapotranspiration and rainfall and was first published in:

Zhang, L., Dawes, W. R., & Walker, G. R. (2001). Response of mean annual evapotranspiration to vegetation changes at catchment scale. Water resources research, 37(3), 701-708.

 I have used NDVI of very dry grasslands and wet-evergreen forests with some success in a previous study: 

1. Krishnaswamy, J., M.C. Kiran, K.S. Bawa and K.N. Ganeshaiah. (2009). Quantifying and mapping biodiversity and ecosystem services using a multi-date NDVI based Mahalanobis distance measure. Remote Sensing of Environment 1(113): 857–867.

 But would like to see other or similar approaches.