Interpolated Landsat Maps Correlation to Corn or Soybean Yield
(Graphics and statistics shown below)
(About Landat and corn or soybean yield correlations and interpolated correlation maps below)
Below, Correlation (R²) between interpolated Landsat (Landsat Yield Prediction Maps [LYPMs; a description is included at bottom of page]) and clean yield monitor data. The difference in detail between clean yield monitor data (finished to a 4 meter resolution) and Landast pixels (30 x 30 meter resolution)
Landsat NIR (band 4), solely, is correlated to soybean in all cases. For corn, NNIR (Spirada, 2006) (NIR / [NIR + red + green]) is applied for most correlations but NDVI (Rouse, 1973) ([NIR - red] / [NIR + red]) is used in two cases (because it is a common and familiar index).
NDVI [NIR - red] / [NIR + red]) correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
Below, correlations between interpolated Landsat and clean yield monitor data with Landsat 30-meter pixel boundaries shown. As opposed to the previous maps, Landsat and yield maps below were produced to an extent bounded by the centers (centroids) of Landsat pixels (for pixels that had corresponding clean yield monitor data). (See About Landsat - yield correlations with pixel boundaries shown page.)
NNIR (NIR / [NIR + red + green]) correlation to clean yield monitor data
NNIR (NIR / [NIR + red + green]) correlation to clean yield monitor data
NNIR (NIR / [NIR + red + green]) correlation to clean yield monitor data
NNIR (NIR / [NIR + red + green]) correlation to clean yield monitor data
NDVI [NIR - red] / [NIR + red]) correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
NIR reflectance correlation to clean yield monitor data
About Landsat Yield Prediction Maps (LYPMs) (top five maps above)
A Landsat Yield Prediction Maps (LYPM) (Hollinger, 2011) is a map customized to the extent of a field that is based on interpolating from the center (centroid) points of 30 x 30 meter Landsat pixels (typically p < 0.0001 for the yield correlation). Essentially, they are developed by extending valid interpolated values to the boundary of the field; valid values are those derived from pixels that represent crop area within the field, not pixels along the outside boundary that are averaging in surface outside the field or pixels within outside boundary that are averaging in non-crop areas that are too significant.
References
Hollinger, D. 2011. Crop Condition and yield prediction at the field scale with geospatial and artificial neural network applications. Dissertation. Kent State University.
Qi, J., Chehbouni, A., Huete, A. R., Kerr, Y.H., and S. Sorooshian. 1994. A Modified Soil Adjusted Vegetation Index. Remote Sensing of Environment 48: 119-126.
Rouse, J.W., R.H. Haas, J.A. Schell, and D.W. Deering. 1973. Monitoring vegetation systems in the Great Plains with ERTS. Third ERTS Symposium, NASA SP-351 I: 309-317.
Sripada, R.P., Heiniger, R.W., White, J.G., and A.D. Meijer. 2006. Aerial color infrared photography for determining early in-season nitrogen requirements in corn. Agronomy Journal 98: 968-977.
Wu, J., Wang, D, and M.E. Bauer. 2007. Assessing broadband vegetation indices and QuickBird data in estimating leaf area index of corn and potato canopies. Field Crops Research 102: 33–42.