Proposed improvement of leaf area index estimation for fruit trees using a two-level automated segmentation to isolate non-leaf material from the analysis where: (a) is the original image; (b), (c) and (d) are the resulting images for the first segmentation level and (e), (f) and (g) are the resulting images for the second segmentation level.Display Omitted An improved image analysis method to estimate leaf area index has been tested.Improvements consist in image segmentation algorithms to exclude non-leaf material.Results suggest that this method is an affordable alternative to compute LAI. Leaf area index (LAI) is a critical parameter in plant physiology for models related to growth, photosynthetic activity and evapotranspiration. It is also important for farm management purposes, since it can be used to assess the vigor of trees within a season with implications in water and fertilizer management. Among the diverse methodologies to estimate LAI, those based on cover photography are of great interest, since they are non-destructive, easy to implement, cost effective and have been demonstrated to be accurate for a range of tree species. However, these methods could have an important source of error in the LAI estimation due to the inclusion within the analysis of non-leaf material, such as trunks, shoots and fruits depending on the complexity of canopy architectures. This paper proposes a modified cover photography method based on specific image segmentation algorithms to exclude contributions from non-leaf materials in the analysis. Results from the implementation of this new image analysis method for cherry tree canopies showed a significant improvement in the estimation of LAI compared to ground truth data using allometric methods and previously available cover photography methods. The proposed methodological improvement is very simple to implement, with numerical relevance in species with complex 3D canopies where the woody elements greatly influence the total leaf area.

A study was carried out to parameterize the METRIC (Mapping EvapoTranspiration at high Resolution with Internalized Calibration) model for estimating instantaneous values of albedo (shortwave albedo) (α_i), net radiation (R_ni) and soil heat flux (G_i), sensible (H_i) and latent heat (LE_i) over a drip-irrigated Merlot vineyard (location: 35°25′ LS; 71°32′ LW; 125 m.a.s. (l). The experiment was carried out in a plot of 4.25 ha, processing 15 Landsat images, which were acquired from 2006 to 2009. An automatic weather station was placed inside the experimental plot to measure α_i, R_ni and G_i. In the same tower an Eddy Covariance (EC) system was mounted to measure H_i and LE_i. Specific sub-models to estimate G_i, leaf area index (LAI) and aerodynamic roughness length for momentum transfer (z_om) were calibrated for the Merlot vineyard as an improvement to the original METRIC model. Results indicated that LAI, z_om and G_i were estimated using the calibrated functions with errors of 4%, 2% and 17%, while those were computed using the original functions with errors of 58%, 81%, and 5%, respectively. At the time of satellite overpass, comparisons between measured and estimated values indicated that METRIC overestimated α_i in 21% and R_ni in 11%. Also, METRIC using the calibrated functions overestimated H_i and LE_i with errors of 16% and 17%, respectively while it using the original functions overestimated H_i and LE_i with errors of 13% and 15%, respectively. Finally, LE_i was estimated with root mean square error (RMSE) between 43 and 60 W∙m⁻² and mean absolute error (MAE) between 35 and 48 W∙m⁻² for both calibrated and original functions, respectively. These results suggested that biases observed for instantaneous pixel-by-pixel values of R_ni, G_i and other intermediate components of the algorithm were presumably absorbed into the computation of sensible heat flux as a result of the internal self-calibration of METRIC.