The March 2014-15 Valencia orange forecast is 20 million cartons. The March 10 National Agricultural Statistics Service (NASS) forecast was based on the results of the 2014-15 Valencia Orange Objective Measurement (O.M.) Survey, which was conducted from January 16 to February 14, 2015.
Estimated fruit set per tree, fruit diameter, trees per acre, bearing acreage, and oranges per carton were used in the statistical models estimating production. The season has been dry in many areas and measurements are indicating an average fruit size, but the fruit set has been negatively impacted.
Survey data indicated an average fruit set per tree of 545, well below the five-year average of 639 and the lowest set since the 2008-09 season. The average March 1 diameter was 2.571 inches, slightly above the five-year average of 2.562.
A Valencia Orange Objective Measurement Survey was conducted from the 1985-86 to 1993-94 seasons before suspension due to a lack of funding. The survey has been conducted since it was reinstated for the 1999-00 season, with the exception of the 2006-07 season due to a substantial freeze. The data from the first three years after the survey was reinstated were used for research purposes in developing crop-estimating models.
A sample of 577 Valencia orange groves was randomly selected proportional to acreage, county and variety representation for the state, with 539 of these groves being utilized in this survey. Once a grove was randomly chosen and grower permission was granted, two trees were randomly selected for each grove. For each randomly selected tree, its trunk was measured along with all connected branches. A random number table was then used to select a branch, and then all connected branches from the randomly-selected branch were measured. This process was repeated until a branch was reached with no significant limbs beyond it.
This randomly-selected branch, called the terminal branch, was then closely inspected to count all fruit connected to it, as well as all of the fruit along the path from the trunk to the terminal branch. Since each selected path has a probability of selection associated with it, a probability-based method was then applied to estimate a fruit count for the entire tree. In the last week of the survey period, fruit diameter measurements were made on the right quadrant of four trees surrounding the two trees of every third sampled grove. These measurements were used to estimate an average fruit diameter per tree. The sampled groves were primarily in the top Valencia orange producing counties of Tulare, Kern, Fresno, Ventura and San Diego.