Findings & Recommendations

Objective 1: Refine sampling techniques & develop early season sampling protocol

Variations in leaf nutrient status of the trees over the growing season were used to establish field sampling methodology and develop an early season sampling protocol.

• Leaf analysis data over two seasons and across four geographic locations in California indicates that nitrogen (N) and potassium (K) distribution among the leaves of a single tree canopy is not uniform. This variability among tissue samples even within a single tree canopy makes effective leaf sampling extremely difficult and has important implications for interpretation of the data and fertilizer management strategies.
• Relationship of leaf N in July from the fruiting and non-fruiting branches with pistachio yield did not provide greater information than any other. It is recommended, therefore, that the existing sampling strategy of collecting leaf samples from non-fruiting branches be maintained.

Update 2013 | Findings | Model

• 

  Figure 1. Canopy Variability. One of the challenges of leaf sampling is variation in leaf nutrient content within the canopy. Photo by M.I. Siddiqui
  To effectively obtain the average nutrient concentration (N, P and K) of a single production area or orchard growers must collect leaves from at least 18 trees in one bag each spaced at least 25 yards apart.
• If the growers goal is to only attain a representative sample of N in a field that is generally uniform, then a sample should be collected from 9 trees that are each 25 yards apart. The sample can then be pooled into a single bag for analysis or analyzed as 9 independent samples, the later approach would be chosen if the growers wanted information on variability in the field.
• The pooled sampling approach used here assumes ‘typical’ coefficient of variation (CoV), if growers wish to obtain their field specific CoV then samples collected from each tree should be analyzed independently and used to determine a site specific CoV to be used to adjust future sampling strategies. 
• If clear areas of differential tree behavior are known, these areas should be sampled and managed separately.
• To identify and correct nutrient problems in the orchard, growers should collect leaf samples early in the season. Three years of leaf sampling data was used to develop an approach to use spring-collected leaf samples to predict late summer leaf nutrient status using stepwise multiple linear regression models. These models were developed based on data from four geographical locations in California and from three seasons (2009, 2010 and 2011) and were extensively validated using data from wide geographic scope in 2012. The general agreement between the observed leaf N and predicted leaf N in July suggest that these models are robust, practical and useful predicting tools and these models can be used to produce effective estimates of the nutritional status of pistachio trees in late summer based on leaf samples measured early in the season. This protocol is of fundamental importance and offers large potential benefits to growers as it allows the growers for adjusting their fertilization programs within the current season.
• To utilize the spring sample nutrient prediction program growers must collect leaves from at least 18 trees in one bag each spaced at least 25 yards apart. Leaf samples should be collected between 30-45 days after full bloom.
• The leaf sampling protocols for sample size determination and early season sampling are for fully exposed non-fruiting sub-terminal leaves to be collected at 6-7 feet height from around the tree canopy. These sampling protocols are valid for orchards of average variability.
  

  Figure 2. Changes in pistachio leaf (non-fruiting branches) nutrient concentations over the seasons (averages: 2009 & 2010).

Top of page

Objective 2: Validate current critical values and determine if nutrient ratio analysis provides useful information to optimize fertility management 

• Extensive correlation and regression analysis between yield and tissue nutrient concentrations were performed on all data sets to determine critical values and determine if nutrient ratio analysis was a viable alternative to traditional critical value analysis. For all elements except magnesium (Mg), no evidence was found that the current critical values are incorrect. Nutrient ratio analysis showed significant relationships at only one site (Madera County). 
• Comparison of leaf tissue value and yield of all experimental tree samples over four sites and four years were used to assess the consistency of the current observational data and published critical values. Critical values of nitrogen (N), phosphorus (P), potassium (K), magnesium (Mg), and calcium (Ca) was validated. Due to contamination by foliar application the critical values for micronutrients including copper (Cu), Iron (Fe), Zinc (Zn) and Boron (B) was not validated. The individual orchards in this trial were selected for their yield and excellent management and as a consequence very few deficient trees were observed. Estimating orchard level CVs is difficult as nutrient rate trials were not attempted and all orchards used here were well managed and not apparently deficient in any nutrient. While the absence of deficient trees prevents the determination of the exact critical value, data from all trees in each year can often be used to determine the nutrient concentration above which yield is optimized. Here we took the approach that the lack of a yield response across a range of sites and years is an indication that the CV is at least lower than the lowest recorded values. With the exception of Mg, we find no evidence that CV’s for nutrients in pistachio need to be adjusted. For Mg, evidence suggests the current critical value is high and should be reduced to 0.45%

Magnesium and Potassium Relationship

• Magnesium (Mg) and potassium (K) are essential plant nutrients. The interactions between these important nutrients and the influence of these interactions on pistachio yield remain largely unknown. In the present study, field observations made over the 2009 and 2010 growing seasons suggested that the Madera County site was suffering from Mg deficiency. These results suggested that K and Mg interact competitively with high K applications resulting in decreased tissue Mg and yield depression. This was a striking result as yield depression as a consequence of high K has not often been observed. Understanding this interaction is of importance for management of these inputs and yield sustainability. To understand the K and Mg relationship, we investigated this phenomenon in the field and under greenhouse conditions over the season in 2012.
• Field results at Madera County site suggested that the yield response to K depends on the Mg status of the plants. This suggests that the Mg status of the plants is important for achieving the positive impact of K on pistachio yield.
• Pot studies on second leaf pistachio trees (Kerman grafted on UCB1 rootstock) in the greenhouse confirm that the K and Mg interaction is a real physiological response and not a result of the unique soil mineral characteristics of the soil at the Madera County site. We can thus predict that any location in which Mg is marginal could be negatively impacted by supplemental K and that this yield decline can be managed by enhancing tissue Mg.

Top of page

Objective 3: Develop a nutrient budget model for pistachio 

• Nitrogen (N) is one of the primary fertilizers used in California pistachio orchards to maintain productivity. However, excessive amounts of N fertilizers can result in ground water contamination without improving crop yield or maximizing grower returns. The current use of leaf sampling to manage N applications does not provide any specific information on the right rate and right time of the fertilizer applications. This uncertainty has led the growers to apply higher rates of N fertilizer without considering the actual tree demand and spatial and temporal variability that inherently exists in every field. The variability assessment (spatial and temporal) indicates that managing N by managing for spatial and temporal variability is critical to efficient N management. Thus N applied at uniform rates across the orchard ignores the spatial variability and is at considerable risk for environmental losses.
• Growing environmental concerns related to N fertilizer are pressuring growers to improve nutrient management practices (especially for N) and one approach that can help achieve that goal is to adjust N applications based on tree demand as determined by N requirement for growth and yield.
• To complement the leaf tissue analysis and provide guidance for fertilization, we have developed yield and phenology based seasonal nutrient removal curves that quantify the time course of nutrient uptake and total plant demand across different environmental conditions for major nutrients, including nitrogen (N), phosphorus (P) and potassium (K).
• Evidence from three years results suggests that considerable improvement in N use efficiency of > 90% could occur with implementation of demand based fertilization programs. Higher NUE would result in decreased residual N in the soil profile below the root zone at the completion of the season. Therefore, fruit load must be considered before application of fertilizers. In the estimate of NUE we have included nutrient demand for growth. Evidence from our research on the fertilizer N economy analysis suggests a potential savings of approximately US$40,000 per 1000 acres.
• The average NPK removal from three season suggest that twenty-eight (28) (lbs) of N, 24 (lbs) of K and 3 (lbs) of P are removed per 1000 (lbs) of marketable yield (CPC). This value includes all nutrients removed in hulls, shells and kernels, blank nuts and other non-marketable yield per 1000 lbs. For example, a crop that results in 4000 lbs CPC yield would remove from the orchard 112 (lbs) N, 96 (lbs) K and 12 (lbs) P. Evidence from prior research suggests that an average of 25 (lbs) N and 22 (lbs) of K is utilized to support tree growth requirements in a mature tree (>10 years old).

Figure 3a & 3b. Yield variability within a single well managed commercial pistachio orchard. High NUE can be achieved if annual nitrogen application rates are synchronized with the actual tree demand.

Figure 3a (above) Kings County experimental site. 2009.

Figure 3b (above) Fresno County experimental site. 2010.

Figure 4. Average nutrient removal (2009-2011) per 1000 (lbs) dry yield (CPC).

2013 Update | Findings | Model