Nitrogen (N) is the most limiting nutrient to crop production in the southwestern United States. The Lower Colorado River Region near Yuma is one of the most productive regions for vegetable production during the fall, winter, and spring growing seasons. To meet the strict quality standard requirements of the produce supply chain, ample amounts of N fertilizer are applied to these crops. In addition to crop uptake, which represents a beneficial disposition of soil applied nitrogen, N fertilizers applied to crops could be lost through various pathways (1). Urea and ammonium forms of N fertilizers can be lost through ammonia volatilization (2). Nitrate-nitrogen can readily be leached below the crop root-zone following irrigations (3) or can be lost to the atmosphere as molecular nitrogen (N₂) and nitrous oxide (N₂O) gases (4). Furthermore, inorganic N might be immobilized into the organic soil fraction through the actions of soil microbes, resulting in delayed crop availability (5). 

Over the past two decades, University of Arizona and University of California researchers have developed strategies for efficient N management. Recommended practices include pre-fertilization plant and soil testing to determine soil nitrogen status, timing of fertilizer applications, and improved irrigation management (1,6,7). Management guidelines and practices pertaining to efficient irrigation practices for vegetables in the Lower Colorado River are available elsewhere (8, 9, 10). The following discussion specifically addresses N fertilizer management. 

Nitrogen fertilizers are available in the form of urea or some blend of urea, ammonium (NH₄), and nitrate (NO₃). Following soil application, each of these forms eventually produce nitrate through the processes of hydrolysis and/or nitrification. As a result, nitrate is typically the most prevalent form of plant available N in the root zone. Notably, the negatively charged nitrate ion is mobile in the soil due to the net negative charge of soil minerals and can readily be leached below the crop root zone by rainfall and irrigation. Therefore, timing N applications in accordance with crops demand is a good strategy to avoid undesirable N losses (1, 7).  Data basis for seasonal crops N accumulation patterns have been compiled by others (7, 11, 12, 13) and are useful for timing N applications.

Vegetable crops require pre-plant application of N fertilizer, when the pre-plant soil N levels are less than 10 ppm nitrate-N. Prior to the start of the fall vegetable growing season, most vegetable fields receive a pre-irrigation to restore salt balance (9). Results from a survey with 316 samples collected from vegetable fields after pre-irrigation and prior to planting showed 68% of the samples collected had nitrate-N levels less than 10 ppm, and 56% had nitrate-N values less than 5 ppm (14). In addition to N, phosphorus (P) fertilizer is applied pre-plant. Monoammonium phosphate (MAP or 11-52-0), which contains N, is the primary source of P fertilizer in the region. As a result, the pre-plant N requirement is often met with P fertilization. However, under conditions where pre-plant P requirements are low due to high soil P levels, pre-plant application of N fertilizer could be required (Table 2-1). This N fertilizer might be applied pre-plant broadcast, or alternatively, the early season N requirements might be applied through sprinkler irrigation during germination and stand establishment. For most crops in the area, solid set sprinklers are used for germination and stand establishment (9) then all following irrigations are by furrow.

For furrow-irrigated vegetable crops, in-season N applications are most often made by tractor-applied side dress injections for optimal uniformity of application. Naturally, there are logistical tradeoffs as to when tillage, irrigations, and crop growth status allow opportunities for tractor-applied side dress N fertilizer applications. In the past, more than one side dress application was often used, with the goal of minimizing various forms of fertilizer loss and maximizing plant uptake. However, more recently for logistical and economic reasons, growers principally rely on one side dress application after thinning and during cultivation. There is also a potential for subtle damage to roots later in the season from side dress operations (14). In the Yuma area, vegetable crops are typically grown in loam to clay loam textured soils and under efficient irrigation management (8, 9) with minimal late season leaching. For coarse textured soils, a more frequent and lighter application of N would be required for optimal N use fertilization efficiency. This would not be logistically feasible for tractor-applied side dress in furrow irrigation systems. In these situations, the use of controlled-release N fertilizer, or fertigation depending on irrigation infrastructure, should be considered. Discussion of potential applications of controlled-release N fertilizers is in section 4.5. Fertigation is discussed in section 4.4. 

Plant tissue tests were often used to diagnose soil N deficiencies. Doerge et al. (7) complied sampling protocols and critical test values for a range of economically important crops in Arizona, including vegetable crops. The protocol typically involved nitrate extraction of plant petioles, midribs, or stems. Subsequent studies of leaf petiole nitrate tests for cotton (15) and stem nitrate tests for wheat (16) confirmed earlier results. For vegetable crops, soil nitrate tests were found to be more reliable than tissue tests. Tissue nitrate levels are dependent on what part the plant the sample was taken and the plant stage of growth at the time of sampling. Therefore, minor deviations from specified sampling protocol may confound interpretation. Furthermore, there is often diurnal variation in nitrate-N levels in plant tissue associated with changing sunlight and temperature, which impacts metabolic processes of plants such as nitrate reductase. The diagnostic accuracy of tissue tests is found to be sound when sampling protocols are rigorously adhered to. However, the application of plant tissue tests appears to be more diagnostic for vegetable crops, whereas soil nitrate-N tests can have prognostic applications. 

In a series of studies conducted in lettuce, broccoli, and cauliflower fields in the Lower Colorado River Region, side dress N application in the context of realistic commercial production scenarios were evaluated. Evaluations conducted in grower fields found 50% of the tissue test predictions were either false positives, where N deficiency was predicted but no response to N application was observed, or 25% false negatives, where N deficiency was not diagnosed but response to N application was detected (see Figure 1-1 and Table 1-2). False positives are often due to extraneous environmental factors that are beyond growers’ control, limiting production below potential, and are expected and tolerable to a certain degree. However, false negatives can lead to yield reduction, and hence, economic consequences. In the same study, soil nitrate-N testing produced less than 5% false negatives. Thus, the in-season N fertilizer recommendations presented in this guideline are based on a soil nitrate-N test. Recommendations, which can vary by crop and management system, are presented under the crop sections. Researchers at the University of California have had similar success using pre-side dress nitrate-N tests (17, 18, 19).

The University of California has developed a quick soil testing protocol using test strips as an alternative to laboratory analysis (20, 21). This quick test correlates well with laboratory nitrate-N test results and is a viable alternative when a rapid turnaround of results is needed. This protocol for using the quick nitrate test is outlined at (21). Results of a study conducted in Arizona also showed that the quick test compares well with the traditional nitrate-N tests (14).

Sometimes supplemental N may be needed later in the season when crop size will not allow for tractor-applied side dress operations. This later need for N might result from N leaching due to above-average rainfall, less than ideal irrigation management, or longer irrigation lengths or production on soils coarser in texture than the norm under which this furrow system was optimized. Thus, in-furrow fertigation may be required. This decision should also be based on in-season soil nitrate-N tests.

Data shows no difference was found among the alternative soluble N sources applied through side dress in terms of their effect on crop yield (see Table 1-1). These findings agree with previous work in Arizona (22). Hydrolysis (2) and/or nitrification processes in soils (23), both processes that produce NO₃ from the commercially available N fertilizers, are slow under cooler temperatures of the vegetable growing season. However, the lower crop growth rates associated with lower temperatures result in correspondingly slower rates for N demand. Furthermore, no difference was found in crop yield between soluble N sources applied through fertigation with furrow or drip systems (14). Urea ammonium nitrate solution (UAN 32) has been successfully used in the context of fertigation with sprinkler systems. Others report benefits from changing to sources with a higher proportion of nitrate-N for sprinkler fertigation during cooler months, and we do not discourage this practice.

            The University of California N fertilizer management protocols stress that the contribution of the background N content of irrigation water to the soil N needs to be considered in determining the crop N fertilizer requirement. This recommendation is appropriate in California where many ground water sources used for irrigation have high nitrate-N concentrations. However, almost all vegetable crops produced in the Yuma region use Colorado River water for irrigation, which has low nitrate-N levels, typically less than 0.1 ppm (less than 0.2 lbs N/acre foot water). Thus, the contribution of background N content of the irrigation water to soil N is negligible in the Yuma area. There are areas of Arizona where ground water sources are used for irrigation. Available data from source wells in the Central Arizona Irrigation and Drainage District (CAIDD) and Maricopa Stanfield Irrigation and Drainage District (MSIDD) (14) show that nitrate-N concentrations ranging from 0.2-39 mg/L. These data suggest that in areas of Arizona where ground water is used for irrigation, the nitrate-N contribution of irrigation water may need to be considered in the determination of total N input.