Potassium (K) fertilizer is another macronutrient required for satisfactory crop growth and yield in many parts of the United States. However, no discernible economic returns are generally observed in response to K fertilization in the Lower Colorado River Region. Much of this lack of response is associated with the mixed clay minerology of the soils of the region. While smectites are the dominant mineral in most soils used for vegetable production in the region, all soils contain K bearing mica and negligible amounts K fixing vermiculite (36). Studies conducted in the region (37) showed potential for substantial K release, by K bearing clay minerals, during and/or following irrigation events. 

There is also K in most ground and surface waters of the region. For example, Colorado River water typically contains 4.9 mg/L of K, which is equivalent to about 13 lbs of K/acre foot water. Surveys showed that most groundwater sources in the Central Arizona Irrigation and Drainage District (CAIDD) and the Maricopa Stanfield Irrigation and Drainage District (MSIDD) have similar concentrations of K with that of the Colorado River water. Although the background concentrations of K in irrigation water alone are insufficient to meet crop demands, they nonetheless are a contributing factor to the general lack of crop response to K fertilization in the region.

Another contributing factor is likely the presence of sodium (Na) in soil and irrigation waters, an ion known to play a substitute role, to a degree, for K in osmotic adjustment mechanisms in many vegetable crops (38, 39, 40, 41, 42). The data in Figure 2-3-1 show celery response to both P and Na soil test levels. These data were collected in the Eastern United States by the author and are presented here to show the potential for Na to play a substituent role for K in crops with high potassium requirements. Note that with potassium consumptions of up to 500 lbs of K/acre at harvest, celery is among the crops with a high K requirement. It is important to stress that the experiments Figure 2-3-1 is based on could not have been conducted in Arizona due to high background levels of both K and Na in the soil and water systems. The Na concentration in the Colorado River water near Yuma is about 100 ppm, or 272 lbs of Na/acre foot water.

High Na in irrigation water can have adverse effects on crops and soils. It can be toxic to crops if accumulated in plant tissues at levels that exceed crop tolerance thresholds, but more often Na has deleterious effects on soil structure (43). The risk related to soil structure is assessed in terms of an index known as sodium adsorption ratio (SAR) of the irrigation water (SAR_w), or the soil solution (approximated by SAR_e), and depends on the ratio of the concentration of Na in the solution to those of the divalent cations of calcium (Ca) and  magnesium (Mg). The SAR_w of Colorado River water in Yuma is about 2.7, a level that can be considered too low to produce adverse effects on soil structure. The low SAR_w of the Colorado River is due to the generally high Ca (about 80 mg/L) and Mg (27 mg/L) concentrations. 

By comparison, the SAR_w varies between 2 and 12 for the groundwater used in CAIDD and MSIDD. The high SAR_w of the irrigation waters used in these districts is the result of a combination of higher Na and generally lower Ca and Mg concentrations in some of the groundwater sources. For parts of the irrigation districts that use water with high SAR_w, sound irrigation and salinity management strategies may need to be followed to prevent adverse effects to soils and crops. However, given that Na can substitute for K to a degree, the presence of Na below problematic can be potentially beneficial to crops. 

Availability of K in soils is typically assessed using an ammonium acetate (AA) extraction test (44). The test is based on the assumption that it is the exchangeable K that constitutes the crop available fraction of soil. 

Because crops in Arizona have generally not responded to K fertilization, we do not have locally calibrated soil test critical levels. The provisional critical soil test levels shown in Table 2-3-1 are compiled from several western sources (45, 46, 47, 48, 49, 50, 51, 52, 53). These reference values should be used for soil test monitoring of K in the event natural soil sources of K become depleted. With a few exceptions, 150 ppm soil test AA K is sufficient for most crops, and most soils currently used for vegetable production in Arizona test above 150 AA K (14, 36, 37). However, in Arizona, even when soil test results show less than 150 ppm AA K, crop response to K fertilization are not certain for the reasons described above. Notably for most Arizona soils, soil test AA Na levels are typically within the same order of magnitude as AA K levels.