Effect of Soil Contamination on the Mineral Composition of Forage Fertilized with Nitrogen

Mineral analysis of forage from a nitrogen (N) fertilizer, field study produced unexpectedly high iron (Fe) concentrations which were correlated with the N fertilizer level (r2=.92) and the percentage N in the forage (r2 =.94). The high Fe values were presumed to be associated with dust on the leaves. The objective of this study was to determine the level of soil contamination on the forage sample and the contribution of mineral in the contaminant to that measured in the sample. Soil contamination of plant tissue samples was calculated from the dilution of soil titanium (Ti) assuming that the uncontaminated tissue contained 0 g Ti/g. Tissue harvested from the 0, 28, 56, or 84 kg N/ha treatments contained 23, 49, 48, and 60 mg soil/ g, respectively. Significant N fertilizer effects would have been accepted for each element tested if soil contamination had been ignored. Correcting for contamination resulted in significant N-fertilizer effects on the concentrations of sodium, potassium, manganese, iron, and zinc but not magnesium or calcium in the forage. Some of these effects may be explained by the acidifying effect of the N fertilizer source. Interest in mineral cycling under semiarid pasture conditions led to the investigation of elemental concentrations in forage that had received annual nitrogen (N) fertilizer applications during a 4-year period. Initial chemical analysis indicated that fertilizer N increased the concentrations of both N and iron (Fe) in the forage. Because Fe concentrations exceeded 300 pg/g, we suspected that the forage samples were contaminated with soil. Procedures based on the ability of the plant to descriminate against the uptake of soil aluminum (Al), Fe, and titanium (Ti) have been developed to measure the degree of soil contamination in plant tissue (Healy et al. 1974, Metson et al. 1979, and Nes 1975). Metson et al. assumed that Fe values of 80, 100, and 100 14/ g for grass, clover, and other herbage, respectively, were threshold concentrations between endogenous (absorbed internally by the plant) and exogenous (external as dust) Fe. They also assumed that threshold concentrations of Al were 70, 100, and 120 ki.g/g for grass, clover, or other herbage, respectively. In each case forage mineral concentrations were corrected for soil contamination based upon these threshold values of Fe or Al. They found reasonable agreement between the two estimates, although in some instances the amount of soil contamination estimated from the Fe values was considerably higher than that derived from the Al data. Nes (1975) reported good correlations between sample ash and Ti concentrations when piint samples were contaminated with known amounts of soil. Metson et al. (1979) reported good agreement in calculated contamination levels when using the Ti, Al, or Fe discrimination technique. However, a limited number of samples were used in the latter study. Titanium might provide a highly sensitive measure of soil contamination because of its low concentration, less than 3 pgig, in Authors are with the USDA, Agricultural Research Service, Kimberly, Ida. 83341 and Burns, Ore. 97720, respectively. Manuscript received May 5, 1982. plants and . the large discrimination factor of 10,000 to 500,000 between the concentration in soils and in plants (Mitchell 1960 and Shacklette 1980). These discrimination values are compared with values of 500 to 3000 for Fe and 1000 to 5000 for Al (Mitche111960 and Shacklette 1980), This advantage for Ti, however, is valid only if the analytical sensitivity is similar for each of the elements. The objectives of this study were to utilize the Ti method to measure the level of soil contamination, to identify endogenous levels of elements measured in plant tissue, and to determine the real effect of N fertilization on endogenous mineral concentrations in the forage tissue. Previous authors have given little detail to the methodology. Therefore, additional discussion is directed to the mineral dilution technique as a means to calculate endogenous mineral values. Methods and Procedure The experiment was conducted on a previously established field of crested wheatgrass (Agropyron desertorum (Fisch. ex Link) Schult.) located on a sandy loam xerollic camborthid soil on the Squaw Butte Experiment Station near Burns, Ore. The experiment consisted of a randomized block design with 3 replications. The plots were fertilized each autumn for 4 years with 0, 28, 56 or 84 kg Nf ha as ammonium nitrate (NH 4 NO3). The fourth year's growth was hand clipped at a 3-cm stubble height in mid summer, dried at 55° C, ground to pass through a 1-mm sieve, and stored over phosphorus pentoxide (P205 ) desiccant. Samples were analyzed for total N by the Kjeldahl procedure which was modified to include nitrate. Mineral concentrations were determined by atomic absorption with either flame or nameless (for Ti only) atomization of samples previously digested in a 3:1 mixture of nitric:perchloric acid (11NO3 :HC104) and appropriately diluted. Methods of soil (<100 pm) or titanium dioxide (TiO 2 ) additions to forage samples produced linear responses (r2 =.99) to Ti measured in the 0 to 500 lag/ g range. The coefficient of variation of these analyses was about 5%. Soil pH was determined on a saturated paste of a composite sample taken from 0 to 15-cm depth in each plot. A composite soil sample, collected from the 0 to 1-cm depth and passing through a 100-m sieve, was dry ashed at 550°C overnight and then digested with HNO3 :HC104 (3:1) prior to elemental analysis. The soil concentration of the tissue samples was determined from the dilution of soil Ti or Fe assuming 0 or 80 plgtg endogenous Ti or Fe in the plant, respectively. The endogenous mineral concentration (Me) in the forage was calculated as follows: Me = [M. (Tii-Mt/ Tii3 )] Tip/ Tiid where MP and Mg were the mineral concentrations in the plant and soil sample, respectively, and Tit; and Tit were the Ti concentrations in the plant and soil samples, respectively. For the Fe procedure, soil Fe concentration was substituted for Tit and plant Fe 286 JOURNAL OF RANGE MANAGEMENT 36(3), May 1983 Table 1. Mean mineral concentrations in soil and forage, the ratio of the mineral concentration in soil to that endogenous in the plant (calculated by the titanium dilution technique) and the endogenous mineral concentration in forage fertilized with four rates of N.