USGS - science for a changing world

South Florida Information Access (SOFIA)

publications > paper > fertilizer-derived uranium and sulfur in rangeland soil and runoff: a case study in central Florida > results and discussion > estimates of fertilizer-derived uranium in soils

4. Results and Discussion

Site Desc. & Land Use
Results & Discussion
- Soil Composition
> Uranium in Soils
- Sulfur in Soils
- Runoff
- U, S, and P Mobility
Figures, Tables, & Equations


Ranchland soils of this study are not subject to tillage, and therefore are more likely to preserve evidence of minor U addition to uppermost soil layers. Fertilizer U stored in these layers could originate from previous direct applications of phosphate fertilizer or be added by runoff originating from historically fertilized upland areas. Another possible source of U could be manure, to the extent that consumed vegetation contains U. Uranium is a nonessential element for plants and is present in low to very low parts-per-billion (ppb) levels in fresh plant tissue (Mortvedt, 1994; Linsalata, 1994). For this reason, native vegetation and vegetation-based nutritional supplements such as molasses should not contain significant U. A 20-fold concentrate of sugar cane leaves produced by ashing contained undetectable U (< 200 ppb) by delayed neutron analysis (Zielinski et al., 2000). In feedlot settings, manure may contain U present in the phosphorus mineral supplements of cattle feeds (Reid et al., 1977). In one study of dairy cattle, virtually all of ingested U was found to pass with the feces (Chapman and Hammons, 1963). For the rangeland cattle at MAERC, contributions of U from ingested plants or artificial feeds are likely to be minimal, eliminating manure as a significant U host. Export of U from improved pasture such as S5 is primarily via dissolution of U that originally resided in soil or fertilizer particles.

Graph showing variations of the uranium isotopic composition of soil extracts with depth in the three soil cores
Figure 3. Graph showing variations of the U isotopic composition of soil extracts with depth in the three soil cores. Analytical precision (2-sigma) is approximated by the error bars. The U isotopic composition of locally applied phosphate fertilizer is indicated for comparison. [larger image]

Graph showing the relation between organic carbon content of soil intervals and the percent of total uranium removed in a selective extraction with 0.1 molar sodium bicarbonate
Figure 4. Graph showing the relation between organic carbon content of soil intervals and the percent of total U removed in a selective extraction with 0.1 M sodium bicarbonate. [larger image]

Concentration-based estimates of the amount of "excess" U in historically fertilized S5 soil can be made by comparing the average U concentration in the upper 15 cm (1.2 ppm) to the average concentration in the lower 15 cm (0.8 ppm) (Table I). A similar comparison for profile 770 indicates similar relative enrichment in the uppermost 15 cm (0.5 ppm) compared to the lower 15 cm (0.35 ppm). The higher U concentrations in near-surface soils could be interpreted as a natural consequence of higher organic matter contents and the ability of soil organic matter to sorb dissolved U. The native grassland profile (Lykes), however, shows no apparent surficial enrichment of U (Table I; Figure 2).

Uranium isotopic compositions of loosely-bound, extractable U were used to estimate the amount of fertilizer-derived U in soil profiles. Uranium isotopes are not fractionated during sorption or desorption and thus the AR of soil extracts represents the time-averaged isotopic composition of dissolved U in contact with soil-based sorbants such as organic matter. Extracts from the upper soil layers of S5 have the lowest measured 234U/238U activity ratios (Table I; Figure 3). These low AR values of 1.039 and 1.050 fall within the narrow range of 1.0 ± 0.05 reported for commercial phosphate fertilizers derived from acid treatment of phosphate rock (Zielinski et al., 2000). Isotope-based estimates of the amount of fertilizer-derived U in these extracts involve calculations based on the contrasting isotopic composition of fertilizer U and natural U. An AR of 1.020 ± 0.002 for the fertilizer end member is based on the average value of locally obtained superphosphate products (Table II) that are applied in pure form or as part of fertilizer blends. Also included in Table 2 are AR values for five nutritional mineral supplements to cattle feed that contain variable amounts of U and P. The narrow range of AR values for these supplements (1.015 to 1.033) also falls within the range for phosphate fertilizers and indicates a probable phosphate-rock origin for the P. An AR of 1.127 ± 0.022 for natural U is based on the average of three extracts from the native grassland profile (Lykes) and two extracts from the lower portion of the semi-native profile 770 (Table I). Extractable U from these lower soil layers is assumed to represent U sorbed onto organic matter under natural or pre-fertilization conditions. Extractable fractions of U in the 10 treated soils ranged from 5 to 42 percent and correlate positively with organic matter content (Table I; Figure 4).

If an extract contains a mixture of desorbed natural U and fertilizer-derived U, then the mass contribution of U from each component can be calculated using a standard isotope mass balance equation:

ARext. = (F) x (ARfert) + (1 - F) x (ARnat)     (1)

where F: mass fraction of U from fertilizer, 1 - F: mass fraction of natural U, AR = 234U/238U activity ratios of fertilizer and natural end members (see above), and the activity ratio of a measured extract.

Rearranging and solving for F yields

F = (ARext - ARnat)/(ARfert - ARnat)     (2)

Based on this calculation, 82 weight percent of the uranium extracted from the 3-6 cm interval of S5 is fertilizer-derived. Extractable U in this interval was 42 percent of total U (Table I), so 34 weight percent of total U in this interval is calculated to be fertilizer-derived. This estimate is a minimum if the extraction did not yield all readily extractable U or if some fertilizer U is present as less soluble particles. A similar calculation for the 9- to 12-cm section of S5 indicates that 72 weight percent of extractable U and 15 weight percent of total U is fertilizer-derived. The concentration of remaining unextractable U in these uppermost intervals is on the order of 0.8-0.9 ppm, similar to concentrations in organic-poor lower intervals (Table I).

The 3- to 6-cm section of semi-native profile 770 may also contain some fertilizer-derived U, based on the AR of 1.050 in its extract. Minor amounts of fertilizer-derived U could be introduced during flooding or by undocumented applications of fertilizer. Calculations indicate that 72 weight percent of extractable U and therefore 27 weight percent of total U in this upper interval may be fertilizer-derived. Unextractable U concentration is on the order of 0.6 ppm, similar to concentrations in organic-poor lower intervals (Table I).

Uranium concentration and isotopic composition of phosphate-bearing fertilizers and cattle feed nutritional supplements used in the study area
Sample U (ppm) 234U/238U activity ratio
  IMC triple superphosphate 0-46-0 176 1.019 ± 0.001 (n = 2 replicates)
  Hi Yield superphosphate 0-49-0 182 1.021 ± 0.001 (n = 2 replicates)
Nutritional Supplements
  PDQ-7 44.6 1.015
  Big Red 11.3 1.033
  RPB 45.4 1.016
  12:12 104 1.016
  BIOFOS 203 1.017
Range of reported values for commercial fertilizers (Zielinski et al., 2000) 20-300 1.0 ± 0.05

< Soil Composition | Sulfur in Soils >

| Disclaimer | Privacy Statement | Accessibility |

U.S. Department of the Interior, U.S. Geological Survey
This page is:
Comments and suggestions? Contact: Heather Henkel - Webmaster
Last updated: 04 September, 2013 @ 02:04 PM(KP)