Phosphorus Characteristics of North Carolina Soils and Riparian Buffers

Abstract

Non-point phosphorus (P) losses from agricultural fields have the potential to accelerate eutrophication of North Carolina (NC) surface waters. In 1998, the US Environmental Protection Agency reported as much as 60% of the US fresh waters were impaired by non-point source nutrient pollution stemming primarily from agricultural contributions (EPA, 1998). Phosphorus has been marked as the most limiting nutrient in fresh waters (Pierynski et al., 2000). Yet in NC, historically excessive applications of P in the form of fertilizer, i.e. tobacco fields, and manures has resulted in elevated soil P concentrations well above levels required to optimize crop performance and yield. In agricultural fields with elevated soil P concentrations, the risk of P loss to surface waters is also elevated.

Soil testing, i.e. Mehlich-3, has been identified as an important factor in determining P losses from soils. However, there in a lack of information on how fertilization and initial Mehlich-3 P (M3P) interact to affect water soluble P (WSP) in soils. Such soil dependent relationships may assist in ranking agricultural fields from low to high on their potential to lose P in the soluble form. Overall, our objectives to validate the use of the Mehlich-3 soil test as a useful indicator of P loss risks from fields receiving animal wastes. In doing so, soil or soil group dependent critical M3P and/or M3P-saturation ratio thresholds (or change points) may be identified.

In an incubation study (Chapter 2), our objectives were to (1) quantify the relationship between WSP and M3P for four textural diverse benchmark soils of North Carolina (NC), and (2) quantify the change in WSP concentrations following P additions to soils over a wide range of initial M3P. One hundred and seven samples known to represent a wide range in M3P were collected from an Autryville loamy sand, Wasda muck, Georgeville silt loam, and Pacolet sandy clay loam and analyzed for M3P, Fe and Al and WSP. An incubation study was also conducted where four samples representing a range in M3P from each series were fertilized at rates of 150 and 300 kg P ha-1, and WSP was measured at 1, 7 and 21 d after fertilization. The Wasda muck exhibited a change point at 115 mg P kg-1 across a broad range of M3P concentrations (60-238 mg kg-1) while Autryville, Georgeville, and Pacolet series (with ranges in M3P of 32-328, 119-524, 0-1034 mg P kg-1, respectively) maintained linear relationships between WSP and M3P. For the fertilized soils, significant increases in WSP occurred regardless of P rate. However, greater increases in WSP were observed following P application to soils of relatively greater initial M3P. As WSP is linked to soluble P losses in runoff, these data suggest that shifting animal waste applications to fields of relatively lower M3P concentrations would have an immediate impact on reducing risk for P losses, if all other factors are equal.

In a P leaching study (Chapter 3), our objectives were i) determine the extent of P leaching in Piedmont and Mountain soils amended with animal wastes, (ii) assess the validity of grouping soils into soil management groups to more accurately predict P leaching, and (iii) evaluate soil factors affecting the relationship between Mehlich-3 P (M3P) and soluble (CaCl2-extractable) P. Twenty-seven soils from 7 Piedmont and Mountain counties were collected to a maximum depth of 91 cm. All soils were analyzed for M3P, Fe, Al, and CaCl2-P. Despite the historically excessive applications of P in the form of animal wastes to NC Piedmont and Mountain soils, the net accumulation of P did not excessively (M3P>53 mg kg-1) accumulate to depths below the plow-layer (upper 30 cm of soil profile) of most soils studied. The finer textured, Fe-oxide dominated soils of these physiographic regions of the state represent the importance of these soils properties in fixing P and inhibiting P leaching. In comparison, a coarser textured floodplain soil exhibited lower P sorption capacities and illustrated a greater P leaching potential. Soil management groups currently employed by the PLAT appear to be justified for use in predicting soil P leaching potentials. From the relationship between the M3PSR and CaCl2-P, we identified mineral-dependent critical environmental M3PSR thresholds for these soils at 0.06(M3PSR) and 0.15(M3PSR), above which potentially greater risks of soluble P losses via surface or subsurface pathways may prevail.

In the middle Coastal Plain of NC, soil dependent P leaching potentials were also evaluated as well as the associated risk of P loss to nearby surface waters via subsurface lateral transport (Chapter 4). Leaching of M3P in NC coastal plain soils to depths exposed to the water table (often <1 m from the soil surface) may increase the risk of soluble P loss to surface waters, but may be mitigated by riparian buffers (RBs). Our objectives were to evaluate NC Coastal Plain soils to (i) validate the use of M3P to predict P leaching in these soils, and (ii) evaluate the effects of RBs for attenuating potential subsurface P losses from high P fields. Eleven soil series were sampled to a maximum depth of 165 cm along transects oriented in the direction of groundwater flow leading from the field into bordering riparian buffers. Soil drainage classification appeared to influence the pathway(s) by which P losses to neighboring RBs occurred from fields. "Drier" soils had greater potentials for surface P loss than "wetter" soils. In general, the "drier" soils exhibited greater P sorption capacities than "wetter" soils as exhibited by their respective M3PSR change points values. The accumulation of humic matter (HM) content in "wetter" soils resulted in lower P sorption capacities and greater potential for P leaching and subsurface lateral P losses. Overall, excessive applications of swine waste resulted in M3P leaching in all NC Coastal Plain mineral soils. In well-drained soils, P loss occurred as a surface phenomenon as opposed to subsurface despite excessive P leaching losses in the field. Yet, greater surface (˜75 cm) HM content resulted in less P sorption capacity (M3PSR) and greater M3P leaching than all other soil series. In reference to the Pantego loam M3P interpolation, the data suggests a real potential for lateral (from the field) subsurface (>120 cm) M3P accumulation in the RB. Due to limited data, no apparent trends in groundwater soluble P were evident. Therefore, further research and emphasis should be placed particularly on soils with relatively greater organic matter to better understand the potential risk for P losses to nearby surface waters so that a more sustainable approach to P management may be developed and implemented.