Skip to content

Landscape sensitivity to P losses: biogeochemical analysis of agricultural soils

Principal Investigator: Merrin L. Macrae

Research Institution: University of Waterloo

Timeline: April 2016 – May 2019


  • Determine soil stratification of plant available phosphorus (P) across different soil textures in agricultural soils.
  • Characterize soil P retention to determine how “tightly” bound P is within soil and which environmental conditions may release P.
  • Determine the soil P sorption capacity to give an idea of how much P can be held in the soil and how close to saturation the soils are and if this varies with depth.


  • Understanding the relationship between surface and subsurface soil test phosphorus (STP) and dissolved phosphorus (P) runoff losses will provide valuable insight into predicting the fate of applied P to agricultural soils, and may allow farmers to apply P more strategically, while minimizing environmental impact and maximizing profits.
  • Understanding the amount and type of P in subsoils may lead to a better understanding of the potential role of tile drains in P loss in different regions.

Scientific Summary: 

Phosphorus (P) losses from agricultural lands are major environmental, economic and political issues in the lower Great Lakes region because of their impacts on downstream water-quality. Predicting the efficacy of best management practices (BMPs) to minimize P loss (dissolved and particulate) has been challenging due to the fact that BMPs that appear to be effective in one region are not necessarily effective in another. Tile drains in the lower Great Lakes region of the USA exhibit different P loss patterns relative to tile drains in Ontario. For example, most Ohio studies report that most P in tile drainage is lost in a dissolved form, whereas most P is lost as particulate in Ontario. Although some of this variability has been attributed to soil cracking and preferential transport, this does not appear to be the only factor controlling P loss. It is critical that differences in soil P loss/retention potential are assessed before the efficacy of BMPs can be fully evaluated and to explain the clear differences seen between studies in the USA and Ontario. Soil texture and biogeochemical properties are important to soil P sorption capacity and thus storage of soil P within agricultural lands. As such, characterizing linkages between soil properties and P stratification is essential for understanding P movement and fate in agricultural subsoils. This is directly relevant to predicating P availability to crops and risk to downstream aquatic ecosystems and can be used to build predictive models.

This project provided a spatial assessment of soil P stratification of agricultural lands across the lower Great Lakes region of Southern Ontario and USA, establishing linkages with soil texture and biogeochemical properties to evaluate and explain the differences in reported P losses between these two regions. Soil cores were collected at 8 sites between Ontario (4 sites) and USA (2 sites in Ohio and 2 sites in Indiana) to establish P stratification of soil test P (STP) concentrations (Olsen, Bray, Mechlich-3) and soil textures (sand, silt and clay). Additional soil biogeochemical analysis on P sorption capacity and retention were also done on selected subsamples to further evaluate the nature of the P stored within these soils.

Objective 1: Soil P stratification of soil test plant available P

We collected soils from 4 sites (3 locations per field and composite of 3 cores per location) in Southern Ontario, 4 USA sites (2 in Ohio and 2 in Indiana). The sites were selected for similar management practices and to capture the natural variability in the landscape around the lower Great Lakes agricultural region (i.e., flat clay-plain soils in the USA, to hummocky loam-textured soils of mid-western Ontario). The University of Waterloo Biogeochemistry Group collected the soil samples from the Ontario fields, while Kevin King, Mark Williams and colleagues at the USDA collected the USA soils (Indiana and Ohio).

Soil P stratification was quantified using 3 soil tests of plant available P (STP) i.e., Olsen, Mehlich-3 and Bray-1 at 5 sampling depths (0-5, 5-15, 15-30, 30-60 and 60-90 cm) using standard methods. Soil texture (%sand, %silt and %clay content) was measured at all 5 sampling depths from each site using the standard pipette method.

Object 2: Geochemical characterization of soil P retention

Geochemical characterization of soil P retention from a subset of 3 soil depths (0-5, 15-30 and 60-90cm) at each site was quantified through a multi-step sequential extraction following Zhang and Kovar (2009) at the University of Waterloo. Soluble reactive P (Pi) was quantified in four operationally defined soil fractions: loosely adsorbed Pi on oxide surfaces (e.g., AIOx and FeOx) and Pi compounds soluble in bases, likely clay bound-P (NaOH + NaCI (i.e., Sol- Pi)); reductant soluble Pi (i.e., redox-sensitive metal oxides) (sodium citrate (Na3C6H5O7.2H2O) sodium dithionite (Na2S2O4)-sodium bicarbonate (i.e., CDB-Pi)); acid-soluble Pi for carbonate and apatite bound-P (HCI (i.e., HCI-Pi)). The residual soil was digested in 10% K2S2O8 in an autoclave at 120 oC for 2 H to target refractory organic and mineral associated Pi (i.e., Res-Pi). Extracts were analyzed colorimetrically according to Murphy and Riley (1962).

Characterization of soil composition i.e., pH (0.01 M CaCI2 at a soil to solution ration of 1:2), organic matter (loss on ignition 550 oC), carbonate content (loss on ignition 950 oC) and reducible Fe oxide content (colormetrically) was quantified for these soils at the University of Waterloo. A subset of soil samples was analysed for mineralogy (XRD).

Objective 3: Soil P sorption capacity

A subsample of 3 soil depths (0-5, 15-30 and 60-90 cm) at each site was analysed for 1) P sorption capacity, 2) degree of P saturation and 3) equilibrium P concentrations following standard EPA isotherm techniques. Briefly, 1.0 g air-dried soil (< 2- mm sieved) was added to 50 mL conical polypropylene tubes and 25 mL of a 0.01 M KCI solution containing standard solutions of 0, 0.05, 0.2, 0.3, 0.5, 5, 10, 25, 75 and 100 mg P/L as KH2PO4. The tubes were agitated on a mechanical shaker for 24h. The soil suspension was centrifuged and filtered (<0.45 um) and the filtrate Pi concentrations were determined colorimetrically using the ammonium-molybdate ascorbic-acid method.


Results of this study demonstrate distinct differences in both the physical and geochemical properties of soils between the two geographic regions. Loam-textured soils from mid-western Ontario, Canada, were calcareous and alkaline, with the highest concentrations of soil Total P. The majority (up to 90%) of the stored soil P was bound tightly in this region (i.e., likely bound with calcium (Ca)). In contrast, clay-textured soils from southern Ontario, Northeast Indiana, and northwestern Ohio, US, were more acidic, had a lower carbonate content and an overall lower P sorption capacity (i.e., lower ability to retain added P). Phosphorus stored in these soils were relatively more soluble (i.e., at higher risk of P mobility to runoff and tile drainage). As such, region specific management strategies, based on both hydrology and soil biogeochemistry, should be considered to help minimize P losses across the agricultural lower Great Lakes region.

Detailed results of these three study objectives were recently published during the last reporting period in the peer-reviewed Journal of Great Lakes Research (see below).

A follow up publication in 2022 used a subset of the study sites and soil results from the initial study to evaluate the influence of landscape position (i.e., landform topography) on soil P retention and mobility within agricultural farmlands. The follow-up study was reported in the peer-review Journal of Environmental Quality (see below).

In addition, the same analytical techniques from the initial project were later applied to determine soil P solubility and risk of P re-mobilization from agricultural riparian soils. The results were since published in 2022 in the peer-reviewed Journal of Environmental Management (see below).

External Funding Partners:

This project was funded in part through Growing Forward 2 (GF2), a federal-provincial-territorial initiative. The Agricultural Adaptation Council assists in the delivery of GF2 in Ontario.

Project Related Publications:

Plach, J.M., Macrae, M. L., Williams, M.R., Lee, B.D. and King, K.W. 2018. Dominant glacial landforms of the lower Great Lakes region exhibit different soil phosphorus chemistry and potential risk for phosphorus loss. Journal of Great Lakes Research, 44(5): 067-1067.

Plach, J.M., Macrae, M.L., Wilson, H.F., Costa, D., Kokulan, V., Lobb, D.A. and King, K.W. 2022. Influence of climate, topography, and soil type on soil extractable phosphorus in croplands of northern glacial‐derived landscapes. Journal of Environmental Quality, 51(4): 731-744.

Pluer, W.T., Plach, J.M., Hassan, A., Price, D. and Macrae, M.L. 2022. Retention of phosphorus in soils receiving bunker silo effluent. Journal of Environmental Management, 323:116147.