Enhanced phosphorus (P) release from flooded, anaerobic soils  have been extensively studied under summer temperatures, but not under cold temperatures with intermittent freeze-thaw events. We investigated the temperature and freeze/ thaw effects during flooding on the release of P to floodwater from soil monoliths (15-cm depth) collected from eight agricultural fields in Manitoba. Soil monoliths were flooded with reverse osmosis water and incubated for 56 d under simulated summer flooding (SSF, 22±1 ℃), or snowmelt flooding with intermittent freeze thaw (IFT, 4±1 ℃ with intermittent freezing) in triplicates. Redox potential (Eh), pore water and floodwater dissolved reactive P (DRP) concentrations, pH and concentrations of Ca, Mg, Fe and Mn were determined weekly. In seven soils, Eh decreased rapidly with days after flooding (DAF) under SSF to values < 200 mV, but not under IFT. Pore water and floodwater DRP concentrations significantly increased with DAF in all soils under SSF, and in seven soils under IFT. While DRP concentrations were consistently greater under SSF than IFT in four soils, other soils had similar concentrations at certain DAFs. Significant relationships between ion concentrations and redox status , that fitted both  IFT and SSF data in most soils, suggests that similar redox-driven mechanisms are responsible for the P release; however, less P is released under IFT than SSF, since soils were not severely reduced  under IFT.  Substantial P release in a few soils under IFT, appeared to be unrelated to redox status, suggesting other P release mechanisms that are not redox driven.

Intact soil monoliths were collected from eight agricultural fields located in the Red River Valley of Manitoba. Soil monoliths with an internal diameter of 10 cm and height of 15 cm were collected using 30 cm length polyvinyl chloride (PVC) tubes, each with two drilled holes (3 mm) at 10 cm from the bottom (5 cm below the soil surface) on  opposite sides of the tube. The bottom of the monolith was sealed using a PVC cap . To extract soil pore water, two Rhizon MOM 10-cm soil solution samplers with an outer diameter of 2.5 mm and 0.15-µm pore size were placed horizontally at 5-cm depth from soil surface through the drilled holes. Soil monoliths were flooded to a 5-cm head. Soon after flooding, a redox potential probe with a platinum sensor was also installed to a 5-cm depth in each monolith to measure the soil redox potential (Eh). Monoliths were incubated for 56 d under two flooding conditions; simulated summer (SSF), or simulated snowmelt with intermittent freeze thaw (IFT). The monoliths under SSF treatment were kept at 22±1 ℃ throughout the 56 d of incubation, while those under IFT were kept in a cooler at 4±1 ℃ with a 24-h freezing cycle at -18 ±1℃ every 7 d during the 56 d of incubation.

Pore water (or soil solution) samples at 5 cm depth were extracted by applying suction using a syringe attached to the end of the Rhizon-MOM samplers. Overlying floodwater samples were collected using syringes from the center of the flooded monolith. Pore water and floodwater samples were collected on a weekly basis throughout the 56-d incubation period and coincided with the third day after each freeze-thaw event.  Floodwater samples were filtered through 0.45-mm membrane filters. Dissolved reactive P concentrations of pore and floodwater samples were determined immediately after sampling by the molybdate blue color method and pH of water samples was measured using a Fisher Accumet AB15 pH meter. Water samples  were acidified with 50 µL of conc. nitric acid and stored at 4 ℃ prior to analysis of concentrations of Ca, Mg, Fe and Mn using flame atomic absorption spectroscopy (AAnalyst 400, PerkinElmer). Soil redox status (Eh) was measured at 7-d intervals (same day of water sampling using the pre-installed Pt redox probe and temporarily inserting a Ag-AgCl reference electrode with KCl electrolyte to the soil-floodwater interface

Iron and manganese concentrations were not not always detectable in pore water and  floodwater. The detectable limits for Fe and Mn were 0.05 and 0.16 mg/L, respectively.