The movement of excess phosphorus (P) into streams, rivers, and lakes poses a significant threat to water quality and the health of aquatic ecosystems and thus, P has been targeted for reduction. In landscapes dominated by agriculture, P is primarily transported through non-point sources, which a number of best management practices aim to target. One such practice is vegetated buffer strips (VBS), which are designed to use dense vegetation above the surface and extensive root systems below the surface to reduce runoff velocity, trap sediments, increase infiltration, and increase plant uptake of nutrients. The effectiveness of VBS in reducing P concentrations has been studied and reviewed, but most studies have been undertaken in warm or temperate climates, where runoff is primarily driven through summer rainfall events and when vegetation is actively growing. In cold climates, the majority of runoff occurs during the snowmelt period, when soils are frozen and vegetation has been flattened by snow and ice over the winter period and is not actively taking up nutrients. These conditions hinder the ability of VBS to work as designed. Additionally, frozen vegetation can release P after undergoing freeze–thaw cycles (FTCs). Thus, this review aimed to (i) summarize research designed to determine the effectiveness of VBS in reducing P transport in cold climates, (ii) collate research on the potential for vegetation to release P after undergoing FTCs, and (iii) identify research gaps to be addressed in determining VBS effectiveness in cold climates. Cold-climate VBS implemented in Canada, the northern United States, and northern Europe have shown P removal efficiencies ranging from −36% to +89%, a range that identifies the uncertainty surrounding the use of VBS in these landscapes. However, there is consensus among researchers globally that vegetation does release P after undergoing FTCs, though P concentrations from different species vary across studies. The design and management of VBS in cold climates requires careful consideration, and VBS may not always be the best management strategy to reduce P transport. Future research should be undertaken at a larger scale in natural systems and focus on VBS design and management strategies.

A vegetated filter strip (VFS) is a measure commonly implemented in agricultural landscapes for the purpose of improving water quality. However, much of the evidence to support their effectiveness comes from warm regions dominated by rainfall driven runoff. This study assessed the performance of VFS plots and compared them with annual crop strips to reduce phosphorus (P) levels in runoff in the cold climate of the Canadian Prairies. Analysis of water samples from 22 events during the study indicated no significant difference in the inflow and outflow concentrations of total dissolved P (TDP) or total P (TP) for either the VFS or the annual crop strips. Although the VFS plots had little effect on TDP or TP during the spring, they performed better during the growing season, reducing mean TP concentrations in five out of seven, or 71%, of these events. The VFS plots did not perform as well during the fall events, with the overall mean TP concentration in runoff increasing after flowing through the filters during this time period.

Documenting the effects of agricultural land retirement on stream-sediment sources is critical to identifying management practices that improve water quality and aquatic habitat. Particularly difficult to quantify are the effects from conservation easements that commonly are discontinuous along channelized streams and ditches throughout the agricultural midwestern United States. Our hypotheses were that sediment from cropland, retired land, stream banks, and roads would be discernible using isotopic and elemental concentrations and that source contributions would vary with land retirement distribution along tributaries of West Fork Beaver Creek in Minnesota. Channel-bed and suspended sediment were sampled at nine locations and compared with local source samples by using linear discriminant analysis and a four-source mixing model that evaluated seven tracers: In, P, total C, Be, Tl, Th, and Ti. The proportion of sediment sources differed significantly between suspended and channelbed sediment. Retired land contributed to channel-bed sediment but was not discernible as a source of suspended sediment, suggesting that retired-land material was not mobilized during high-flow conditions. Stream banks were a large contributor to suspended sediment; however, the percentage of stream-bank sediment in the channel bed was lower in basins with more continuous retired land along the riparian corridor. Cropland sediments had the highest P concentrations; basins with the highest cropland-sediment contributions also had the highest P concentrations. Along stream reaches with retired land, there was a lower proportion of cropland material in suspended sediment relative to sites that had almost no land retirement, indicating less movement of nutrients and sediment from cropland to the channel as a result of land retirement.

No abstract

No abstract