Language selection


Beneficial management practices for decreasing nutrient losses

The instillation of small dams and reservoirs as a beneficial management practice (BMP) was discovered to help reduce nutrient losses at the South Tobacco Creek Watershed. Alternatively, the application of multiple BMPs at the same watershed were found to accomplish the same results.

Small dams and reservoirs

Small on-farm earthen dams can reduce downstream peak flow and associated flooding in agricultural watersheds, and can significantly reduce sediment, nitrogen (N) and phosphorus (P) loadings to streams. Researchers at the South Tobacco Creek Watershed in Manitoba observed these sediment and nutrient reductions immediately downstream of the dams.

The following three types of earthen dams were constructed with funding provided through an agreement administered by federal and provincial governments:

Twenty-six of these small on-farm dams are located in the South Tobacco Creek Watershed, such that nearly 30% of the watershed’s total drainage area is now managed for flow reduction. Of the 26 dams constructed within the South Tobacco Creek Watershed, two - the Madill and Steppler dams - were chosen for evaluation as part of the Watershed Evaluation of Beneficial Management Practices study.


Collectively, it is estimated that the entire network of 26 small dams reduced peak flow due to snowmelt by 9 to 19%, and rainfall runoff by 13 to 25%. It shows that, despite differences in construction, both reservoirs significantly reduced not only the export of sediment, as expected, but also reduced the export of total N and total P. This reduction occurred during both snowmelt, and rainfall, generated runoff events.

During rainfall, the reservoirs were occasionally sources of particulate P. However, since dissolved nutrients are the year-round dominant form of N and P in this watershed, the two reservoirs were successful in reducing total N and P loads overall.

Economic considerations

The small dams in the South Tobacco Creek Watershed have the clear potential to regulate peak flow, and thereby directly benefit those landowners who live within the watershed. Additionally, some of the dams may provide back-flood irrigation or have the storage capacity to serve as livestock watering sources.

Although these small dams are largely located on private property, they also provide a public benefit by mitigating downstream flooding and sediment loading. In addition, it is possible that the nutrient reductions that have been measured at the outlet of the dams could be reflected further downstream at the watershed outlet or beyond. Should this prove to be the case, the construction and operating costs of the dams may be further offset by these public benefits.

In combination with improved flood and erosion control, it is clear that small headwater storage dams are an effective conservation tool that may also reduce downstream nutrient loading into rivers and water bodies. This BMP should merit consideration by watershed managers and policy makers when developing resource protection plans, particularly for agricultural escarpment regions on the Great Plains.

Flooded reservoir with partially submerged bushes

Steppler reservoir with high flood-retention levels (~ 40,000 m3). The water level meter is virtually submerged (lower right of photo).

Application of multiple beneficial management practices

Research at the South Tobacco Creek Watershed in south-central Manitoba discovered that the application of multiple BMPs in agricultural areas can substantially reduce nutrient losses to surface water.

The implementation of five BMPs reduced nutrient export to the stream and resulted in little variation in the flow and volume of surface runoff over the entire sub-watershed (as monitored at the sub-watershed outlet). The five BMPs collectively reduced the average annual total P, dissolved P, and particulate P export by 38%, 41% and 42%, respectively. The average annual total N, dissolved N, and particulate N export was reduced by 41%, 43% and 38%, respectively.

The nutrient reduction resulting from each individual BMP is difficult to estimate due to varying landscapes, soils, crops and other agricultural practices. Of the five BMPs implemented, the holding pond and nutrient management BMPs appeared to provide the largest proportion of nutrient reduction.

The holding pond captured all of the nutrient-enriched runoff from the cattle feedlot. Before the holding pond was built, the feedlot drained directly into the stream. Not all nutrients from the feedlot would have travelled the distance to the sub-watershed outlet because of opportunities for nutrient capture along the flow path. These may include sedimentation (deposition or accumulation of eroded sediments), infiltration and adsorption to soil or leaching losses, gaseous losses and crop uptake. The extent of nutrient capture along the flow path is not known. However, if it is assumed that these losses are negligible, the maximum possible nutrient export reduction from the watershed due to the holding pond would be 64% of total P and 57% of total N.

The nutrient management BMP may have played a role in the reduction of nutrient export from the sub-watershed. Nutrient budget analysis revealed N and P inputs were reduced in the treatment sub-watershed following implementation of nutrient management strategies. These reductions resulted from lower fertilizer application rates on annual cropland and from minimal fertilizer applications on the land converted to perennial forage.

These watershed studies clearly demonstrated that a combination of multiple BMPs can be effective at reducing nutrient losses from agricultural lands into water bodies. However, the relative contribution or non-contribution of individual BMPs has yet to be quantified. Research findings may also lead to the enhancement of current BMPs, as well as the development of new BMPs to further minimize nutrient losses to the environment and maximize efficiency of on-farm nutrient use.

Report a problem on this page
Please select all that apply:

Date modified: