Modelling the impacts of agriculture in mixed-use landscapes: a review and case study involving two species of dabbling ducks

Context

This study synthesizes evidence from three separate surveys of American Black Duck and Mallard breeding habitat usage to quantify the effects of agriculture at the landscape scale.

Objectives

To assess duck breeding activity in agricultural landscapes within the Canadian maritimes in order to measure the overall impact of agricultural land use, the response to particular agricultural activities, and the influence of landscape configuration.

Methods

Models, constructed using a long-term census (SURVEY1), were used to predict habitat selection for two other independent surveys (SURVEY2, SURVEY3). Predictions incorporated information about wetland area and diversity, as well as anthropogenic factors, allowing subsequent analyses to focus on the remaining residual error attributable to agricultural effects.

Results

SURVEY2 results demonstrated that the proportion of active agriculture is an important indicator of the severity of human disturbance, yielding threshold estimates of 39% for Mallards and 60% for Black Ducks, with an overall average of 50%. Agricultural conversion beyond these thresholds deterred breeding ducks independently of other factors. SURVEY3 land cover information demonstrated that the presence of cropland intensified this deterrence effect, even at levels as low as 10%. Woodland cover (in excess of 30%) was important for both species, but its configuration was not.

Conclusions

In addition to quantifying threshold effects, this study reaffirms that woodland is an important part of the maritime landscape matrix, and contributes positively to habitat diversity in mixed use, moderate intensity agricultural regions. Wetland restoration in agricultural landscapes should monitor and promote less crop-intensive, mixed-use practices.