Return to search

Impacts of Deforestation on Water Quality and Quantity in a Canadian Agricultural Watershed

Around the world, many forested areas have been and continue to be cleared for expanding agriculture. Canada’s remaining forested lands account for around 9% of the world’s forest cover. Although only a fraction is lost to deforestation annually (0.02%, 2013), Statistics Canada reports that conversion to agriculture is the most significant driver of forest loss. As climate changes and agricultural demand expands, this trend is expected to continue, and ecosystems will continue to be impacted by resulting habitat loss and hydrological changes that can impact infrastructure and communities. Additionally, changes to sediment and nutrient loadings can harm ecosystems and affect the downstream usability of freshwater supplies.

The impact of increased sediment and nutrient concentrations in freshwater systems has been extensively documented in the literature. In some extreme cases, it can lead to anoxic ‘dead zones’ in riverine, lacustrine, and marine habitats. Many river systems in Canada have shown elevated nutrient levels in recent years, often tied to the expansion of agricultural land use and destruction of natural forests to increasing nutrient levels in downstream rivers, lakes, and oceans.

This study applies numerical modelling to quantify the influence of forest loss, agricultural expansion and the application of best management practices (BMPs) on water quality and quantity in the South Nation Watershed in eastern Ontario, Canada. The land use in the watershed is mainly agricultural (over 60%) with forest (27%) that is unevenly distributed in the basin. Aerial photography surveys from 2008 and 2014 show a steady decline in forest cover. Recent water quality monitoring has shown nutrient concentrations at or above Canadian water quality standards in many parts of the basin. The Soil and Water Assessment Tool (SWAT) was used to model the watershed because of its capacity to simulate comprehensive land management scenarios and assess their impact on a variety of water quantity and parameters quickly and effectively.

The work was performed in four steps:
1. Recent land use configurations (2008-2014) in the watershed were acquired, and simplified land use projections based on the direct substitution of cropland for forest land were developed.
2. A numerical model was calibrated and validated for the initial land use scenario.
3. These land use scenarios, as well as more hypothetical scenarios representing more extensive deforestation and reforestation, were used as the basis for hydrological modelling using 31 years of real-world meteorological observations.
4. Idealized vegetated filter strips (VFSs) and grassed waterways (GWWs) were added to the cropped land packages to study the potential of these practices to contribute to the management of water quality.

Analysis of the 33 output datasets derived from simulations of the suite of land use scenarios with and without VFSs and GWWs leads to several conclusions, while also raising some questions. Generally, forests significantly reduce sediment, nitrate and phosphorus outputs to streams as well as slightly reducing water yield compared to cropped areas due to an increase in surface runoff, groundwater and lateral flow combined with the absence of tile drainage. Across subbasins, this translates to significant reductions in sediment, nitrate and total phosphorus loadings entering the river reaches and a slight increase in water yield. At the basin outlet near Plantagenet, Ontario, streamflow and sediment loading show to have little sensitivity to changes in forest and crop cover, while increased forest cover leads to significantly reduced nutrient loadings, particularly in late spring and early winter.

It is clear from this work that continued deforestation will continue to drive further nutrient enrichment in the South Nation River, while VFSs seems to have a significant potential for offsetting some of this enrichment. Streamflow and sediment loadings, however, are not significantly impacted by foreseeable deforestation. The influence of land use change and BMPs was much more significant in the runoff than in exports from the basin, suggesting there would be value in further examination of water quality and quantity at a higher spatial density to expand on assumptions of in-stream processes made here.

Identiferoai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/40967
Date10 September 2020
CreatorsNoteboom, Matthew
ContributorsSeidou, Ousmane, Lapen, David
PublisherUniversité d'Ottawa / University of Ottawa
Source SetsUniversité d’Ottawa
LanguageEnglish
Detected LanguageEnglish
TypeThesis
Formatapplication/pdf

Page generated in 0.0019 seconds