Scale issues in connected landscapes: Resolving emergent behaviour for improved hydrologic modelling

点击翻译按钮获取中文摘要

Over the past fifty years, computational models of surface water and groundwater hydrology have become increasingly complex. Physically-based integrated hydrologic models with millions of nodes are now regularly deployed for flood forecasting and prediction, climate change investigation, and assessment of landscape sensitivity to changes in land use, land cover, management practices, or climate. These models are used to support critical management decisions about infrastructure, environmental practices, forestry, floodplain delineation, and land management in Canada and abroad. While prolific, these models still suffer from critical drawbacks in their ability to represent some common Canadian landscapes and phenomena, such as the wetlands in the Boreal plains of Alberta, thawing of permafrost hydrology in the Northwest Territories, and groundwater-surface water interactions in agricultural Ontario. The common impediment is our inability to map our process understanding generated through small-scale field studies to understanding of the larger phenomena occurring at the watershed scale, particularly when lateral water movement is a controlling factor. We are hampered by both our inability to measure critical quantities in the field for use in hydrologic models and by our limited understanding about how to algorithmically describe physical phenomena at scales of interest, especially with the presence of limited subsurface data. Many existing models can be fit to observations of streamflow, but often for the wrong reasons.