Will Our Bridges Be Over Troubled Waters?

Climate change has gained much interest in the past few years both within the scientific community and the general public. Its effects have been related to a number of factors including an increase of greenhouse gases, such as carbon dioxide. Recent studies indicate that climate change is a principal cause of increases in extreme storm events and high intensities of precipitation. It has now piqued the interest of public and private infrastructure managers.

How will climate change impact the transportation sector? It may require adaptive design measures to handle changes in climate. Can we continue to design drainage systems based on historical climate data and the assumption of a fixed climate? What actions can we take now to protect the highway infrastructure in the long run, in preparation for the potential impacts of climate change?

To answer these questions, the Ministry of Transportation sponsored a research project at McMaster University through the Highway Infrastructure Innovation Funding Program (HIIFP). The purpose of the study (directed by Dr. Paulin Coulibaly) was to investigate the potential impact of climate change on highway drainage infrastructure including bridges, culverts, storm sewers and stormwater management facilities. Dr. Coulibaly's research used models that predict the earth's climate to identify trends in precipitation and water flow on different catchment areas. These trends would better predict runoff flow rates that would need to be accommodated by the highway drainage infrastructure.

Current prediction models of the earth's climate are able to reasonably forecast changes in temperature over time. There are a number of different climate prediction models, generally referred to as Global Climate Models (GCMs). These models provide temperature and precipitation predictions using a 350 km by 350 km grid system. However, this scale is considered to be very large, especially when trying to predict temperatures and precipitation at a specific location. Furthermore, GCMs have been unable to accurately predict precipitation, compared to actual values measured at monitoring locations. This lack of accuracy is due to the added complexity of predicting the behaviour of a number of systems including storm patterns, wind speed, and air moisture content.

To overcome the limitations of the GCMs, Dr. Coulibaly used techniques to relate the GCM temperature and precipitation predictions to local precipitation data. By identifying the relationship of the GCM simulations and local weather conditions based on historic data, future predictions of the GCM can be translated into local precipitation predictions. To determine estimates of local and regional values of future daily precipitation and variability, eight rainfall stations were selected to represent typical southern and northern Ontario regions. Four stations were in the Grand River watershed and the other four rainfall stations were located in the Kenora/Rainy River watershed.

To show how much, how long and how often precipitation occurs in the two test regions, rainfall intensity-duration-frequency (IDF) curves were derived using the GCMs and daily precipitation data from the rainfall stations. To detect rainfall trends, IDF curves were developed for four time periods, the present, 2020, 2050 and 2100.

The IDF curves showed changes in precipitation intensity, suggesting that by 2050 and 2100 a 24% and 35% increase in heavy, and more frequent, rainfall events can be expected. All correlations showed these increasing trends, except for the 2020s when there was a decrease. In his study, Dr. Coulibaly indicated that this decrease is consistent with the decrease observed in the raw GCM simulations and can therefore be related to the GCM predictions rather than the statistical techniques used to determine the local conditions.

Dr. Coulibaly suggests that highway drainage infrastructure may be significantly affected by climate change. He predicted that, by 2050, highway drainage systems designed to accommodate storms that occur once in 10 years may only be able to accommodate storms that occur once in 5-years. Larger highway drainage systems designed for a once in 50-year storm period might only be able to accommodate a once in 20-year storm period. Dr. Coulibaly suggests that design flow rates (the estimated runoff flow rates) may need to be increased leading to larger bridges, culverts, storm sewers and storm water management ponds to maintain the level of service provided today and to avoid potential constricted water flow and possible future flooding events.

While this research provides some insight into the potential impact of climate change on highway drainage infrastructure based on a limited number of sites and a short period of rainfall record, more research is required to fully understand future climate effects on rainfall and flow rates before considering modifications to highway drainage design standards.