The top 30% of weighted sum result using slope as the highest weighted factor resulted in 57,448 km^2 of the affected area. The subbasin with the most area affected in our analysis was the Tanana River subbasin with 22.6% of affected area. The Lower Yukon subbasin was also found to have significant flood risk with 14.4 % area affected. Based on our results, more efforts can be diverted to Tanana River subbasin and the Lower Yukon subbasin in minimizing flood damage on human properties and future flood analyses. Towns within the affected area, i.e. the top 30% of weighted sum results, are included in Table 4. Fairbanks is one of the largest cities in Alaska. Since it is located in the Tanana River subbasin, the highe level of precipitation and flat slope characteristics both contribute to the high flood risk and therefore the city falls into the one of the high risk zones. Other major cities such as North Pole and Fort Yukon also fall into the areas at high risk of flooding; total estimated population at risk is around 36,830 people. In addition, there is a large portion of the secondary roads, highways and pipelines that can be potentially affected. The total lengths of affected infrustucture is listed below in Table 5. As a result, floods that happen in particularly populated regions of the Yukon Basin can have quite a large impact on the safety of the residences and the economy of the region by damaging infrastructure.

Figure  9.  Final map    [click on image to enlarge]

Table  4.  Affected towns




Figure 10. Affected infrastructure   [click on image to enlarge]

Table 5.   Affected infrastructure summary table

In our sensitivity analysis, the largest affected area was obtained using the highest weights for ‘Soil’ and ‘Vegetation’ layers while the smallest affected area was obtained using ‘Accumulation’ and ‘Slope’ layers. Smaller differences from our results with regards to total area affected was observed when using equal weights (equalwtd)  and when varying the weights of the top 3 factors (wtd_precip). However, in the ‘equalwtd’ result, the percentage of affected area in each subbasin was slightly different than the final result. The subbasins with the largest affected  area and second largest affected area were switched around in ‘equalwtd’ compared to our final result. The percentage of affected area in ‘wtd_precip’ was similar to our results except that the values were more exaggerated. Although we are able to compare the percentage of affected areas in each subbasin, the spatial spread of affected areas can be quite different as can be seen in the sensitivity maps (Figure 11- Discussions).