Discussion of Results

Based on the predicted increases in MWAT ranges, species composition predictions can be made for the study area. By comparing the upper limit of the current MWAT range to the predicted MWAT range for each location that the species has been found in, it can be reasonably predicted whether the species will be found in that same location in the future. The table below shows the number of locations that the species has been found in (given the scope of this study) and compares this to the number of predicted locations it will be found in out of the locations that are currently occupied.

As Table 3 shows, for all species there is a decline in the number of locations that will be suitable habitat in the 2080s. The locations that Rainbow trout (RB) and Kokanee (KO) have been found in have higher upper limits for the current MWAT ranges, and the predicted MWAT ranges for the locations that these species are found in are wider, so these species will likely be more prevalent than Bull trout (BT) and Slimy sculpin (CCG). The column showing the predicted fraction of future locations a species will inhabit, shows a severe decrease of BT in the area; so extreme a decrease that the species may not even be found in the area. This conclusion agrees with the trend found in the 2016 Parksinon et al. paper, which analyzed fish species composition based on the MWAT model for the Thompson River and Upper Fraser River drainages. In this paper, it was found that RB were much more prevalent than BT at higher MWAT values.

As  our results and Table 4 summarize, suitable habitat locations will decrease by at least 30% for all study species. It is predicted there will be a severe decrease in Bull trout populations as characterized by a decrease in suitable locations by about 76%. As Figure 10 showed, much of the study area will become unsuitable for Bull trout by the 2080s. Within the study area there does not appear to be any distinct region where conditions will remain suitable for Bull trout, other than possibly the Northwest section (further research would be required to confirm this). Although suitable conditions for Rainbow trout and Kokanee decrease by 30%, there will remain a fair amount of potential habitat for both of these species. As many of the projected suitable locations for these two species overlap, it is reasonable to assume that there will probably be increased competition between these species for the suitable habitat and resources. Slimy sculpin’s future in the study area appears to be bleak as an estimated 56% decrease in suitable habitat for this species is predicted.

Discussion of Errors

It should be noted that there are a variety of errors associated with our results,  brought about through the selection and omission of data to accommodate the scope of this project. Firstly, we acknowledge that only a small subset of data from recordings of fish species was taken into account.  However, the process outlined in this project can be implemented for a much greater number of recordings, and is encouraged for further research. With increased recordings, predictions would become inherently more valuable. Secondly, in an ideal scenario a glacial change model would have been used in conjunction with the MWAT model to more accurately predict the how fractional glacial cover would change over time, and subsequently the MWAT inputs. This procedure was not pursued for this study, as further data retrieval would have been required which could not be accommodated. Thirdly, this study only took into account the broad effect of climatic warming (such as mean summer temperature increase, and glacial melt)  on fish species in the channel systems of our study areas. Research into the ecological, migratory, and behavioural patterns of the fish species was not studied. Therefore it is plausible that fish populations may shift habitats due to changing conditions, and population characteristics may fluctuate in the future due to reasons not directly predicted by the MWAT model. Nevertheless, it can also be argued that the effects of climate warming mentioned would be the trigger to any consecutive conflicts that may arise within a population, and thus subsequent population dynamics are not relevant for this project’s aim.  Fourthly, the retreat of glaciers in the Columbia River Basin may give rise to additional channels and lakes which may affect the results of this model, but this element is difficult to predict and thus was omitted as we could not find a way to accurately represent this. Instead, we offered current and predicted MWAT values based on the variables we had available. Finally, Q3 is the 75th percentile of the fish point distribution data and it was set as the upper limit of the MWAT model. This value was chosen as seeming reasonable based on the 2016 Parkinson et al. paper, but this may not be the best choice to represent the uppermost boundary. While this assumption is valid for the BT and CCG species, the upper limits for the RB and KO species are not as clearly defined.

Future Research

For future studies within this field we wish to offer some recommendations. Although time-consuming, the processing of a greater volume of fish data would yield much more conclusive results. Likewise, the upper limits of data for Rainbow trout and Kokanee were not as well defined as for Bull trout and Slimy sculpin. More research into better MWAT model limitations is advised. At the very least, this point should be taken into account for analysis and interpretation of results. Furthermore, we suggest the use of a glacial change model to predict the fractional glacial cover for the 2080s. In the interest of meeting this project’s deadline, fractional glacial cover was set as zero, yet this probably does not accurately represent actual model predictions. Using a glacial change model such as the one recommended in the 2015 Clarke et al. paper (see references) will produce more accurate predictions. Finally, depending on the intentions of future research, it could prove useful to take into account some additional specifics on the fish species.  Research on ecology or population dynamics would lead to better informed predictions on habitat occupation and interspecies relations.