4. Discussion
Figure 10: Temperature grouping analysis map
4.2 Grouping Analysis
The grouping analysis tool is an exploratory method that reveals patterns in the data and cannot be used to draw conclusions about the correlation between land use and water quality. However, this analysis confirmed that that urban, agricultural and industrial sites exhibit medium to high temperatures levels and forest sites exhibit low temperature levels (figures 5 and 10). Low flows during drought conditions through July and August likely exacerbate water temperatures especially in Bertrand Creek.
The grouping analysis tool is an exploratory method that reveals patterns in the data and cannot be used to draw conclusions about the correlation between land use and water quality. However, this analysis confirmed that that urban, agricultural and industrial sites exhibit medium to high temperatures levels and forest sites exhibit low temperature levels (figures 5 and 10). Low flows during drought conditions through July and August likely exacerbate water temperatures especially in Bertrand Creek.
Figure 11: Dissolved oxygen grouping analysis map
Typically higher water temperatures are related to low dissolved oxygen levels, however the grouping analysis showed that some agriculture sites had medium or high temperatures and good or saturated dissolved oxygen (figures 10 and 11). This may be attributed to increased photosynthesis in eutrophic unshaded reaches as the measurements were taken during the day. The reverse situation was also highlighted in some industry sites that had low temperatures and low dissolved oxygen, which does not support previous studies (Hicks, Hall, Bisson, & Sedell, 1991; Knothe, 2012). The low dissolved oxygen levels may be due to lack of flow and mixing in the stream and larger ponds, and an increase in oxygen demand due to decomposition of organic matter. The urban sites, on the other hand, were grouped as medium (warm) temperature and low dissolved oxygen, which do support previous literature.
Figure 12: Conductivity grouping analysis map
Typically areas that are heavily urbanized or have intense agricultural impacts are associated with high conductivity levels due to an increase in chloride, phosphates and/or nitrates (Hall & Schreier, 1996; Paul & Meyer, 2008). In Bertrand Creek, the analysis showed only two urban sites with medium conductivity levels and a few agriculture sites with medium or high levels (figures 7 and 12). If those results are compared to Pepin Creek, which only has lower conductivity levels, than the assumption could be verified. However, the conductivity levels at all sites in both creeks are considered higher than typical levels in other watersheds (personal obs.). The higher levels could be due to either an abundance of natural minerals in these watersheds or an excessive amount of chemicals from fertilizers and other sources at the headwater and/or all along the stream. Thus, resulting in above average levels in the entire watershed. The interesting result is the area in Bertrand Creek where the conductivity levels decrease from high to medium to low as the water moves downstream (figure 12). At the upper end of the stream, where there are high levels of conductivity, there is a mushroom barn on agriculture land (personal ob.). Effluent from the mushroom barn could cause the high levels that are observed in this data.
Figure 13: All water quality parameters grouping analysis map
The final grouping analysis combined all the water quality indicators, looked for natural clusters based on similarities and divided them into four groups (figures 8 and 13). The most striking observation is that only the forest land type has average/good water quality at all sites. All other areas show either high conductivity, high temperatures/high dissolved oxygen or low dissolved oxygen. These results suggest that land use is having a significant impact on water quality in the Bertrand and Pepin Creek watersheds. This also suggests that the removal of riparian vegetation contributes to the degradation of water quality.