2. Methods and Data Sources
2.1 Spatial Data
Langley and Aldergrove land use zoning shapefiles were obtained from the Township of Langley website. Abbotsford land use zoning shapefiles were obtained from Abacus Canmap route logistics 2011. Bertrand and Pepin Creek shapefiles were obtained from and created by Pearson Ecological, Agassiz, BC. All data was projected as NAD83 UTMzone10.
2.2 Water Quality Data
Dissolved oxygen (DO; mg/L), specific conductivity (uS), and temperature (oC), measurements were made twice during the spring (May 15 and June 12) and twice during the summer (July 14 and August 25) of 2012. YSI PRO DO meter with optic probe was used to measure DO and temperature and an ECTester 11+ for conductivity. These parameters where chosen because they are often used as indicators of water quality and fish habitat quality levels. Data was collected from 55 sites in Bertrand Creek and 66 sites Pepin Creek and each location marked with a GPS (Garmin 60CSx; precision +/-3m). An average was calculated for both summer and spring water quality measurements and used for analysis.
2.3 Land Use Classification
As land use classifications in the zoning shapefiles were very generalized, individual parcels were reclassified, using ArcGIS 10.1, into one of five categories based on information extracted from the metadata and personal observation. The new categories were: agriculture, industry, forest, urban and rural.
2.4 Creating Land Use and Water Quality Layers
Using ArcGIS 10.1 the Langley and Abbotsford land use shapefiles, were joined using the union analysis tool. The water quality data was imported as a ‘csv’ file and sampling points were displayed on the LandUse layer. The LandUse polygons and the water quality data points were intersected, and this layer was used to create a 100m buffer around each point.
2.5 Grouping Analysis
The grouping analysis tool in ArcGIS 10.1 classifies data into clusters based on specified attributes. It looks for natural clusters where features within each group are similar and between groups are different. Spatial or temporal properties can be specified. Four grouping analyses were conducted using the buffered summer water quality layer as the input feature and grouped into three categories (low, medium and high). Each water quality parameter, (DO, temperature, and conductivity), was used as an analysis field individually and then the fourth grouping analysis combined the three water quality measurements and was grouped into four categories. No spatial constraints were set, so a K Means algorithm was used for grouping. This minimizes the differences among the features in a group, over all three groups.
2.6 Graphs and Tables
Spring and summer, dissolved oxygen measurements were plotted as a function of temperature using the ‘create scatterplot matrix graph’ function in ArcGIS 10.1. Each land use was assigned a colour to assist visualization of relationships or trends between temperature, dissolved oxygen and land use. An additional scatterplot graph was created to display the grouping analysis of all water quality measurements analyzing the relationship between temperature and dissolved oxygen. The attribute tables of each grouping analysis layer were joined with the summerWQ buffer layer in order to look at land use and water quality groups. These tables were exported to excel, where pivot tables and charts were created.
Langley and Aldergrove land use zoning shapefiles were obtained from the Township of Langley website. Abbotsford land use zoning shapefiles were obtained from Abacus Canmap route logistics 2011. Bertrand and Pepin Creek shapefiles were obtained from and created by Pearson Ecological, Agassiz, BC. All data was projected as NAD83 UTMzone10.
2.2 Water Quality Data
Dissolved oxygen (DO; mg/L), specific conductivity (uS), and temperature (oC), measurements were made twice during the spring (May 15 and June 12) and twice during the summer (July 14 and August 25) of 2012. YSI PRO DO meter with optic probe was used to measure DO and temperature and an ECTester 11+ for conductivity. These parameters where chosen because they are often used as indicators of water quality and fish habitat quality levels. Data was collected from 55 sites in Bertrand Creek and 66 sites Pepin Creek and each location marked with a GPS (Garmin 60CSx; precision +/-3m). An average was calculated for both summer and spring water quality measurements and used for analysis.
2.3 Land Use Classification
As land use classifications in the zoning shapefiles were very generalized, individual parcels were reclassified, using ArcGIS 10.1, into one of five categories based on information extracted from the metadata and personal observation. The new categories were: agriculture, industry, forest, urban and rural.
2.4 Creating Land Use and Water Quality Layers
Using ArcGIS 10.1 the Langley and Abbotsford land use shapefiles, were joined using the union analysis tool. The water quality data was imported as a ‘csv’ file and sampling points were displayed on the LandUse layer. The LandUse polygons and the water quality data points were intersected, and this layer was used to create a 100m buffer around each point.
2.5 Grouping Analysis
The grouping analysis tool in ArcGIS 10.1 classifies data into clusters based on specified attributes. It looks for natural clusters where features within each group are similar and between groups are different. Spatial or temporal properties can be specified. Four grouping analyses were conducted using the buffered summer water quality layer as the input feature and grouped into three categories (low, medium and high). Each water quality parameter, (DO, temperature, and conductivity), was used as an analysis field individually and then the fourth grouping analysis combined the three water quality measurements and was grouped into four categories. No spatial constraints were set, so a K Means algorithm was used for grouping. This minimizes the differences among the features in a group, over all three groups.
2.6 Graphs and Tables
Spring and summer, dissolved oxygen measurements were plotted as a function of temperature using the ‘create scatterplot matrix graph’ function in ArcGIS 10.1. Each land use was assigned a colour to assist visualization of relationships or trends between temperature, dissolved oxygen and land use. An additional scatterplot graph was created to display the grouping analysis of all water quality measurements analyzing the relationship between temperature and dissolved oxygen. The attribute tables of each grouping analysis layer were joined with the summerWQ buffer layer in order to look at land use and water quality groups. These tables were exported to excel, where pivot tables and charts were created.