Historical temporal trends in groundwater levels from British Columbia, Canada

Abstract

(S. Rayne).Management of groundwater quantity and quality is an important resource development issue in Canada [1]. Overuse and/or contamination of groundwater supplies has the potential to restrict future use of this resource [2]. In the western Canadian coastal province of British Columbia, increasing attention is being paid to the scientific and legislative/regulatory aspects of surface and groundwaters. Aquifer types in the province have been classified [3] and various investigations have employed case-study type approaches to determine the status of groundwater resources, developed rigorous quantitative models of representative aquifers with predictive capacity, and attempted to assess future changes due to natural and/or anthropogenic forcings (see, e.g., ref. [4][5][6][7][8][9][10][11][12][13][14][15][16] and references therein). In our previous work, we examined temporal trends in groundwater monitoring wells for the western Canada prairie province of Saskatchewan [17]. We found that groundwater resources in this region generally appear to be increasing over time and do not appear to be under current threat from depletion. As part of the current study, we have conducted a similar investigation on time trends for groundwater monitoring well levels in British Columbia.Groundwater data was obtained from the British Columbia Ministry of Environment online Groundwater Observation Well Network database [18]. Trend data is available for a total of 210 groundwater monitoring well stations throughout the province (Figure 1 and Table 1) with end-of-month water levels (presented as meters below ground surface) over the period of record. Substantial end-ofmonth water level data gaps are present throughout the available historical record for the large majority of monitoring wells. At many sites, significant interannual variability in water levels is expected to have occurred due to natural and/or anthropogenic causes, precluding reliable reconstruction of complete inter-and intraannual monthly well level records using interpolative techniques. Consequently, the dataset was screened and correspondingly reduced to only include years with a complete set of monthly records. Average annual groundwater levels were calculated as the mean of the twelve individual endof-month groundwater levels for each year over the resulting usable historical record at each site. Temporal trends in average annual groundwater levels at each site were examined using standard parametric linear regression [19].Of the 210 groundwater monitoring well stations, 166 stations (79.0%) contained records of greater than two years (Table 2). The remaining stations had either 0, 1, or 2 years of complete annual records, and were not considered further. The average usable record length is 9.5 years with a standard deviation of 8.2 years and a range from 0 to 39 years. The majority (62.4%) of the usable record lengths are between 0 and 10 years, with 10% of the records at 0 years ( Figure 2). Only 10% of the records are longer than 21 years, and 1% of the ...