3. Results
As for the first year of the study six wells were selected in the Waikakihi catchment (Figure 1) with well depths ranging from 4.6 to 11 m below ground level. The wells were selected to avoid possible contamination from septic tanks and to ensure good wellhead protection. Three wells were slightly modified to improve the wellhead protection. Details for each well are given in Table 1. The second year well sampling program started on September 4 2003 (winter sample with no irrigation effects) and continued at varying intervals during the irrigation season until April 2004. Intensive sampling (12 samples) was undertaken at two locations, which involved sampling twice a week for one month. Less intensive sampling (between 3 and 5 samples) was undertaken at the three remaining wells. This gave a total of 44 samples complementing the 42 samples collected in the first sampling round resulting in a total of 86 samples.
Table 1: Details for each well
| Site |
Well diameter (mm) |
Well screen (m) |
Well depth (m) |
Usage |
| J40/0134 | 200 | 4.6 | Stock | |
| J40/0131 | 200 | 6.1 | Domestic/Dairy | |
| J41/0008 | 150 | 8 - 10 | 11* | Dairy/Stock |
| J41/0025 | 150 | 6.5 | Domestic/Stock/Dairy | |
| J41/0026 | 200 | 5.7 | Domestic/Stock | |
| J41/0031 | 150 | 7.1 | Domestic/Stock |
* Well owner indicated that this well has been deepened to about 15m.
The full set of results from the groundwater samples are given in Appendix 1 and are summarised in Table 2. Total coliforms were detected on 99% of sampling occasions at levels ranging from <1 to >2,400 MPN/100ml with an average for all wells of 538 MPN/100ml.
E. coli was detected in all wells, with the levels ranging from <1 to 2400 MPN/100 ml with an average for all wells of 59 MPN/100ml. The mean level of E. coli in the well samples ranged from 9 to 252 MPN/100 ml (Table 2). The drinking water standard for E. coli is <1 MPN/100 ml. The overall detection rate for all samples for E. coli was 83%.
Over the two years Campylobacter have been detected in 5 out of the 6 wells on at least one sampling occasion with levels ranging from 0.6 to >3.1 MPN/L. There have been 9 samples out of a total of 86 with positive detections of Campylobacter, giving an overall detection rate of 10.5%. Six of the detections have been C. jejuni and the remaining three have been thermophillic Campylobacter but probably not C. jejuni or C. coli. In year two Campylobacter was detected in well J41-0008 in a sample taken in September 2003 and two samples taken in February 2004 coinciding with the large increase in E. coli counts.
There is a poor correlation between both Campylobacter and E. coli (r=0.065, p=0.57, n=79) and between Campylobacter and total coliforms (r=0.035, p=0.76, n=79). This is due mainly to the relatively low number of positive campylobacter results (9) in comparison to positive E. coli and total coliform results (almost 100%). This is a change from the first year results where there was a reasonably good correlation between E. coli and Campylobacter (r=0.52, p=0.001, n=36).
E.coli laboratory results are plotted with stocking and irrigation data from each well are presented in Figures 2-6. Intensive (twice weekly) second year sampling of wells J41/0008 and J41/0026 was undertaken in February and March 2004.
Year one data for well J41-0026 (Figure 2) show an elevated E.coli count of 100 MPN/100ml in a sample collected on 7 November. This is after the commencement of flood irrigation and approximately 16 days after paddocks 32 and 33 were stocked and 11 days after a large rainfall event (22mm). Paddock 34 was continually stocked with calves throughout the season. Stock numbers recorded in paddocks 32 and 33 generally ranged from 270-285 cows. A second elevated E.coli count was recorded on 17 January, approximately 20 days after the paddock 32 was stocked in conjunction with an irrigation event.
Year two data for the same well (Figure 2) show an increase in sampling density in January and February 2004. An elevated E.coli count of 38 MPN/100ml was recorded on 16 February. There were significant rainfall events in late January and February 2004, negating the need for irrigation at this time. The E.coli peak occurred approximately 20 days after paddock 33 was stocked in conjunction with a major rainfall event. Reduced E.coli counts of 2 and 5.2 MPN/100ml recorded on 12 and 19 February respectively indicate that the pulse of contaminated groundwater moved quickly past the sampling point.
Samples collected from well J41-0008 (Figure 3) indicated an elevated concentration (32 MPN/100ml) of E.coli on 12 December 2002 (year one). This sample was collected approximately 20-25 days after paddock D7 was stocked in conjunction with two irrigation events.
The 2003/2004 data (Figure 3) show a large E.coli peak (1100 MPN/100ml) on 16 February 2004, the same time as well J41-0026. This peak is broader than the peak recorded from well J41-0026 with elevated counts of E.coli recorded both before (190 MPN/100ml on 12 February) and after 16 February (34 MPN/100ml on 23 February). This peak occurs approximately 25-30 days after paddock D7 was stocked in conjunction with an irrigation event and also after the period of high rainfall in late January to early February.
Elevated E.coli counts of 81 and 16 MPN/100ml were recorded in samples collected from well J41-0025 on 13 March 2003 and 15 January 2004 respectively. Paddocks 21, 22 and 23 were regressed in December 2002 resulting in no stocking or irrigation of these paddocks until 22 February 2003. The elevated concentration of E.coli in March 2003 is approximately 20 days after irrigation recommenced, though stocking of the paddocks did not occur until 9 March 2003. Stock numbers recorded for paddocks 20-23 ranged from 280-320 cows.
Data collected in relation to well J41-0031 show extremely high E.coli counts in mid- January of 2003 and 2004 (120 and 2400 MPN/100ml E.coli respectively). Discussions with the site manager indicated that the small paddock in which the well is located was used to keep ten bull calves from October to mid-December 2003. The well is situated in a small hollow with relatively large (10cm) river cobbles at the surface, adjacent to the well. It is likely that increased infiltration and runoff could occur in this area immediately surrounding the well. At the time of writing it was unknown if stock was kept in this paddock in late 2002 also. During a field visit to the site it was recommended that the area be filled and to seal around the well with concrete if animals were going to be kept in this paddock in the future. Stock numbers supplied for the paddocks near this well generally ranged from 50-105 cows.
Samples collected from well J40-0134 indicated two peaks of E.coli with 53 MPN/100ml recorded on 13 February 2003 and 22 MPN/100ml recorded on 4 December 2003.
Table 2: Summary of groundwater quality data for each well. Mean with range in brackets
| Site | Water level (m bgl) | C. jejuni (MPN/L) | C. Therm (MPN/L) | E. coli (MPN/100ml) | Total coliforms (MPN/100ml) |
| J40/0134 | 1.77 (1.23 - 2.6) |
0.05 (<0.6 - 0.6) |
0 (<0.6 - <0.6) |
9 (<1 - 53) |
780 (25 - >2400) |
| J40/0131 | 2.47 (1.29 - 3.29) |
0.78 (<0.6 - >3.1) |
0.39 (<0.6 - 3.1) |
31 (2 - 120) |
514 (20 - 2400) |
| J41/0008 | 7.43 (7.29 - 7.52) |
0.09 (<0.6 - 0.6) |
0.16 (<0.6 - 3.1) |
76 (<1 - 1100) |
989 (2- >2400) |
| J41/0025 | 2.1 (1.81 - 2.50) |
0 (<0.6 - <0.6) |
0.05 (<0.6 - 0.6) |
9.9 (<1 - 81) |
845 (14 - >2400) |
| J41/0026 | 3.72 (3.31 - 4.04) |
0.03 (<0.6 - 0.6) |
0 (<0.6 - <0.6) |
12 (<1 - 100) |
514 (20 - >2400) |
| J41/0031 | 3.29 (3.0 - 3.64) |
0 (<0.6 - <0.6) |
0 (<0.6 - <0.6) |
252 (<1 - 2400) |
780 (25 - >2400) |
Note:
1. For the purpose of calculation of the mean, values less than detection were set to zero and values
greater than a certain MPN number were set equal to twice the maximum limit. The full set of data is given in Appendix 1.
2. C. Therm are thermophillic Campylobacter that are probably not C. jejuni or C. coli.
Groundwater level fluctuations continued as noted in the first year report with the two wells near the top of the catchment had the most variation (1.3 to 2 m range) with a peak level in January or February. The other three wells near the lower end of the catchment had much less variation (range <0.7 m) and no consistent pattern.
Figure 2: Years one and two E.coli counts and stocking and irrigation dates for paddocks 32, 33 and 34 for well J41-0026. Paddock 34 was continually stocked with calves. Stock numbers in paddocks 32 & 33 ranged from 270-285.
Figure 3: Years one and two E.coli counts and stocking and irrigation dates for paddocks D7, D6 and D13 for well J41-0008. Stock numbers were not provided. Note the log scale for E.coli in year two.
Figure 4: Years one and two E.coli counts and stocking and irrigation dates for paddocks 20, 21, 22 and 23 for well J41-0025. Paddocks 21, 22 & 23 were regrassed in December 2002. Stock numbers ranged from 280-320.
Figure 5: Years one and two E.coli counts and stocking and irrigation dates for paddocks 1, 2, 4 and 6 for well J41-0031. Ten bull calves were kept in well paddock from October to mid-December. Stock numbers in the remainder of the paddocks ranged from 50-105. Note year two log scale for E. coli.
Figure 6: Years one and two E.coli counts and stocking and irrigation dates for paddocks 82/83 and 88 (600 stock) for well J40-0134.
Contact for Enquiries
MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND
Fax: +64 4 894 0721
Contact this person
