The view from a small corner of Anglesey...
Contact Harvest Your Rain Home



The relative effects of a Pinus plantation on the hydrology of an Atlantic dune system. (Newborough Warren case study)

Martin Hollingham (2008)

Discussion

Figure 5 is a graphical representation of the annual water balance for the years on record and summarises tables 3. The series have been ordered by increasing effective precipitation.

 

Figure 5: Graphical representation of the Forest, Border and Warren water balance.

 

 

There is a strong relationship between effective precipitation and drainage for all areas, but not for effective precipitation and storage. This was also found by Freeman, 2008. As effective precipitation increases so does drainage. The Forest, Border and Warren drainage all show similar responses to effective precipitation, which implies that there is not a big difference in aquifer properties or received drainage. This is not surprising as ultimately the estimate for discharge is based on both the effective precipitation and the change in storage which is a small fraction of effective precipitation. As water levels and storage predominantly fell in the years studied, the proportion of drainage is always greater than effective precipitation.

Regression plots of effective precipitation against drainage fitted through the zero intercept are shown in figure 6. Table 4 presents the regression values. It is clear that the proportion of drainage to effective precipitation is similar for all groups, drainage is 8-10% greater than effective precipitation and is greatest in the Warren (110.0% of EP) which is 0.3% > than the Border (109.7%) and 1.4% > the Forest (108.3%); the differences though are small and insignificant. The r 2 values for the Forest are better than the Border and Warren because water levels fluctuate less. This suggests that the Forest is having no net effect on the annual hydrological balance.

Figure 6: Regression plot of Effective Precipitation against Drainage

 

Table 4: Regression parameters for Warren discharge against Forest, Border and Warren storage.

Forest
Border
Warren
slope
1.083
1.097
1.110
r 2
0.87
0.72
0.63

There is a strong correlation between Warren , Forest and Border storage which is plotted and regressed in figure 7, and the regression parameters presented in table 5. These both show that for a given change in Warren storage, the change in Border storage will be 90%, and Forest storage will be 55% of that of the Warren . This also applies to water level movements. Interception could explain the limited response of the forest, but it does not explain why Forest storage does not fall more than 40mm. This suggests that the storage capacity of the Forest is limited, either by the depth of aquifer or by water levels being kept lower in comparison to the Warren .

Figure 7: Regression plot of Warren discharge against Forest, Border and Warren storage

 

Table 5: Regression parameters for Warren against Forest and Border Storage, fitted through zero intercept.

Forest
Border
Warren
slope
0.59
0.86
1
r 2
0.69
0.83
1

A regression plot of Warren drainage against Forest, Border and Warren storage is shown in Figure 8: Table 6 presents the regression values. In general the Forest change in storage is 10mm lower than the Warren , while the Border storage is up to 15mm lower, as Warren drainage decreases (Q w ). The Forest and Warren storage responses to Warren drainage are similar (36 % of Q w and 38% of Q w respectively), while the response of Border storage is greater (43 % of Q w ). The Warren drainage (180mm) at which there is no change in storage, is greater for the Forest and Border (-200mm). This implies that the Warren storage falls before falling in the Forest and Border.

Figure 8: Regression plot of Warren discharge against Forest, Border and Warren storage

 

Table 6: Regression parameters for Warren discharge against Forest, Border and Warren storage.

 

Forest
Border
Warren
intercept
71.7
84.9
67.7
slope
0.36
0.43
0.38
r 2
0.81
0.54
0.39

 

 

 

Results                            Contents                                    Conclusions

Copyright © Martin Hollingham