178
C. Saliternik
Statistical tests
In order to examine the influence of various parameters on nitrate concentration in
groundwater, linear simple correlation and multiple correlation tests were made. In each
case, the dependent variable (Y) was defined as the average (or maximum) nitrate
concentration over a certain period. The parameters (independent variables) whose
influence on nitrate concentration in the groundwater was examined were the intensity
of pollution, soil permeability, depth of saturated layer, geological profile, quantity of
annually pumped water, degree of land use in the vicinity of the well, and the intensity of
pollution relative to the directions of groundwater flow and of well location.
The results of the linear correlation were negative. No clear significant connection was
found between any of the parameters and the rate of groundwater pollution by nitrate.
In the second stage, multiple regression tests were held with the aim of examining the
joint effect of different parameters on the rate of groundwater pollution by nitrates. The
computer programme is arranged so that it first holds the simple test with every possible
combination of pairs of variables, then selects the independent variable which shows the
best adaptability to the dependent variable (Y), and computes the correlation coefficient.
After that, the programme adds the variable which has shown the second best
adaptability, etc.
Successful statistical processing by this method requires a large amount of data.
Unfortunately, as a result of insufficient results of chemical analyses, the sample was not
sufficiently large to be representative. The general result was that the parameters which
most affect groundwater pollution by nitrates are the land use in the vicinity of the well
and the geological profile.
DISCUSSION
The presence of large quantities of organic nitrogen deep below the soil surface raises
the question of its sources and of the possibilities of its decomposition in the course of
time as a result of environmental changes. The author hopes that the application of
advanced test methods which will make it possible to differentiate between N
1 4 and N
1 5 ,
and of radioactive methods for the determination of the age of the water in the aquifer
and in the saturated soil profile will make it possible to establish clearly the role of
organic nitrogen decomposition, in groundwater pollution and other potential pollutant.
The scope of the research will be enlarged to include also chlorides and pesticides. The
findings of the research conducted so far have shown that one of the key questions with
regard to the rate of groundwater pollution by nitrate is the fate of the organic nitrogen
in deep soil profiles.
On the basis of this information more should be known about the relative contribution
of the various potential pollutants, and it should be possible to make more exact forecasts
on the rate of groundwater pollution.
C. Saliternik
Statistical tests
In order to examine the influence of various parameters on nitrate concentration in
groundwater, linear simple correlation and multiple correlation tests were made. In each
case, the dependent variable (Y) was defined as the average (or maximum) nitrate
concentration over a certain period. The parameters (independent variables) whose
influence on nitrate concentration in the groundwater was examined were the intensity
of pollution, soil permeability, depth of saturated layer, geological profile, quantity of
annually pumped water, degree of land use in the vicinity of the well, and the intensity of
pollution relative to the directions of groundwater flow and of well location.
The results of the linear correlation were negative. No clear significant connection was
found between any of the parameters and the rate of groundwater pollution by nitrate.
In the second stage, multiple regression tests were held with the aim of examining the
joint effect of different parameters on the rate of groundwater pollution by nitrates. The
computer programme is arranged so that it first holds the simple test with every possible
combination of pairs of variables, then selects the independent variable which shows the
best adaptability to the dependent variable (Y), and computes the correlation coefficient.
After that, the programme adds the variable which has shown the second best
adaptability, etc.
Successful statistical processing by this method requires a large amount of data.
Unfortunately, as a result of insufficient results of chemical analyses, the sample was not
sufficiently large to be representative. The general result was that the parameters which
most affect groundwater pollution by nitrates are the land use in the vicinity of the well
and the geological profile.
DISCUSSION
The presence of large quantities of organic nitrogen deep below the soil surface raises
the question of its sources and of the possibilities of its decomposition in the course of
time as a result of environmental changes. The author hopes that the application of
advanced test methods which will make it possible to differentiate between N
1 4 and N
1 5 ,
and of radioactive methods for the determination of the age of the water in the aquifer
and in the saturated soil profile will make it possible to establish clearly the role of
organic nitrogen decomposition, in groundwater pollution and other potential pollutant.
The scope of the research will be enlarged to include also chlorides and pesticides. The
findings of the research conducted so far have shown that one of the key questions with
regard to the rate of groundwater pollution by nitrate is the fate of the organic nitrogen
in deep soil profiles.
On the basis of this information more should be known about the relative contribution
of the various potential pollutants, and it should be possible to make more exact forecasts
on the rate of groundwater pollution.