Figure 1: Map showing the location of Watershed 5 at the Hubbard Brook Experimental Forest (US Forest Service, 2013)
The Hubbard Brook experiment (Fig. 1) was established on the belief that it was the model location for small watershed experiments as it is a forest ecosystem where readings could be straightly measured (Hong et. al., 2010). One of the key researches have been regarding the effects of whole-tree harvesting on the Nitrogen cycle in temperate forest conditions at Watershed 5 (Burns and Murdoch, 2005).
Nitrogen is an important nutrient for all living organisms within forested ecosystems. Elements and ions such as ammonium (NH4), ammonia (NH3), and organic nitrogen (N) are forms of nitrogen that usually occur in forest organisms. Though, nitrate (NO3) is frequently desired by plants in temperate forests as a source for nutrients. The nitrogen supply in forests is regularly sufficient but can sometimes be finite3 in its abundance. Martin et. al. (2010) states that this is speculated to be as a result of whole-tree harvesting which disrupts the nitrogen cycle.
Figure 2: Image showing the nitrification process before and after forest cut (
Nitrogen also goes through a number of internal fluxes and alterations within forest ecosystems that impacts its presence as a growth-limiting nutrient (Groffman, 2016). One of the key processes is that of nitrification. Nitrification is the oxidation process of the reduced form of nitrogen, ultimately to nitrate (NO3). Prosser (2005) states that one of the key indicators of increased nitrification as a result of deforestation (Fig. 2) is also the uptake in the concentration of nitrate (NO3).
In this study, the objectives were to measure the changes in nitrate (NO3) concentrations of stream water in Watershed 5 during the period before whole-tree cut (1980-1984) and the period after (1985-1989). The results of this study will be used to determine which of the following hypotheses is supported.
H0 : Nitrification is not significantly affected by whole-tree harvesting at Watershed 5.
H1 : Nitrification is significantly affected by whole-tree harvesting at Watershed 5.
ANALYSIS OF DATA
The data analysis for the quantity aspect was first done by viewing the line graph of average monthly streamflow for the total period before and after whole-tree cut (1980-1989). The specific period was chosen to see whether the streamflow readings differed throughout the years as a result of the deforestation or not. A monthly time series plot using MINITAB was then made to give a much more comprehensive picture of the average monthly streamflow within that period.
The data analysis for the quality aspect was carried out by impacts of whole-tree harvesting on the concentrations of nitrate (NO3) in stream water 5 were first analysed by viewing the histogram of NO3 concentrations during the period before whole-tree harvesting (1980-1984) and the year after (1985-1989). The 5-year period data were also selected to average out any anomalies associated within certain years, such as soil frost events and ice storms.
The data was found to be non-normally distributed based on the Anderson-Darling normality test. As the data was non-normally distributed, the non-parametric test of Mann-Whitney ‘U’ was used. Differences between the measured parameters were considered to be statistically significant if they surpassed the 95% confidence intervals. MINITAB was used to conduct the normality and hypothesis tests for the sample data collected.
PRESENTATION OF THE QUANTITY DATA
Figure 3: Line graph showing the average streamflow (l/s) at Watershed 5 (1980-1989)
The graph (Fig. 3) shows the average streamflow at Watershed 5 from the period before (1980-1984) and after (1985-1989) whole-tree cut. We could observe that the graph peaks during the spring months (March, April, May) regardless of the whole-tree cut.
Figure 4: Time series plot of monthly streamflow at Watershed 5 (1980-1989) by MINITAB
The plot (Fig. 4) shows the monthly streamflow at Watershed 5 between 1980 to 1989. The time series also shows that streamflow is generally highest during the spring months.
PRESENTATION OF THE QUALITY (CHEMISTRY) DATA
Figure 5: Variance plot for all anions and cations of Streamwater 5 from (1964-2014)
The variance plot (Fig. 5) shows how far the spread of nitrate (NO3) concentrations are during the years 1984 and 1985. Therefore, the 5-year period before (1980-1984) and after (1985-1989) the whole-tree harvesting was chosen to analyse the NO3 concentrations.
Figure 6: Histogram showing the NO3 concentrations before whole tree cut (1980-1984) with standard error bars.
The histogram (Fig. 6) of nitrate (NO3) concentrations before whole-tree cut (1980-1984) shows a non-normally distributed data with a total of (n1= 54), mean of ( and standard error of (SE= 0.09).
Figure 7: Plot of NO3 concentrations (1980-1984) produced by MINITAB normality test.
The Anderson-Darling normality test (Fig. 7) conducted on the average nitrate (NO3) concentrations during the period before whole-tree cut (1980-1984) shows a negatively-skewed diagram. This proves that the data during this period are non-normally distributed.
Figure 8: Histogram showing the NO3 concentrations after whole-tree cut (1985-1989) with standard error bars.
The histogram (Fig. 8) of nitrate (NO3) concentrations after whole-tree cut (1985-1989) shows a non-normally distributed data with a total of (n2=50), mean of (, and standard error of (SE=0.01).
Figure 9: Plot of NO3 concentrations (1985-1989) produced by MINITAB normality test.
The Anderson-Darling normality test (Fig. 9) done on the average nitrate (NO3) concentrations during the period after whole-tree harvesting (1985-1989) also shows a negatively-skewed diagram. This verifies that the nitrate (NO3) during this period are non-normally distributed.
INTERPRETATION OF DATA
The two normality tests done on the data confirms that both periods have non-normally distributed data. Some of the reasons why the data is non-normally distributed could be due to the fact that the sample size (n1=54,n2=58) is quite large and that outliers and mixed distribution are present within the sample data.
As such, standard parametric tests could not be conducted to test the hypotheses. Hence, the Mann-Whitney ‘U’ test was chosen as the interval data collected was significantly skewed.
Figure 10: Output of Mann-Whitney 'U' test conducted by MINITAB
The Mann-Whitney ‘U’ test (Fig. 10) was carried out using MINITAB. This was done with a 95% confidence range, so there is only a 5% probability that the results were due to chance.
Based on the Mann-Whitney ‘U’ test conducted, the p-value of 0.000 was the output. Since, this value is lesser than the significance level of 0.05. Hence, the null hypothesis (H0) is rejected and the alternative hypothesis (H1) must be accepted. Therefore, based on the sample data and statistical analysis conducted, we can conclude that we are 95% confident that nitrification is significantly affected by whole-tree harvesting at Watershed 5.
LIMITATIONS AND FUTURE RESEARCH
There are several limitations of this current study. Firstly, results of this study are limited to the Northern Hemisphere forest conditions. The results cannot be generalized globally as we must take into account the difference of geographical regions. However, the findings of this study offer a starting point for a comparative analysis of the effect of whole-tree harvesting on the nitrification process at other types of forest.
An unanswered question concerning the effects of whole-tree harvesting is whether the subsequent changes in nitrate (NO3) concentrations of stream waters are produced mainly by the main effect of whole-tree cutting, or by the minor effects of soil disruption and resulting alterations in rates of microbial Nitrogen-cycling processes (Bernhardt et. al., 2002).
Conclusion
The whole-tree harvesting at Watershed 5 is believed to only have a minimal impact on the quantity of waterflow. This is due to the fact that the average waterflow during the spring months remained the highest regardless of the period before (1980-1984) or after (1985-1989) the whole-tree cutting. This could possibly be due to the fact that trees larger than about 50-cm in basal diameter were cut with chain saws which would allow the roots to remain and retain the integrity of the soil.
The quality of watershed 5 is shown by the greatly increased concentration of nitrate (NO3) from the deforested ecosystem. This was due to an alteration of the nitrogen cycle within the ecosystem. Although nitrogen is generally preserved in the undisturbed ecosystem, in the deforested ecosystem nitrate is quickly expelled from the system. The movement of nitrate out of the system is believed to be a result of increased nitrification as the hypothesis of this study is supported.