Soil Ph of Algonquin Park

Soil pH of Algonquin Provincial Park

C. Cui

FIT 1 Imaging Technician and soil pH Analyst

J. Fung

FIT 1 Mammal Specialist and soil pH Analyst

B. Li

FIT 1 Probe Technician and soil pH Analyst

R. Lin

FIT 1 Recorder and soil pH Analyst

(Received 01 June 2015)

Soil pH data from 67 sample stations in Algonquin Provincial Park over the span of 8 years was collected and analyzed to determine trends and establish correlations with water ions, and tree data. After processing for reliability, the data indicated that Algonquin Park exhibits an overall increase in acidity over time, with the central region being more acidic than soil samples collected in the east and west regions. A direct, positive correlation was also established between soil pH and ammonia ion concentrations in neighbouring water systems, likely a result of chemical equilibria and surface runoff. Soil pH was shown to have some degree of influence over the tree species present in an area and vice versa, however further studies pertaining to the health of the specimens are required to produce more conclusive correlations.

 I Introduction

Soil pH is a measurement (the negative logarithm) of the molar hydronium ion concentration in soil. It is measured in soil pH units, which range from 0 (acidic) to 14 (alkaline), with 7 being neutral. As the concentration of hydrogen ions increases, the pH decreases, resulting in more acidic soil. Conversely, as the concentration of hydrogen ions decreases, the pH increases, resulting in more alkaline soil.

        Soil data was collected from Algonquin Provincial Park at each sample station by the imaging technician(s). To obtain the soil data at each sample station, a soil sample from 4” deep was taken and placed in a plastic comparator. Next, the contents of a green capsule were poured inside. Using the provided dropper, the comparator was filled with distilled water up to the indicated line. The comparator was shaken thoroughly, and its contents left to settle for roughly one minute. The comparator was then placed on the pH chart for a photo to be taken by the imaging technician. After the expedition, the data was combined with data collected every year since AP9. A decision was made to analyze the soil pH data together with water ion and tree data to identify correlations.

II Observations

A consensus was reached to create a piece of software that would extract the relevant data from each sample station and transfer it directly onto a spreadsheet. The program consisted of three processing stages. The first stage read the raw data directly from the source files provided by H. van Bemmel; if the substring of the soil pH data contained an asterisk or a blank space, the data was rendered invalid and marked as “N/A” on the spreadsheet. As shown in Fig. 1., raw data for soil pH in AP16 showed blank spaces. After the extraction stage, the data was eliminated and marked as “N/A”.

The second stage of the process was noise reduction, wherein outliers from each sample station were eliminated to allow for more accurate analysis. Data that displayed pH values greater than 14 were deemed invalid and marked as “N/A”. Work by Iglewicz and Hoaglin (1993) recommended the use of a modified Z-score for smaller sample sizes, demonstrated in the following relation,

                [1][pic 1]

Where:

        Mi = The modified Z-score of the evaluated data point

        xi = The value of the evaluated data point

        xm = The median of the data

MAD (Median Absolute Deviation) = The median of the absolute differences between each of the data points being evaluated and the median of the data

Iglewicz and Hoaglin recommended that a modified Z-score be labeled as a potential outlier if its absolute value was greater than 3.5. Following this recommendation, the data was processed a second time and outliers were relabeled on the spreadsheet as “N/A”. This can be seen in Fig. 1.; a data point for soil pH from AP9 had a modified Z-score of -4.3664, which qualified as an outlier and was marked “N/A”.

Fig. 1. The above table shows the soil pH for Sample Station #121 at different stages of data processing. The Stage 3 trend recognition identified this Sample Station as “falling”

The final stage of data processing involved basic trend recognition. It was agreed upon that data from any sample station containing fewer than four valid data points after noise reduction was insufficient and should not be considered for further analysis and marked as having a trend of “unknown”. All the valid data points were then evaluated, ignoring those marked as “N/A”. If the absolute difference between two subsequent data points was greater than 2, it was deemed an inconsistency; else it was deemed as either rising or falling. In the end, each sample station was marked “Inconsistent”, “Consistent”, “Rising”, or “Falling” based on the number of inconsistencies and rising or falling points.