Results & Discussion

Changes to Peat Physicochemical Conditions

Box plots were used to compare the differences in soil moisture, depth of rust, surface temperature, and below ground temperature for each peat extraction treatment (Figure 32.).

Figure 32. Box plot displaying surface peat physicochemical parameters for each peat extraction treatment. Different colours represent different peatland treatments. Boxplots with a dot pattern represent the Avenir field location; no pattern corresponds to the Seba Beach field location. Whiskers indicate the minimum and maximum values not including outliers. Outliers are represented by black dots. (a) Soil moisture measured in percent volume. High values indicate saturated soil conditions. (b) Surface temperature measured in degrees Celsius. (c) Depth of rust observed on a steel rod inserted into the ground, measured in centimeters. Negative values indicate depth below ground surface where rust was observed. Increasing negativity indicates deeper aerobic soil conditions. (d) Temperature 10 cm below the ground surface measured in degrees Celsius.

Natural Sites

Natural site predictions:

    • Hummock

      • ↓ soil moisture

      • ↑ surface temperature

      • ↑ below ground temperature

      • ↑ depth of rust

    • Hollow

      • ↑ soil moisture

      • ↓ surface temperature

      • ↓ below ground temperature

      • ↓ depth of rust

Natural site observations:

    • Hummock

      • low soil moisture

      • moderate surface temperature

      • high below ground temperature

      • high depth of rust

    • Hollow

      • moderate soil moisture

      • low surface temperature

      • low below ground temperature

      • low depth of rust

Key take-away and discussion points: As expected, the measured factors were very different between the hummock and hollow locations. However, there is a large range of natural variability especially for the soil moisture and depth of rust. This may be due to different moss types or sampling error (ie. the steel rod was not inserted to the same depth at each site because of ice, roots, and other debris below ground).

Extracted Sites

Extracted site predictions:

    • Extracted

      • ↓ soil moisture

      • ↑ surface temperature

      • ↑ below ground temperature

      • ↑ depth of rust

Extracted site observations:

    • Extracted (young)

      • moderate soil moisture

      • moderate surface temperature

      • moderate below ground temperature

      • high depth of rust

    • Extracted (mature)

      • moderate soil moisture

      • high surface temperature

      • moderate below ground temperature

      • moderate depth of rust

    • Extracted (complete)

      • moderate soil moisture

      • moderate surface temperature

      • moderate below ground temperature

      • moderate depth of rust

Key take-away and discussion points: Soil moisture was higher than expected in all three locations. Depth of rust was lower than expected in the mature and complete restored sites, but the aerobic layer was within the PRS probe depth. This may be due to smaller pore sizes in the older peat deposits that may inhibit oxygen. Surface temperatures were elevated at the mature site, but all three locations recorded temperatures >30°C.

Restored Sites

Restored site predictions:

    • Restored

      • ↑ soil moisture

      • ↓ surface temperature

      • ↓ below ground temperature

      • ↓ depth of rust

Restored site observations:

    • Restored (unsaturated)

      • moderate soil moisture

      • moderate surface temperature

      • high below ground temperature

      • moderate depth of rust

    • Restored (saturated)

      • high soil moisture

      • low surface temperature

      • high below ground temperature

      • low depth of rust

Key take-away and discussion points: The restored sites differed greatly in soil moisture, depth of rust and surface temperature. The restored "saturated" site had soil moisture and depth of rust values as expected; however, the below ground temperature was higher and the surface temperature was lower than all other sites. This may be due to the presence of shallow water at or near the surface that is exposed to the sun and could influence the below ground temperature. The presence of vegetation may shade surface peat, and evapotranspiration from the open water may keep temperatures cool. The unsaturated restoration site behaved more similarly to the extracted sites in terms of soil moisture, depth of rust, and surface temperature. This is likely due to the presence of the ditch which would partially lower the water table and the relative lack of vegetation that may increase surface temperatures.

Single Factor ANOVA with Random Blocks

Table 3. Single factor ANOVA with random blocks using a mixed model for each physicochemical parameter (depth of rust, soil moisture, surface temperature, and below ground temperature). Model indicates the code used in R, Treatment is the peat extraction treatment (natural hummock, natural hollow, extracted, restored), SS is the sum of squares, MS is the mean squares value, Num df is the numerator degrees of freedom, Den df is the denominator degrees of freedom. An alpha level of 0.05 was used; significant p-values are identified with *.

A single factor ANOVA with random blocks was used to observe any significant relationships between physicochemical parameters and the peatland treatments.

Soil Moisture

There is a significant difference in soil moisture between different peat extraction treatments (Table 3.). The least square means and confidence intervals displayed in Figure 33. indicate that soil moisture levels are higher in restored peatlands compared to natural hummocks. There was no significant difference in soil moisture between natural hummocks, natural hollows, and extracted sites, or between natural hollows, extracted sites and restored sites.

Below Ground Temperature

There is a significant difference in below ground temperature between different peat extraction treatments (Table 3.). The least square means and confidence intervals in Figure 34. indicate that below ground temperatures are significantly different between natural hummocks, hollows, and extracted sites. The below-ground temperature for the restored treatments was not significantly different from any of the other treatments.

Figure 33. Least squares means for soil moisture in percent volume for each peat extraction treatment. Error bars represent upper and lower 95% confidence intervals. Lowercase letters indicate significant differences among treatments. Treatments with the same letter are not significantly different at alpha = 0.05.

Figure 34. Least squares means for below ground temperature in degrees Celcius for each peat extraction treatment. Error bars represent upper and lower 95% confidence intervals. Lowercase letters indicate significant differences among treatments. Treatments with the same letter are not significantly different at alpha = 0.05.

Take Home Messages: Physicochemical Conditions

The overall lack of significant difference between extraction treatments when observing the depth of rust, surface temperature, and overall soil moisture initially suggests that ditching and vegetation removal do not have a substantial impact on those in situ physicochemical properties between treatments (Table 3., Figure 33.). This is not surprising, considering the large amount of variability and small sample size in this study. However, it is surprising that soil moisture does not appear to decrease in extracted peatlands when compared to natural hollows, yet the depth of rust (indicator of oxygen content) is generally deeper (Figure 32.). Although not significant, this suggests that oxic conditions may persist deeper into the soil profile and allow for enhanced decomposition. Similarly, Price et al. (2003) observed elevated soil moisture content in areas with lowered water tables, however, they suggest this impedes oxygen and decreases microbial activity. Because this study observes the top 10 cm of the peat surface, the soil may be just wet enough to enhance decomposition, but not so wet that oxygen is decreased.

Further it appears that extraction activities may increase below ground temperature when compared to natural hollows and provide more hospitable microbial conditions; however, when considering the variability within the natural site, there does not appear to be any major change (Figure 34.). Harris et al. (2020) reported cooler peat temperatures in a drained bog compared to an undrained bog. The current evaluation of significance in this study does not separate the different extracted and restored peatland types. With a larger, more representative sample size for each location, the variability between locations observed in Figure 32. may become more pronounced and help differentiate more clearly between treatment types.

Changes in Nitrogen Availability

Box plots were used to compare the differences in nitrogen composition and availability for each peat extraction treatment (Figure 35.).

Figure 35. Box plot displaying nitrogen availability at each peat extraction treatment. Whiskers indicate the minimum and maximum values not including outliers. Outliers are represented by black dots. Data are displayed on a logarithmic scale, but have not been transformed. Different colours represent different peatland treatments. Boxplots with a dot pattern represent the Avenir field location; no pattern corresponds to the Seba Beach field location. PRS probes were installed for 4 weeks at 5 to 10 cm below the ground surface. (a) Nitrogen available as ammonium in µg / 10 cm2 / 4 weeks. (b) Nitrogen available as nitrate in µg / 10 cm2 / 4 weeks.

Changes in Ammonium (NH4) Availability

The presence of ammonium indicates some form of decomposition has occurred - either in aerobic or anaerobic conditions

Natural Sites

Natural site predictions:

    • Hummock - ↓ ammonium

    • Hollow - ↑ ammonium

Natural site observations (Figure 35a.):

    • Hummock - low

    • Hollow - low

Key take-away and discussion points: There was almost no variability between peatlands or between hummocks and hollows. The very low concentration of ammonium suggests limited decomposition is occurring and/or living plants are removing all available ammonium from the system before it can be adsorbed onto the PRS probe.

Extracted Sites

Extracted site prediction:

    • ↑ ammonium

Extracted site observations (Figure 35a):

    • Extracted (new) - low

    • Extracted (active) - high

    • Extracted (old) - high

Key take-away and discussion points: There was huge variability within and between the extracted sites. The active and old sites contain high amounts of available ammonium. This suggests that there is decomposition occurring and/or decreased plant uptake has resulted in more available ammonium. The lower relative ammonium availability in the young extraction site may be due to differences in peat substrate. Higher carbon to nitrogen ratio material takes longer to break down and may result in lower decomposition rates. This requires future testing. Harris et al. (2020) observed no advanced decomposition in hummocks in a drained bog, suggesting that similar factors may be at play at the you extraction site.

Restored Sites

Restored site predictions:

    • ↑ ammonium

Restored site observations (Figure 35a.):

    • Restored (unsaturated) - moderate

    • Restored (saturated) - low

Key take-away and discussion points: The unsaturated restored site had more ammonium compared to the saturated site. This is as likely because elevated soil moisture, anoxia, and lower surface temperatures have slowed decomposition in the saturated site. Further, vegetation uptake may be removing any available ammonium that is produced.

Changes in Nitrate (NO3) Availability

The presence of nitrate indicates that ammonium was present and was nitrified into nitrate under aerobic conditions

Natural Sites

Natural site predictions:

    • Hummock - ↓ nitrate

    • Hollow - ↓ nitrate

Natural site observations (Figure 35b):

    • Hummock - absent

    • Hollow - absent

Key take-away and discussion points: As with the ammonium, there was no variability between peatlands or between hummocks and hollows. The absence of nitrate is expected because there is very limited ammonium available to be nitrified into nitrate.

Extracted Sites

Extracted site prediction:

    • ↑ ammonium

Extracted site observations (Figure 35b):

    • Extracted (young) - absent

    • Extracted (mature) - moderate

    • Extracted (complete) - high

Key take-away and discussion points: The mature and complete extraction sites had elevated nitrate availability, while the young site does not. This is likely due to lowered decomposition in the young site. Low soil moisture in the complete extraction site may account for the increase in nitrate availability because aerobic decomposition may have increased (Figure 32.).

Restored Sites

Restored site predictions:

    • ↑ ammonium

Restored site observations (Figure 35b):

    • Restored (unsaturated) - absent

    • Restored (saturated) - absent

Key take-away and discussion points: Available nitrate was absent in both restored sites. Low ammonium availability and flooded conditions impede aerobic decomposition, which may account for the absence of nitrate (Figure 32.).

Changes in Phosphorus Availability

Box plots were used to compare the differences in phosphorus composition and availability for each peat extraction treatment (Figure 36.).

Figure 36. Box plot displaying phosphorus, aluminum, iron, and calcium availability at each peat extraction treatment. Whiskers indicate the minimum and maximum values not including outliers. Outliers are represented by black dots. Data are displayed on a logarithmic scale. Different colours represent different peatland treatments. Boxplots with a dot pattern represent the Avenir field location; no pattern corresponds to the Seba Beach field location. PRS probes were installed for 4 weeks at 5 to 10 cm below the ground surface. (a) Available phosphorus in µg / 10 cm2 / 4 weeks. (b) Available aluminum in µg / 10 cm2 / 4 weeks. (c) Available iron in µg / 10 cm2 / 4 weeks. (d) Available calcium in µg / 10 cm2 / 4 weeks.

Changes in Phosphate (PO4) Availability

Phosphate mobility is complex and is driven by decomposition as well as the availability of calcium, aluminum, and iron.

Natural Sites

Natural site predictions:

    • Hummock - ↓ phosphate

    • Hollow - ↑ phosphate

Natural site observations (Figure 32.):

    • Hummock - low

    • Hollow - high

Key take-away and discussion points: Phosphate availability was low in the hummocks and high in the hollows as predicted, but the variability within the hollows was large and outliers suggest the range of values may be more variable.


Extracted Sites

Extracted site prediction:

    • ↓ phosphate

Extracted site observations (Figure 32.):

    • Extracted (new) - moderate

    • Extracted (active) - low

    • Extracted (old) - low

Key take-away and discussion points: Extracted sites were low in phosphate with the exception of the young site. Low levels coincide with higher aluminum availability.


Restored Sites

Restored site predictions:

    • ↑ phosphate

Restored site observations (Figure 32.):

    • Restored (unsaturated) - high

    • Restored (saturated) - low

Key take-away and discussion points: The variability between both restored sites was very large. The unsaturated site had high phosphate availability, while the saturated site had very low availability. High iron levels in the saturated site as well as active plant uptake may be decreasing the phosphate availability.

influence of Physicochemical Parameters on Nutrient availability

Table 4. Spearman rank correlations results correlating each physicochemical parameter (depth of rust, soil moisture, surface temperature, and below ground temperature) with each nutrient supply rate (NH4, NO3, P, Al, Ca, Fe). R indicates the Spearman correlation coefficient, R2 is the proportion of the variance described, Adjusted p-value is the p-value adjusted for multiple inference for 6 comparisons. An alpha level of 0.05 was used; significant p-values are identified with *.

Table 4. shows that after adjustment for multiple comparisons, there are no significant correlations between physicochemical parameters and nutrient availability for alpha = 0.05. This may suggest that there are additional factors at play that may be controlling how nutrients move within the system, or that the sample size in this study is too small to have much statistical power. Despite the lack of significance, it appears that there is a weak positive relationship between depth of rust and Fe and Ca availability, as well as a weak positive relationship between Ca and soil moisture. This may inadvertently have an impact on phosphorus concentrations; however, additional analysis is required to explore these complex relationships.

Further, there are weak positive relationships between both ammonium and nitrate and surface temperature (Table 4.). This suggests that warmer locations are more likely to have increased products of decomposition (Table 4.). Further, there appears to be a weak, positive correlation between ammonium and soil moisture and nitrate appears to have a weak, negative correlation with soil moisture (Table 4.). This is as predicted, and tentatively suggests that ammonium is present in areas with more moisture, while nitrate is in locations with lower water saturation.

Conclusions

Peat extraction activities appear to have a limited impact on depth of rust and surface temperatures when compared to the natural variability found in hummocks and hollows. There appears to be a difference in below ground temperatures, with extracted sites significantly warmer than natural hollows, but cooler than natural hummocks. Soil moisture was low in natural hummocks and high in restored sites, but it is surprising that soil moisture was not lower in extracted peatlands despite ditching. The combination of warmer soils and appropriate levels of soil moisture my promote increased decomposition and nutrient availability in extracted sites. Thus, this study suggests that peat extraction activities may influence soil moisture and below ground temperatures, but have a limited impact on soil aeration and surface temperature. Additional samples are required to further improve the confidence of these findings.

Despite these relative similarities in physicochemical properties between treatments, mature and complete extracted peatlands had an elevated availability of ammonium and nitrate which could potentially leach off site. The young extracted site had higher soil moisture levels, moderate phosphate availability, lower available ammonium, and no available nitrate in comparison, suggesting that decomposition was slowed. Therefore, other factors such as the peat age and decomposition level (fibric, mesic, humic) may play a role in determining the decomposition rate and nitrogen availability in combination with changes to peat physicochemical properties.

Restored sites had inconsistent phosphate availability, with higher levels in the unsaturated site despite lower soil moisture. There appears to be weak positive relationships between Fe and Ca and depth of rust which may be influencing phosphorus availability. This suggests that phosphate availability is complicated, but increased leaching may occur in fields with low vegetation cover and low iron. Further, adding additional sampling units and differentiating between the extracted and restored treatments in future ANOVA and correlation analyses may help understand the parameters influencing nutrient availability more clearly as there is currently insufficient evidence to correlate the tested physicochemical properties with nutrient availability.

This suggests that mature and complete extracted sites, as well as restored sites with unsaturated soil conditions could be at higher risk for leaching and should be monitored. Further research is required to pinpoint the environmental conditions necessary to increase nutrient availability in each peat extraction phase. However, although there is the potential for available nutrients to move through the peat and leach off site at these locations, the hydrological flow path within each extraction phase is necessary to understand. Even if there is a high potential for many nutrients to leach, they cannot leave site if there is no water to transport them!