Humboldt Lake, Saskatchewan

1. Salinity Plots

Using diatoms to infer lake salinity, the adjacent plot was generated for the 1000+ core years from Humboldt Lake. Notice the recent (~1900) shift to much "fresher" water.

Figure 1.1: Plot of diatom-inferred salinity over time in Humboldt Lake.

The above record of fluctuations in salinity is converted to a detrended salinity time series by fitting a regression and plotting the residuals. This detrended salinity series is then used in further analyses of droughts (above zero) and floods (below zero).

Figure 1.2: Plot of detrended diatom-inferred salinity over time in Humboldt Lake.

2. Drought Analyses

For a drought of a given intensity (standardised among lakes and identified much like the standard deviations shown in the salinity plot above) we can calculate various statistics that describe its past behaviour over time. Below are the stats for a drought that was equal to the Moon Lake drought of 1988. In the Humboldt area, a drought of this intensity was experienced in 1909.

We can also calculate the frequency at which droughts have occurred in the historical record. The adjacent graph shows results from the extraction of a signal from the apparent "noise" of the salinity plot. The analysis extracts frequencies of 32.5 and 81.2 years from the Humboldt drought record using the upper portion of the detrended salinity plot.

Figure 2.1: Spectral periodicity of Humboldt area droughts.

To look into the future, we add a proabilistic element to our drought time series and predict future likelihoods based on the past distribution of events (see Analytical Protocol). The adjacent graph shows predictions of future probabilities for a drought similar to that in 1909 (severity of 1.4) over the next 30 years. The bottom panel demonstrates how these probabilities change as we change the severity of the drought (most severe droughts are lowest in the plot).

Figure 2.2: Probability plots of Humboldt area droughts.

3. Flood Analyses

We can calculate the same statistics for floods of a given intensity as those we obtained for droughts (except using the negative values from the detrended salinity plot). Below are the stats for a flood with a defined severity of 1.4. In the Humboldt area, a flood of this intensity occurred in 1979.

We can also calculate the frequency at which floods have occurred in the historical record. In this case, the bottom half of the detrended salinity plot is analysed. The adjacent graph shows results from the extraction of a signal from the apparent "noise" of the salinity plot. The analysis extracts frequencies of 81.2 and 162.3 years, as well as a shorter peak of 11-12 years, from the Humboldt flood record.

Figure 3.1: Spectral periodicity of Humboldt area floods.

As for droughts, we can also predict future flood likelihoods based on the past distribution of events (see Analytical Summary or Details). The adjacent graph shows predictions of future probabilities for a flood of 1979 severity (1.4 on climatic severity scale) over the next 30 years. The bottom panel demonstrates how these probabilities change as we change the severity of the flood (most severe floods are lowest in the plot).

Figure 3.2: Probability plots of Humboldt area floods.

4. Pigment Analyses

To examine the effects of climate on long-term algal community composition, we analysed sediment cores for algal pigment concentrations. Photosynthetic pigments are also preserved in the sediments and by creating pigment time series at the same resolution as the salinity time series, comparisons between climatic shifts and ecosystem productivity can be made.

We generated algal pigment time series via HPLC (High Pressure Liquid Chromatography) analyses of the same sediment layers from which diatoms were counted. The various plots shown here represent shifts in different components of the algal community over time.

Figure 4.1: Plot of algal pigments over time in Humboldt Lake.

Comparisons of salinity-pigment relationships to date have revealed that changes in climate do not appear to affect ecosystem productivity (total algal abundance), but the composition of the algal community does appear to shift to bloom-forming taxa (greens and blue-greens) under a drier climate.

Figure 4.2: Plot of pigment concentrations vs. salinity in Humboldt Lake.

5. Recent History

Establishing relationships between recorded historical data and the drought record is another component of the drought project. For instance, defining a relationship between drought and crop data can provide a measure of the responsiveness of crop yield to drought events in recent history. In the Humboldt watershed there have been a number of "disturbances" (sewage diversion, conversion of native prairie to cropland, fish stocking, etc.) following settlement of the area that may have also had an impact on the diatoms and the algal community in general. Attempting to tease these effects apart is an ongoing endeavour.

As a first cut at understanding the relationship between recent climate and salinity, we examined the correspondence of a number of different drought indices (the Bhalme-Mooley is shown here) and our record of salinity changes. As the adjacent plots shows, the relationships were not strong. The comparative lack of recent drought events in the Humboldt area may prove hard to capture in the salinity record. However, many different factors contribute to a "drought" from a crop's perspective. Analysis of crop data is in progress...

Figure 5.1: Comparison of the Bhalme-Mooley drought Index (at Muenster, SK) and recent salinity in Humboldt Lake.

One recent impact on Humboldt lake has been the diversion of sewage from the town's lagoon. The sediment record of diatoms appears to faithfully record an increase in the abundance of taxa that are typically associated with "enriched" or eutrophic conditions around this same time. Other factors will also be examined, but this pattern appears to be clear evidence of an anthropogenic impact on the lake ecosystem, over and above any climatic effects.

Figure 5.2: Abundance plots of selected taxa from Humboldt Lake in recent history.

6. Preliminary Conclusions - Humboldt Lake

Last updated on 15 January 2001 by Jim Rusak