Palaeoecology of Three Lakes - Data Page
The interactive map of Three Lakes can be found on the map page (opens in a new tab).
Three sediment cores have been taken from the Three Lakes site, either in or near the Middle Lake, as being central to the site as a whole. One core (core TL1) was taken near the old railway line about 200 metres from the northern lake shore and at a height about 3 metres above lake level. The second core (core TL2) was taken in a more or less central position of the main lake basin through 3 metres of water. The third core (core TL3) was from on the edge of the floating mat in about 1.5 metres water depth.
TL1 revealed 5.9 metres of peat, and lake mud, the upper layers containing many pieces of wood. The peat is underlain by 10 cm of blue grey silt with a rapid transition from lake bed gyttja to silt over 2 cm. The silt appears to be unoxidised and fairly free of organic matter. This core was taken using a russian corer which extracts 0.5 metres of sediment at a time. Bottom was reached purely on account of the difficulty in pushing in, and extracting, the full length of corer and rods. Once the silt was struck, further progress was not possible; there might well be more silt or further organic sediment below the level reached. This location is to be revisited with a gouge corer to hopefully find bottom and explore, and extract, the lowest levels of sediment. If the bottom most silt from base up to the top of the upper silt can be extracted in one length, this might clarify the depths of the layers across the area where the lowest long core and the bottom short core join in TL3 (see below).
TL2 was an exploratory core using a gouge corer, a narrow hemispherical 1 metre long sampler. At a location just to the north of the lake on dry (ish) ground off the floating mat, almost 7 metres of sediment was found, with two layers of silt, one at about 5.80 metres, followed by organic sediment, and a lower silt layer at bottom.
TL3 was taken in the lake using a piston corer, through c. 1.5 metres of water. About 6.5 metres of sediment was obtained and this is in the process of being analysed. Two layers of silt were encountered, one at the bottom - again stones were felt. Overlying these stones was the bottom layer of silt, fine grained, clear of organic material and cohesive. Above this a layer of organic sediment. Above this again a layer of blue grey silt, and then further up and all the way to the surface, dark brown organic sediment.
TL4, taken in the lake using a piston corer, acquired just c 3.5 metres of lakebed gyttja, organic, and fine grained, dark brown to black in colour. A shallow depth of silt was found at bottom, but this fell out of the piston corer as it was extracted and was lost. A bottom of stones was found below this 10 to 15 cm of silt and is assumed to be a glacial till.
Initial interpretation is that the lowest bed of silt is post glacial sediment deposited as water and regolith was released from freezing conditions. Colonisation of the land and water followed, resulting in an organic sediment. A return to cold conditions resulted in vegetation dying and a final organic sediment before the second silt layer was deposited. Climatic improvement then gave rise to a return of vegetation and other life and deposition of organic sediment. The climatic downturn is possibly that known as the Younger Dryas, which occurred around 12,800 BP and remained for just over 1000 years. The second silt layer from the bottom may be associated with this period.
Radiocarbon dating of three whole peat samples core TL1 at 1.0 m, 2.9m and 5.8 m depths, gave dates as follows:
- Sample no. Depth Age Date of Deposition
- (radiocarbon years) (calibrated)
- UBA-39773 1.0 m 1417 +/- 25 AD 597- 659
- UBA-39775 2.9 m 8868 +/- 43 BC 8223- 7830
- UBA-39774 5.8 m 10026 +/- 49 BC 9808- 9359
Radiocarbon dating for these three samples was paid for by a generous grant under the Bill Watts Chrono Award from IQUA, for which I am most grateful.
The following charts illustrate these radiocarbon dates, the calibrated dates, age plotted against depth, sediment accumulation and rate of deposition. They have been calculated and drawn using the Bchron package in R through RStudio. Bchron uses the statistical programming functions in R to illustrate the Bayesian interpretation of the chronology. Bchron is available here.
The important thing to remember is that radiocarbon dates give a bounded set of probable dates, in which, graphically, the highest part of the curve represents the highest probability. The total area of the curve represents all possible dates at all probabilities. It is because of the calibration that a radiocarbon date cannot result in a single definitive date; the calibration takes into account the variability of the propotion of the unstable Carbon 14 isotope of carbon in the environment over time. The calibration curve illustrates this - this has been constructed using intcal20, a calibration curve that is based entirely on tree rings as far back as 13,900 BP. Intcal20 actually extends back further in time to 55,000 BP, using other environmental factors - refer to the paper by Reimer et al "The IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0–55 cal kBP)" (open Access) published in the journal Radiocarbon. For the Holocene therefore the calibration is based on dendrochronology, that is the radiocarbon date of a sample of wood is calibrated against dating by tree ring analysis.
First, an age / depth chart based on the radiocarbon dates obtained for the samples from TL1. An additional date and depth has been inserted, namely depth zero and age zero, representing the surface sediment being dated to the present. This is a fair assumption since the current state of the coring location does not suggest any disturbance or extraction of peat etc.
The grey date curves represent the possible real dates calibrated from the calibration curve, with a total of 95% probability. The higher parts of the grey curves represent the higher probability of that date. The blue areas illustrate the possible intermediate ages at depths between those actually dated - again, this is a probability range. Note that the only true points of reference are the dated points, and even they are probabilities. Estimations of age between those depth points are just that - estimations. It is entirely possible that sediment deposition halted or even sediment was eroded (less likely), but on the basis of what we know, the age-depth chart is a best guess.
The second chart illustrates the possible sediment accumulation rates. The lines are percentiles, so each line marks that percentage of possible points occurs below that - for example, the 50th percentile line indicates that 50 percent of the possible accumulation rates occur below that line.
And finally, the possible sedimentation rates for the whole core from 580 cm depth to surface.
It is clear from these two charts that sediment accumulation rate (cm / year) and sedimentation rate (years / cm) changed over time. With only three defined dates at the specified depths the estimated changes occur at those points. In reality changes in these rates were probably more variable and took place over longer periods of time. The only way to get closer to this reality is to obtain more radiocarbon dates, but they are not cheap. But analysis of the microfossils may well shed light on other changes which will imply the fluctuation of sediment deposition, and maybe even the reasons.
The cores, TL1 and TL3 specifically, and also probably TL4, though it is incomplete, will be processed and then analysed in detail, using the methods described on the Laboratory Methods page. Results of analysis for the various depths examined, for each core, will be stored and made available for public access. Physical samples, processed samples, and prepared slides, will also be stored for the purpose of later retrieval and re-examination as necessary.
Radiocarbon dating of samples from core TL3 will be undertaken. The samples to be dated will focus on the earlier organic sediment below the upper silt - proably the lowest of the top layer of organic sediment, and the topmost and bottommost of the lower organic sediment.
Prior to radiocarbon dating, preliminary comparison of centimetre slices between the transition to the top of the silt layer at the bottom of TL1 and the transition to the top of the upper silt layer of TL3 suggests that TL1 593-594 slices compare to TL3 589-590 slices in both colour and texture.
- Items of interest to be investigated are:-
- Diatom, sponge, chrysophyte, and other microfossils occurrence in the lower silt. Assuming this silt is immediately post glacial, can we detect the colonisation of water bodies following the ice melt?
- Following dating of samples immediately before (below) and immediately after (above) the upper silt, can we ascertain the deposition rate of this non organic deposit? Will this shed some light on how sediment was transported in environments with no or minimal vegetation cover?
- On examination of the lower organic layer, if this does relate to pre-younger Dryas, can we therefore ascertain how rapidly the vegetation disappeared with a climatic downturn?
- Likewise, can we ascertain how quickly the vegetation returned after that downturn? This would be demonstrated by microfossils in the lowest levels of the organic sediment above the upper silt.
Results will be posted as investigation continues.
The assumption was that the deepest part of the valley would be beneath the lake, somewhat closer to the south side of the valley. From the depths of bottom during coring it would appear that this is not the case. Either the deepest part of the basin lies to the north of the middle lake, or there are several basins. Indeed, the most likely explanation is that following glacial erosion and formation of the valley basin, ice melt and deposition of sediment with further fluvial influx of sediment resulted in a series of mounds in the basin area, of which the currently visible hillock is the highest. These highpoints might equally well be ridges of bedrock, or a combination of bedrock and sediment drapes. Further investigation to ascertain this may involve taking depth soundings in the lake, ground penetrating radar, or physical probing, in all cases to ascertain the bottom of the sediment and the start of stony till or even bedrock.
As described on the Three Lakes page - see link below - a cursory survey of vegetation in the area has been undertaken and related to Fossitt's Guide to the Habitats of Ireland. Further more detailed vegetation surveys will be undertaken.
- Fossitt, J. 2007. Guide to Habitats in Ireland. The Heritage Council.
- JNCC. 2010. Handbook for Phase 1 habitat survey - a technique for environmental audit.
- Website http://webgis.archaeology.ie/historicenvironment/