Weighted Averaging Partial Least Squares Regression Research Articles

Pollen-based seasonal temperature reconstruction in Northeast China over the past 10,000 years, and its implications for understanding the Holocene Temperature Conundrum

The Holocene Temperature Conundrum refers to the mismatch between proxy-based temperature records and those based on climate model simulations. A possible reason for this mismatch is a putative proxy-based bias in reconstructed summer temperatures, and therefore, regional reconstructions of seasonal temperature are crucial for resolving the conundrum. In this paper, we reconstruct vegetation and climate changes over the last ∼10,000 years BP based on a high-resolution pollen record from Gushantun peatland, Changbai Mountains, Northeast China. Multiple quantitative reconstruction approaches were used and weighted averaging partial least squares regression (WAPLS) was found to be the most appropriate method for reconstructing Holocene temperature and precipitation. The reconstructed climate record shows that the Holocene Climate Optimum occurred between 8 ka and 6 ka and exhibited a cold month mean temperature that was 3 °C warmer than modern temperatures. Climate gradually cooled during late Holocene with a minimum cold month temperature of −19.6 °C. Four prominent cold events occurred around 8.7 ka BP, 7.8 ka BP, 5.7 ka BP, and 2.5 ka BP with an amplitude variation up to 3 °C. The synthesized seasonal temperature time series and a comparison with other proxies show that the decreasing trend in mean annual temperature is not a seasonal bias caused by summer temperature change. This study provides evidence of a Holocene seasonal temperature change at a regional scale and insights for further understanding of the Holocene Temperature Conundrum.

Reliability of presence‐only data for assessing plant community responses to climate warming

Climate warming has triggered shifts in plant distributions, resulting in changes within communities, characterized by an increase in warm‐demanding species and a decrease in cold‐adapted species – referred to as thermophilization. Researchers conventionally rely on co‐occurrence data from vegetation assemblages to examine these community dynamics. Despite the increasing availability of presence‐only data in recent decades, their potential has largely remained unexplored due to concerns about their reliability. Our study aimed to determine whether climate‐induced changes in community dynamics, as inferred from presence‐only data from the Global Biodiversity Information Facility (GBIF), corresponded with those derived from co‐occurrence plot data. To assess the differences between these datasets, we computed a community temperature index (CTI) using a transfer function, weighted‐averaging partial least squares regression (WA‐PLS). We calibrated the transfect function model based on the species–temperature relationship using data before recent climate warming. Then we assessed the differences in CTI and examined the temporal trend in thermophilization. In a preliminary analysis, we assessed the performance of this calibration using three datasets: 1) Norwegian co‐occurrence data, 2) presence‐only data from a broader European region organized into pseudo‐plots (potentially capturing a larger part of the species niches), and 3) a combined dataset merging 1) and 2). The transfer function including the combined dataset performed best. Subsequently, we compared the CTI for the co‐occurrence plots paired up spatially and temporally with presence‐only pseudo‐plots. The results demonstrated that presence‐only data can effectively evaluate species assemblage responses to climate warming, with consistent CTI and thermophilization values to what was found for the co‐occurrence data. Employing presence‐only data for evaluating community responses opens up better spatial and temporal resolution and much more detailed analyses of such responses. Our results therefore outline how a large amount of presence‐only data can be used to enhance our understanding of community dynamics in a warmer world.

LegacyClimate 1.0: a dataset of pollen-based climate reconstructions from 2594 Northern Hemisphere sites covering the last 30 kyr and beyond

Abstract. Here we describe LegacyClimate 1.0, a dataset of the reconstruction of the mean July temperature (TJuly), mean annual temperature (Tann), and annual precipitation (Pann) from 2594 fossil pollen records from the Northern Hemisphere, spanning the entire Holocene, with some records reaching back to the Last Glacial Period. Two reconstruction methods, the modern analog technique (MAT) and weighted averaging partial least squares regression (WA-PLS), reveal similar results regarding spatial and temporal patterns. To reduce the impact of precipitation on temperature reconstruction, and vice versa, we also provide reconstructions using tailored modern pollen data, limiting the range of the corresponding other climate variables. We assess the reliability of the reconstructions, using information from the spatial distributions of the root mean squared error in the prediction and reconstruction significance tests. The dataset is beneficial for synthesis studies of proxy-based reconstructions and to evaluate the output of climate models and thus help to improve the models themselves. We provide our compilation of reconstructed TJuly, Tann, and Pann as open-access datasets at PANGAEA (https://doi.org/10.1594/PANGAEA.930512; Herzschuh et al., 2023a). The R code for the reconstructions is provided at Zenodo (https://doi.org/10.5281/zenodo.7887565; Herzschuh et al., 2023b), including the harmonized open-access modern and fossil datasets used for the reconstructions, so that customized reconstructions can be easily established.

Phytolith Assemblages as a Promising Tool for Quantitative Canopy Coverage Reconstruction in Subtropical Forests, China.

This study investigates the reliability of phytolith assemblage analysis for characterizing subtropical vegetation and explores the potential for using these modern phytolith–vegetation relationships for paleoenvironmental interpretation in southeastern China. The samples were collected from five common subtropical vegetation communities in the Daiyun Mountains, southeastern China, with the above-ground vegetation recorded at each plot. Constrained ordination analysis was used to determine the most important factor governing the variations in phytolith assemblages that could be quantitatively reconstructed with weighted averaging partial least squares regression (WAPLS). The relationship between modern phytolith assemblages and the parent vegetation, as well as production, dispersal, and taphonomic processes, was discussed. Results demonstrated that the main subtropical biomes in southeastern China could be well distinguished by soil phytolith assemblages. In particular, the overall amount of tree coverage was well represented by topsoil phytolith assemblages. Grass silica short cell phytoliths (GSSCP) tended to occur in higher proportions in open habitats (shrub-meadow) at higher elevations, whereas non-grass phytolith morphotypes attained higher frequencies under mixed and broadleaf forests at lower elevations. Human-induced deforestation might increase the frequency of GSSCP within the bulk phytolith assemblage. Our results constitute the primary phytolith reference data for the subtropical zone in southeastern Asia where vegetation change during the Holocene period, particularly forest shifts, anthropogenic deforestation, and early agriculture are poorly documented.

Comparison of boosted regression trees vs WA-PLS regression on diatom-inferred glacial-interglacial climate reconstruction in Lake Tiancai (southwest China)

Quantitative paleoclimate reconstructions based on biological fossils over glacial-interglacial timescales are a major source of information on long-term climate variability. However, such reconstructions can present major methodological challenges, as calibration methods based on modern climatic and biological patterns may yield biased results, especially when data are particularly sensitive to the influence of secondary environmental variables in the training set or fossil assemblage. A machine-learning technique, boosted regression tree (BRT), is compared with a weighted averaging-partial least squares (WA-PLS) regression in a diatom-based summer (mean July) temperature transfer function with a calibration set of 273 samples collected from China. Using these calibration methods and a fossil diatom sequence from Lake Tiancai (southwest China; 3898 m a.s.l.), the summer temperatures covering the last glaciation and entire Holocene (18000 cal yr BP to present) are reconstructed. The record is validated by applying several statistical reconstruction diagnostics and compared with pollen and diatom records from the same sediment core. The results suggest that the BRT-based calibration model, which produced a coefficient of determination (r2 = 0.85) and root-mean-square error of prediction (RMSEP = 1.65 °C) offers substantial improvements over WA-PLS. The record shows that the summer temperature varies by ~2.5 °C across the entire period. The temperature evolution inferred by diatom-based BRT model is characterized by an early cool phase (with a MJT of 9.3 °C) prior to 10000 cal yr BP, followed by a clear mid-Holocene thermal maximum (~1.1 °C above the present-day temperature) from 6000 to 3500 cal yr BP and a subsequent cooling that ended at 1500 cal yr BP. This contrasts with the chironomid-based temperature transfer function in the same sequence showing a warmer period between 8500 and 6000 cal yr BP. The overall pattern of diatom-based summer temperature reconstructions broadly matches the climate change inferred from pollen records from Tiancai Lake and other records from the surrounding regions, suggesting that summer temperatures respond to insolation forcing and the variability of the southwest monsoon. We argue that the BRT-based climate model potentially outperforms the other models in southwest China when using a large and non-linear calibration set of diatoms for reconstructing long-term temperature.

An improved statistical approach for reconstructing past climates from biotic assemblages.

Quantitative reconstructions of past climates are an important resource for evaluating how well climate models reproduce climate changes. One widely used statistical approach for making such reconstructions from fossil biotic assemblages is weighted averaging partial least-squares regression (WA-PLS). There is however a known tendency for WA-PLS to yield reconstructions compressed towards the centre of the climate range used for calibration, potentially biasing the reconstructed past climates. We present an improvement of WA-PLS by assuming that: (i)the theoretical abundance of each taxon is unimodal with respect to the climate variable considered; (ii)observed taxon abundances follow a multinomial distribution in which the total abundance of a sample is climatically uninformative; and (iii) the estimate of the climate value at a given site and time makes the observation most probable, i.e. it maximizes the log-likelihood function. This climate estimate is approximated by weighting taxon abundances in WA-PLS by the inverse square of their climate tolerances. We further improve the approach by considering the frequency ( fx) of the climate variable in the training dataset. Tolerance-weighted WA-PLS with fx correction greatly reduces the compression bias, compared with WA-PLS, and improves model performance in reconstructions based on an extensive modern pollen dataset.

Soil-surface pollen assemblages and quantitative relationships with vegetation and climate from the Inner Mongolian Plateau and adjacent mountain areas of northern China

We statistically analyzed 396 soil-surface pollen samples from the Inner Mongolian Plateau and its adjacent mountain areas of northern China to gain insights into pollen–vegetation–climate. The results of indicator species analysis, canonical correspondence analysis (CCA) and boosted regression tree (BRT) analysis suggest that: 1) soil-surface pollen assemblages can differentiate between desert, steppe, meadow steppe, cool broadleaved forest, cool conifer forest, and temperate forest, but fail to differentiate between steppe and forest steppe; shrub and temperate forest; and temperate broadleaved forest, temperate mixed conifer-broadleaved forest, and eurythermic conifer forest; 2) pollen taxa, with low percentages but frequent occurrences, strongly indicate presence in the vegetation; 3) mean annual precipitation (MAP) is the most dominant variable influencing soil-surface pollen assemblages and the most promising climate variable for quantitative reconstructions. Weighted averaging partial least squares regression (WA-PLS), modern analogue technique (MAT), and boosted regression trees (BRT) were used to construct a series of pollen-climate calibration sets. WA-PLS and MAT outperform BRT under leave-one-out cross-validation. WA-PLS models are less susceptible to spatial auto-correlation than MAT models, making WA-PLS models generally the best choice. Our soil-surface pollen data will contribute to modern pollen data in eastern Asia.

Establishment of a diatom-total phosphorus transfer function for lakes on the Songnen Plain in northeast China

To understand the relationship between planktonic diatoms and environmental variables in the lakes on the Songnen Plain, northeast (NE) China, we investigated water quality and planktonic diatoms from 71 sampling points in 27 lakes, based on which Canonical Correspondence Analysis (CCA) was conducted. The result show that planktonic diatoms displayed certain responses to environment gradients, and the total phosphorus (TP) explained the maximum variation of planktonic diatom species among the 15 environmental variables, suggesting that TP was the most important and significant environmental parameter affecting the distribution of diatom species. In addition, we established a diatom-total phosphorus transfer (DTPT) function, of which component 5 of the weighted averaging partial least squares regression (WAPLS) was chosen to and compared with a series of weighted average regression (WA) models and WA-PLS models. According to the jackknifing statistical test, the component 5 of WA-PLS models provided a lower root-mean-square error prediction (RMSEP=0.202) and a higher correlation coefficient between observation and prediction (R2_Jack=0.759). After deletion of three outliers, the root-mean-square error prediction of the DTPT function was reduced (RMSEP=0.169) while the correlation coefficient between observation and prediction was increased (R2_Jack=0.823). Therefore, this DTPT function performed better than other regional TP models in the world. However, it remains demanding for expanding the background dataset to improve the prediction ability of the model.

Holocene vegetation and climate dynamics of NE China based on the pollen record from Sihailongwan Maar Lake

High-resolution palynological analysis on annually laminated sediments of Sihailongwan Maar Lake (SHL) provides new insights into the Holocene vegetation and climate dynamics of NE China. The robust chronology of the presented record is based on varve counting and AMS radiocarbon dates from terrestrial plant macro-remains. In addition to the qualitative interpretation of the pollen data, we provide quantitative reconstructions of vegetation and climate based on the method of biomization and weighted averaging partial least squares regression (WA-PLS) technique, respectively. Power spectra were computed to investigate the frequency domain distribution of proxy signals and potential natural periodicities. Pollen assemblages, pollen-derived biome scores and climate variables as well as the cyclicity pattern indicate that NE China experienced significant changes in temperature and moisture conditions during the Holocene. Within the earliest phase of the Holocene, a large-scale reorganization of vegetation occurred, reflecting the reconstructed shift towards higher temperatures and precipitation values and the initial Holocene strengthening and northward expansion of the East Asian summer monsoon (EASM). Afterwards, summer temperatures remain at a high level, whereas the reconstructed precipitation shows an increasing trend until approximately 4000 cal. yr BP. Since 3500 cal. yr BP, temperature and precipitation values decline, indicating moderate cooling and weakening of the EASM. A distinct periodicity of 550–600 years and evidence of a Mid–Holocene transition from a temperature-triggered to a predominantly moisture-triggered climate regime are derived from the power spectra analysis. The results obtained from SHL are largely consistent with other palaeoenvironmental records from NE China, substantiating the regional nature of the reconstructed vegetation and climate patterns. However, the reconstructed climate changes contrast with the moisture evolution recorded in S China and the mid-latitude (semi-)arid regions of N China. Whereas a clear insolation-related trend of monsoon intensity over the Holocene is lacking from the SHL record, variations in the coupled atmosphere-Pacific Ocean system can largely explain the reconstructed changes in NE China.

FORAMINIFERA AS ENVIRONMENTAL INDICATORS AND QUANTITATIVE SALINITY RECONSTRUCTIONS IN THE PEARL RIVER ESTUARY, SOUTHERN CHINA

A total of 117 surface sediment samples from sites across the Pearl River Estuary, southern China, were collected to analyze relationships between foraminiferal distributions and environmental parameters such as salinity, depth, substrate, and water energy. A total of 123 species of foraminifera were identified. A Q-mode cluster analysis resulted in five clusters (relative abundance ≥5%), which each corresponded to a particular species association, reflecting different sub-environments: (A) the Ammonia tepida–Elphidium excavatum association, found in distributary channels and river mouths of the estuary, with low salinity, shallow water, strong currents, and comparatively coarse sediments; (B) the Ammobaculites formosensis–Cavarotalia annectens association, found in brackish waters near Lantau Island and southeastern Lingding Bay, characterizing turbid brackish-water conditions and sediment consisting of a mixture of sand-clay-silt; (C) the Haplophragmoides canariensis–Ammobaculites agglutinans var. association, found in the artificial harbor near Macao and greatly affected by human activities; (D) the Ammonia beccarii var. –Cavarotalia annectens association, mainly found in the deep channels of two bays and representing euryhaline conditions; and (E) the Ammonia tepida–Elphidium advenum association, mainly found at depths >10 m, higher salinity, low energy, and fine-grained substrate, indicating relatively stable conditions. Agglutinated foraminiferal taxa are better indicators than calcareous taxa of littoral zone environments with a mixture of sand-clay-silt substrate. Benthic foraminiferal distributions in the surface sediments were strongly influenced by salinity. Consequently, a salinity transfer function was developed using weighted-averaging partial least-squares regression. The transfer function was tested on a sediment core collected near the Modaomen distributary, with the goal to reconstruct the palaeosalinity from ~10,100 to ~5960 cal yr BP, an interval with sufficient foraminiferal tests for analysis. At ~10,100 cal yr BP, marine waters flooded the Da’ao plain. Then, a semi-enclosed estuarine environment developed and persisted until ~5960 cal yr BP. Salinity at the core site increased gradually during the early Holocene (from ~10,100 to ~8200 cal yr BP), although a short-term decrease can be recognized. The peak salinity occurred between ~8520 and 8200 cal yr BP, when brackish-marine taxa reached their highest abundances. The core site experienced declining salinity from ~8200 cal yr BP onwards as a result of strong river discharge and a decrease of marine water influence at the core site.

A modern pollen–climate calibration set from central‐western Mongolia and its application to a late glacial–Holocene record

AbstractAimFossil pollen spectra from lake sediments in central and western Mongolia have been used to interpret past climatic variations, but hitherto no suitable modern pollen–climate calibration set has been available to infer past climate changes quantitatively. We established such a modern pollen dataset and used it to develop a transfer function model that we applied to a fossil pollen record in order to investigate: (1) whether there was a significant moisture response to the Younger Dryas event in north‐western Mongolia; and (2) whether the early Holocene was characterized by dry or wet climatic conditions.LocationCentral and western Mongolia.MethodsWe analysed pollen data from surface sediments from 90 lakes. A transfer function for mean annual precipitation (Pann) was developed with weighted averaging partial least squares regression (WA‐PLS) and applied to a fossil pollen record from Lake Bayan Nuur (49.98° N, 93.95° E, 932 m a.s.l.). Statistical approaches were used to investigate the modern pollen–climate relationships and assess model performance and reconstruction output.ResultsRedundancy analysis shows that the modern pollen spectra are characteristic of their respective vegetation types and local climate. Spatial autocorrelation and significance tests of environmental variables show that the WA‐PLS model for Pann is the most valid function for our dataset, and possesses the lowest root mean squared error of prediction.Main conclusionsPrecipitation is the most important predictor of pollen and vegetation distributions in our study area. Our quantitative climate reconstruction indicates a dry Younger Dryas, a relatively dry early Holocene, a wet mid‐Holocene and a dry late Holocene.

An expanded ostracod-based conductivity transfer function for climate reconstruction in the Levant

We present the first modern calibration dataset linking ostracod assemblage composition to water chemistry, and other site-specific variables, in the hydrologically and geopolitically sensitive southern Levant region. A total of 42 ostracod taxa were recorded from the 178 sampled sites in Israel and Jordan. Ilyocypris spp., Heterocypris salina and Cypridopsis vidua are the most abundant taxa. Species strictly confined to freshwater conditions are Prionocypris zenkeri, Gomphocythere ortali and Prionocypris olivaceus. In contrast, H. salina, Bradleytriebella lineata and Cyprideis torosa show high frequencies in brackish waters (waters with higher conductivity). Humphcypris subterranea, G. ortali, P. olivaceus and Cypridopsis elongata apparently prefer flowing waters. Specific conductivity optima and tolerance ranges were calculated for the recorded ostracod species and may be used for the palaeoenvironmental assessment of fossil ostracod assemblages. In addition, a transfer-function for quantitative specific conductivity estimation based on 141 samples was established with weighted averaging partial least squares regression (WA-PLS). The resulting coefficient of determination r2 between observed and predicted conductivity values (0.72) and the root-mean-square error of prediction (RMSEP) in % gradient length (13.1) indicate that conductivity may be reliably estimated from ostracod assemblage data. The transfer function was first applied to last glacial ostracod assemblage data from an archaeological trench in the Sea of Galilee (northern Israel). Relatively large conductivity fluctuations between ca 1 and 7 mS cm−1 were inferred for the period 24–20 cal ka BP. In addition, four episodes of freshwater influx near the site of the trench were identified from the presence of shells of freshwater and stream-dwelling species intermingled with very abundant shells of Cyprideis torosa. The results of our study allow a better use of Quaternary ostracods from the Levant as palaeoenvironmental indicators of water-body types and past conductivity levels and will contribute to a better understanding of Quaternary environmental and climate change in the Levant.

‘Diatoms and pH reconstruction’ (1990) revisited

The 167 sample lake-water pH-diatom calibration data-set created as part of the Palaeolimnology Programme within the Surface Water Acidification Project (SWAP) is re-analysed numerically using nine different numerical methods, six based on simple two-way weighted-averaging (WA), and the other three involving Gaussian logit regression (GLR) and maximum-likelihood (ML) calibration, the modern analogue technique, or weighted-averaging partial least-squares regression and calibration. Root mean squared error of prediction and maximum bias were estimated for all nine methods based on 10,000 internal and 10,000 external cross-validations involving a training-set, an optimisation-set, and a test-set. The results show that WA with a monotonic deshrinking spline equals or slightly outperforms WA with linear inverse deshrinking, especially in external cross-validation. Methods that employ tolerance downweighting generally have an inferior performance except when combined with monotonic deshrinking. It appears that simple two-way WA extensively used in SWAP cannot be significantly bettered. Thanks to increased computing power, better software, and more rigorous cross-validations, GLR shows good performance, especially in external cross-validation.

Developing a modern pollen–climate calibration data set for Norway

Bjune, A. E., Birks, H. J. B., Peglar, S. M. & Odland, A. 2010: Developing a modern pollen–climate calibration data set for Norway. Boreas, Vol. 39, pp. 674–688. 10.1111/j.1502‐3885.2010.00158.x. ISSN 0300‐9483.Modern pollen–climate data sets consisting of modern pollen assemblages and modern climate data (mean July temperature and mean annual precipitation) have been developed for Norway based on 191 lakes and 321 lakes. The original 191‐lake data set was designed to optimize the distribution of the lakes sampled along the mean July temperature gradient, thereby fulfilling one of the most critical assumptions of weighted‐averaging regression and calibration and its relative, weighted‐averaging partial least‐squares regression. A further 130 surface samples of comparable taphonomy, taxonomic detail and analyst became available as a result of other projects. These 130 samples, all from new lakes, were added to the 191‐lake data set to create the 321‐lake data set. The collection and construction of these data sets are outlined. Numerical analyses involving generalized linear modelling, constrained ordination techniques, weighted‐averaging partial least‐squares regression, and two different cross‐validation procedures are used to asses the effects of increasing the size of the calibration data set from 191 to 321 lakes. The two data sets are used to reconstruct mean July temperature and mean annual precipitation for a Holocene site in northwest Norway and a Lateglacial site in west‐central Norway. Overall, little is to be gained by increasing the modern data set beyond about 200 lakes in terms of modern model performance statistics, but the down‐core reconstructions show less between‐sample variability and are thus potentially more plausible and realistic when based on the 321‐lake data set.

Calibrating vascular plant abundance for detecting future climate changes in Oregon and Washington, USA

We propose using future vascular plant abundances as indicators of future climate in a way analogous to the reconstruction of past environments by many palaeoecologists. To begin monitoring future short-term climate changes in the forests of Oregon and Washington, USA, we developed a set of transfer functions for a present-day calibration set consisting of climate parameters estimated, and species abundances measured, at 107 USDA Forest Service FIA (Forest Inventory and Analysis) Phase 3 plots. For each plot, we derived climate estimates from the Daymet model database, and we computed species abundance as quadrat frequency and subplot frequency. We submitted three climate variables (mean January temperature, MJAT; mean July temperature, MJUT; and mean annual precipitation transformed to natural logarithms, MANPt) to canonical correspondence analysis (CCA), and verified their importances in structuring the species frequency data. Weighted averaging-partial least squares regression (WA-PLS) provided the means for calculating six transfer functions. In all cases, based on performance statistics generated by leave-one-out cross-validation, we identified two-component WA-PLS models as the most desirable. The predictive abilities of our transfer functions are comparable to, or better than, those reported in the literature, probably due both data quality and statistical considerations. However, model overfitting as a result of spatial autocorrelation remains a possibility. The large errors associated with our MJAT transfer functions connote that even the highest amount of change in mean January temperature predicted for Oregon and Washington for 2010–2039 would be indistinguishable from current conditions. The higher predictions indicate that our MJAT transfer functions may be able to track climate changes by the 2040s. Our MJUT transfer functions can detect change in mean July temperature under the highest projection for 2010–2039. Our MANPt transfer functions will be of limited use until the 2070s, given the predictions of only slight changes in mean annual precipitation during the early part of the twenty-first century. Our MJAT and MANPt transfer functions may prove useful at the present time to verify relative climatic stability. Because the predicted climate values sometimes deviate substantially from the observed values for individual plots, our transfer functions are appropriate for monitoring climatic trends over the entire Pacific Northwest or large regions within it, not for assessing climate change at individual plots.

Quantitative reconstruction of lake conductivity in the Quaternary of the Near East (Israel) using ostracods

Surface sediments, water samples and environmental data from 37 lakes, ponds and streams in Israel were analysed to determine the main variables controlling ostracod species distributions. Multivariate statistical analysis revealed that the greatest amounts of variation in the distribution of the ostracod taxa among the 37 water bodies were explained by the host water δD value (12.9%), water temperature (11.0%), mean January air temperature (10.5%), electrical conductivity (9.5%), and the Mg and NO3 concentrations (7.8 and 7.1%, ion concentrations as % of the anions or cations). A supplementary data set comprising ostracod species composition and electrical conductivity readings for 24 water bodies was available from previous research and was merged with the 37 samples data set to develop an ostracod-based transfer function for the reconstruction of electrical conductivities. A weighted averaging partial least squares regression (WA-PLS) provided the best results with a relatively high coefficient of determination (r2) between measured and inferred electrical conductivity values of 0.73, a root mean square error of prediction of 0.13 (13.4% of gradient length) and a maximum bias of 0.24 (23.9% of gradient length), as assessed by leave-one-out cross-validation based on 56 water bodies. The application of the EC transfer function onto (sub)fossil ostracod assemblages from Holocene and early to mid Pleistocene lake sediments provided EC values consistent with other proxies and demonstrated that Quaternary ostracod assemblages from subaqueous sediments can now be used to trace the hydrological history of water bodies in the Near East. A better understanding of past hydrological conditions in response to the natural climate variability is crucial in regions that face restricted water resources and rising demands in times of rapid climate and environmental change.

Holocene climate change and its effect on lake ecosystem production on Northern Victoria Island, Canadian Arctic

A multi-proxy paleoenvironmental study from Lake WB02 (72.29°N, 109.87°W) on Northern Victoria Island, Nunavut, Canada provides an 8.4-ka record of chironomid and ecosystem production. Mean July air temperatures for this region during the Holocene were inferred from the fossil record. The chironomid assemblages contained 33 taxa and were dominated by Paracladius and Heterotrissocladius maeaeri-type. Primary production and chironomid food availability inferred from sediment biogenic silica and loss on ignition at 550°C, and chironomid concentrations, all exhibited synchronous patterns of change through time. Similar to other climate records from across the Arctic, the sediment and fossil data from Lake WB02 support the hypothesis of a warm and productive early to middle Holocene, a cool and generally unproductive middle to late Holocene and a return to a warmer, more productive environment in the past 100 years. Mean July air temperature reconstructions based on both the modern analogue technique (MAT) and weighted averaging partial least squares regression (WAPLS), however, failed to reflect these same changes. The difference between the qualitative and quantitative environmental reconstructions may be due to the restrictions associated with the use of these inference tools, the effects of which are more significant in unproductive ecosystems such as Arctic lakes.

Environmental controls on peatland testate amoebae (Protozoa: Rhizopoda) in the North of Ireland: Implications for Holocene palaeoclimate studies

The environmental controls on modern peatland testate amoebae (Protozoa: Rhizopoda) in the North of Ireland were investigated to assess the potential for Holocene palaeoclimate research within this region. Canonical Correspondence Analysis (CCA) revealed that hydrological factors (water table depth and moisture content) are the most important abiotic controls on organism distribution. A series of partial CCAs showed that water table depth explains 15.8% and moisture content explains 5.5% of the total variance. Monte-Carlo permutation tests showed that the results are highly significant (p < 0.002; p < 0.040 respectively). Transfer functions were generated for water table depth using weighted averaging tolerance downweighted (WA-Tol) regression and for moisture content using weighted averaging partial least squares regression (WA-PLS). The performance of the models was assessed using leave-one-out cross-validation (jacknifing). After removal of outlier samples, the improved transfer functions were found to perform well with an r jack 2 and root mean square error of predictionjack of 0.83, 4.99 cm for water table depth and 0.76, 4.60% for moisture content respectively. The water table transfer function was applied to a fossil peat sequence from this region and reconstruction errors were generated by 1,000 bootstrap cycles. The water table reconstruction was also carried out using an established pan-European transfer function and was found to be similar to that based on the North of Ireland dataset. This demonstrates the persistent and comparable control of hydrological variables on the distribution of testate amoebae taxa across Europe and implies that regional training sets can suffice as long as no-analogue situations are not encountered.

Lateglacial climate development in NW Romania — Comparative results from three quantitative pollen-based methods

This paper provides quantitative reconstructions of the Lateglacial changes in four climate parameters from two fine-resolution pollen profiles in the Gutaiului Mountains, NW Romania. Climate estimates are based on two modern analogue techniques (with and without considering vegetation types) and weighted averaging partial least squares regression (WA-PLS), giving evidence for several climatic fluctuations during the period from > 14,700 to 11,500 cal. yr BP. The comparative results of the two modern analogue techniques show consistent trends of climate changes that are also coherent at both sites, but these results appear to largely disagree compared with climate reconstruction provided by WA-PLS. The modern analogue techniques revealed four intervals with low temperatures: prior to 14,700 cal. yr BP, between 13,950 and 13,800; 13,400 and 13,200; and 12,700 and 11,700 cal. yr BP. The temperature declines were more pronounced for winter than for summer, suggesting an intensification of seasonality, which together with a drop in precipitation indicates an increase in continentality. The Younger Dryas is the most pronounced cooling phase with winter temperatures ~ 14–16 °C colder than modern conditions, annual and summer temperatures ~ 2–5 °C and ~ 2 °C, respectively below present ones. Precipitation was ~ 400–500 mm, half that of present. During the Bølling and Allerød, summer temperatures were close to modern values (13 to 17 °C), whereas winter (− 6 to − 12 °C) and annual temperatures (0.5 to 6 °C) as well as precipitation were (550 to 700 mm) lower, indicating more continental conditions compared to the present-day climate.

An ostracod-conductivity transfer function for Tibetan lakes

About 145 freshwater to hypersaline lakes of the eastern Tibetan Plateau were investigated to develop a transfer function for quantitative palaeoenvironmental reconstructions using ostracods. A total of 100 lakes provided sufficient numbers of ostracod shells. Multivariate statistical techniques were used to analyse the influence of a number of environmental variables on the distributions of surface sediment ostracod assemblages. Of 23 variables determined for each site, 19 were included in the statistical analysis. Lake water electrical conductivity (8.2%), Ca% (7.6%) and Fe% (4.8%, ion concentrations as % of the cations) explained the greatest amounts of variation in the distribution of ostracod taxa among the 100 lakes. Electrical conductivity optima and tolerances were calculated for abundant taxa. A transfer function, based on weighted averaging partial least squares regression (WA-PLS), was developed for electrical conductivity (r 2 = 0.71, root-mean-square-error of prediction [RMSEP] = 0.35 [12.4% of gradient length], maximum bias = 0.64 [22.4% of gradient length], as assessed by leave-one-out cross-validation) based on 96 lakes. Our results show that ostracods provide reliable estimates of electrical conductivity and can be used for quantitative palaeoenvironmental reconstructions similarly to more commonly used diatom, chironomid or pollen data.