Content uploaded by Eric Christopher Grimm
Author content
All content in this area was uploaded by Eric Christopher Grimm on Dec 12, 2018
Content may be subject to copyright.
A preview of the PDF is not available
CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares
Abstract and Figures
Stratigraphically constrained cluster analysis is a multivariate method for quantitative definition of stratigraphic zones. As opposed to ordinary, unconstrained cluster analysis, only stratigraphically adjacent clusters are considered for merging. The method of incremental sum of squares has been used widely for unconstrained analyses and has proved particularly satisfactory for pollen frequency data. CONISS is a FORTRAN 77 program for stratigraphically constrained cluster analysis by this method. Several data transformations lead to different implicit dissimilarity coefficients. As an option, the program also will perform an unconstrained analysis, which can be useful for comparison with the constrained analysis.
Figures - uploaded by Eric Christopher Grimm
Author content
All figure content in this area was uploaded by Eric Christopher Grimm
Content may be subject to copyright.
... The relative abundance of each macrofossil type or unidentified organic matter was estimated as the percentage by volume. A stratigraphically constrained cluster analysis (CONISS) was used to determine the zonation of macrofossil assemblages using the Tilia program (Grimm, 1987). Fossil pollens were analyzed at 1-cm intervals for cores UIC13-3 and KRP13-2. ...
... Pollen grains from the samples were identified and counted to a sum of at least 100 grains under a compound microscope. A stratigraphically constrained cluster analysis (CONISS) was similarly used to determine the zonation of pollen assemblages using the Tilia program (Grimm, 1987). ...
... The macrofossil and pollen diagrams are made on the Tilia program, available from https://www.neotomadb.org/apps/tilia (Grimm, 1987). ...
Recent amplified climate warming in the Arctic has caused profound changes in terrestrial ecosystems, with the potential for strong feedback on climate change. Arctic tundra landscapes have developed patchy and thin organic soil (peat) layers at the surface that may continue to grow into mature peatlands and become a larger carbon sink under future warming. Here we use paleoecological analyses of multiple soil and peat cores collected from the North Slope of Alaska to document and understand the formation and development histories of tundra peat patches and permafrost peatlands. We find a consistent peat development sequence for peat patches, first from mineral soils to sedge peat during the Little Ice Age, and then to Sphagnum peat during the recent warming with high carbon accumulation rates. These findings suggest that climate cooling is likely critical for the initial peat buildup on tundra soils due to reduced decomposition, whereas climate warming triggers the regime shift into an increased abundance of Sphagnum mosses that are likely central to enhancing their carbon sink capacity. Additionally, peat patches become similar to permafrost peatlands in the vicinity in terms of ecosystem processes and carbon dynamics, and therefore may have developed the same ecohydrological feedback system to maintain their long‐term stability. This study implies that the potential future expansion of peat patches into peatlands may strongly alter the carbon balance of Arctic tundra, supporting the new United Nations Environment Programme's report that calls for incorporating widespread shallow peat into understanding the peatland–carbon–climate nexus.
... In this study, fossil pollen spectra were divided into 7 pollen zones in terms of the cluster analysis result of the main pollen percentages [50]. A PCA (principal component analysis) of the main pollen percentage data was performed to assist the vegetational interpretation of fossil pollen spectra and zones. ...
... (Minneapolis, IL, USA) [49]. In this study, fossil pollen spectra were divided into 7 pollen zones in terms of the cluster analysis result of the main pollen percentages [50]. A PCA (principal component analysis) of the main pollen percentage data was performed to assist the vegetational interpretation of fossil pollen spectra and zones. ...
Vegetation dynamics data since the Last Glacial Maximum (LGM) are essential for our understanding of ecosystem shifts and vegetation responses to climate change. Here, we present a pollen record covering the last 25,000 years from Lake Fuxian in central Yunnan, southwest China. Our study shows seven stages of vegetation dynamics since the LGM: The early LGM (stage 7 of 25,000–21,200 cal. a BP) witnessed less dense regional vegetation dominated by pine forests, evergreen broadleaved forests (EBFs), deciduous broadleaved forests (DBFs), montane hemlock forests, and fir/spruce forests. The late LGM (stage 6 of 21,200–17,500 cal. a BP) saw an expansion of grasslands, wetlands, and montane fir/spruce forests as well as a shrinkage of EBFs and DBFs. During the last deglaciation (stage 5 of 17,500–13,300 cal. a BP), dense regional vegetation was dominated by EBFs as well as deciduous oak and alder forests. The densest regional vegetation occurred in stage 4 of 13,300–11,200 cal. a BP, roughly equal to the Younger Dryas Chron, when pine forests, DBFs, EBFs, grasslands, and wetlands grew in the Lake Fuxian catchment. During the early to mid-Holocene (stage 3 of 11,200–5000 cal. a BP), dense regional vegetation was dominated by sweetgum forests, in addition to some pine forests and EBFs. After 5000 cal. a BP, the regional vegetation density became lower and lower, and forests became thinner and thinner. Pine forests expanded to their maximum of the entire sequence in stage 2 of 5000–2500 cal. a BP. A big deforestation event occurred in stage 1 (the last 2500 years), when grasslands, wetlands, and cultivated vegetation dominated regional vegetation in the catchment of Lake Fuxian. The regional vegetation since the LGM in the catchment of Lake Fuxian also experienced six major transitions, five centennial shift events, and one big large-scale and long-term deforestation event. These resulted from the responses of regional vegetation to climate changes during the LGM, last deglaciation, and early–mid-Holocene, as well as human influence in the late Holocene. The vegetation density since the LGM has changed with the 25° N summer insolation.
... (2) pCI = palaeo-climate index assigned for each 100-year time slice; (dry = 1; intensified = 11; Tables 2 and 3; Figure 3); w is the weight, and the subscripts j and t represent the number of publications and the number of time slices respectively. Using the assigned WApCI for the 29 available datasets, cluster analysis was performed on TILIA software (Grimm, 1987; Figure 3). The cluster analysis helped us in identifying different climatic zones temporally. ...
The Indian Central Himalaya Region (ICHR), the northern topographic front of the Indian summer monsoon (ISM), is an ideal location to study topography-climate interactions because two weather systems—the ISM and mid-latitude westerlies (MLW)—create distinct eco-climatic regimes from tundra (north) to tropical (south). The region’s paleoclimatic studies show considerable climatic variations since the late Pleistocene. We evaluated 29 paleoclimatic records from the region and synthesized the results semi-quantitatively using the weighted palaeoclimate index (WApCI) to better understand the important climatic events and their driving mechanisms. According to the WApCI, the region has at least six enhanced monsoonal periods and eight drier spells during the past 34 ka. The cold-dry climatic events, such as the last glacial maxima (LGM), Younger Dryas (YD), 8.2 ka, and 4.2 ka, are associated with northern-hemisphere (NH) climate dynamics and propagated via MLWs. While, the warmer phases are dictated by the insolation-driven ISM dynamics. The WApCI’s reconstructed rainfall anomaly aligns with paleoclimatic-model experiments for dynamically generated Paleoclimate Modeling Intercomparison Projects (PMIP3/PMIP4) rainfall for chosen time-slices (last-millenium, historical, mid-Holocene, and LGM). Finally, the Granger causality test determines the temporal relationship between the climatic drivers/forcing indices and primary meteorological parameters. The results showed that summer and post-monsoon precipitation is primarily influenced by total solar irradiation, winter precipitation is driven by a complex mix of variables, and pre-monsoon precipitation is driven by the Arctic oscillation. Based on the facts, we hypothesize that past climate variability demonstrates a complex interplay of local and hemisphere teleconnections in ICHR’s climate dynamics.
... Data visualization for stratigraphic diagrams was done using Tilia 3.0.1 (Grimm, 2011). We performed Bray-Courtis dissimilarity (Faith et al., 1987) based temporally constrained hierarchical clustering (CONISS) analysis (Grimm, 1987) to distinguish changes in the Cladocera community structure. The number of significant clusters was determined using the broken stick method (Jackson, 1993). ...
Our research aimed to evaluate, how urbanization affects lake ecosystems and Cladocera in particular. For this purpose, we chose a small urban lake with a well-documented history. Lake Velnezers (located in Riga, Latvia) is currently surrounded by apartment building complexes. Construction works around this lake started in the 1950s and continued up until the 1970s. To investigate how the transition from forested to agricultural and further urbanized land affected the lake ecosystem we took a sediment core that covers the time period from before 1875–2018. We evaluated ecological changes in the lake based on chemical and Cladocera species composition in sediment records and linked these changes to the historical information about alterations in the landscape around Velnezers. Our results show lake transitioned from oligotrophic to eutrophic conditions already before urbanization. The Lake ecosystem reacted to urbanization gradually, showing small changes in the beginning. However, in the 1980s lake experienced rapid deterioration in water quality – sediment records show an increase in heavy metal pollution, anoxia, and nutrient input. These stressors resulted in Cladocera functional group structure changes and loss of Cladocera species richness and diversity. Improvements in nature protection – such as wastewater management have reduced heavy metal and nutrient input into Lake Velnezers towards the present. However, previous deterioration, i.e. loss of species diversity and phosphorous legacy effect do not allow natural lake recovery under current conditions.
The chronology of the Mikulino Interglacial and its individual phases have been the subject of discussion. The goal of this study was to evaluate the time limits of the main stages of the Mikulino Interglacial on the Russian Plain according to ²³⁰Th/U dating and paleobotanical studies of lake and peat sediments from the known sections located within the Tver region on the Bolshaya Dubenka River, Malaya Kosha River, Granichnaya River, and Sizhina River (“Kileshino-2” section). An improved geochronological approach has been applied to identify layers suitable for the ²³⁰Th/U isochronous approximation. In combination with pollen and carpological studies of the deposits, this made it possible to date units corresponding to relatively narrow time intervals in the development of plant formations at different stages of the Last Interglacial. New paleobotanical studies of buried lake and peat sediments from the sections located on the Bolshaya Dubenka River, Malaya Kosha River, and Granichnaya River allowed us to restore the vegetation development during the Mikulino Interglacial in the interval of pollen zones M1–M7, i.e., more pollen zones have been analyzed and in greater detail than in 1960–1970. A chronological scheme of the main stages of vegetation development in the Mikulino Interglacial is proposed based on the results of ²³⁰Th/U dating and paleobotanical studies of organic-rich deposits from the Tver region sections in combination with previously published data obtained for the “Nizhnyaya Boyarshchina” section from the Smolensk region. The Mikulino Interglacial had begun about 130–126 kyr ago. Its first phase, corresponding to the M2 zone, ended ca. 118 kyr ago. The pre-optimal stages of vegetation development (M3 and M4 zones) fit into the time range of ca. 118–112 kyr ago. The climatic optimum of the interglacial (M5 and M6 zones) began ca. 112 kyr ago and ended ca. 100 kyr ago. The duration of the Mikulino Interglacial was probably at least 25 thousand years.
This research explores the vegetation compositional changes between the fifth and third millennium BC on the Karelian Isthmus (north-west Russia). Special emphasis is placed on studying the timing and magnitude of the impact of hunter-fisher-gatherers on the vegetation. First, we reconstruct the local vegetation around Lake Bolshoye Zavetnoye by using the Landscape Reconstruction Algorithm modelling. The application of different scenarios to the relevant source area of pollen is used to assess the local vegetation around Lake Bolshoye Zavetnoye (BZL) at specific distances from 500 to 3000 m. The regional vegetation reconstruction is assessed by using the REVEALS model. Second, we present new pollen and charcoal data from adjacent Lake Ikmenlampi (LI). Third, we calculate indices of vegetation change for BZL, LI and the region. We further explore the potential causes for these differences between the spatial scales and compare all these outcomes with local and regional archaeological data. The results show that foragers actively manipulated their environment. A shift towards more heliophilous conditions is shown by ca 4800 BC. Between 4000 and 3000 BC, an increase in archaeological material suggests intensive resource procurement and landscape management activities, particularly near settlements. Around the BZL site a local decrease in birch is observed from 4100 BC, coinciding with an increase in the rate of vegetation changes. Additionally, a decline in tree species (spruce, hazel, Alder) from 3500 to 3300 BC suggests human activities. The high fire frequency recorded between 4090 and 3150 BC further supports the presence of local human disturbances by the deliberate use of fire to create favourable living conditions. The results from the LI site go in the same direction with fluctuating abundances of spruce and the presence of pollen from Cannabis, Hordeum, Urtica and Plantago lanceolata from ca 4000 to 3600 BC, indicating the use of spruce-dominated forests and an early, incipient cultivation.
Studies of sediment, chronology, fossil pollen and charcoal from cores from Lost Trail Pass Bog (2152 m) provide the first postglacial bog, forest, and fire history for the Bitterroot Mountains. The 6.27 m of sediment, dated by 16 radiocarbon dates and two volcanic ashes, represent lake, fen, and bog deposition spanning the last 12,000 yr. Lycopodium spores were introduced as tracers into the 81 constant-volume samples to estimate pollen and charcoal influx. Because of considerable variation between samples, pollen and charcoal estimates were averaged by pollen zones. Glacial ice withdrew leaving a lake by 12,000 yr ago and sagebrush steppe dominated the landscape for the next 400 to 500 yr. If lodgepole and whitebark pine are the diploxylon and haploxylon pine pollen in the record, then by 11,500 yr ago whitebark pine forests replaced the steppe and persisted for the next 3000 to 4000 yr under climatic conditions that were probably cooler than present. Two falls of Glacier Peak volcanic ash, separated by less than 25 yr, occurred about 11,250 BP. By 7000 yr ago, under warmer but not necessarily drier climatic conditions, Douglas-fir and lodgepole pine replaced whitebark pine and charcoal influx increased. The fall of Mazama volcanic ash was dated at about 6700 yr ago. By 5000 yr ago aquatic, fen, and bog microfossils became important. With the return to cooler climates, about 4000 yr ago, Douglas-fir was no longer common in the pine forest. Little vegetational change is indicated after 4000 years ago. However, more charcoal was deposited during the last 2000 yr than during the previous 9500 yr.
The recent interest in numerical classification and its application in the biological sciences to the evaluation of taxa has prompted the introduction of a large number of clustering processes, many of which are justified by empirical results. There is clearly a need for the formulation of a theoretical approach to the subject, and this is aided by the comparison and generalisation of existing processes by analytic methods. The work of Lance and Williams in this direction is supplemented here by results derived for the method of optimizing an error sum of squares objective function, and an algorithm is given which can be programmed to compute a general classification process for six accepted methods.
A method of classification is proposed for the situation where there are external constraints on which objects may come together to form a group. An example of such constraints occurs in palaeoecology, where the objects to be classified are samples down a vertical bore, and it is desired to preserve the stratigraphical information in the classification. The method inserts a barrier between neighboring objects, the height of the barrier being related to the similarity of the sets of objects on either side of it. The relationship of this approach to other classificatory procedures is considered. The method is applied to a palaeoecological example, and more general applications are indicated.