Testing the use of molluscs to infer climate with special reference to the late Cenozoic molluscs of the Meade Basin, southwest Kansas.
Date on Master's Thesis/Doctoral Dissertation
Alexander, James E.
Paleoecology; Freshwater molluscs; Meade Basin; Land molluscs; Kansas; molluscan assemblages
Mollusks--Kansas; Mollusks, Fossil--Kansas
Twenty-six fossil assemblages of land and freshwater molluscs from the Pliocene to the present, were collected from locations near the town of Meade, Kansas, U.S.A. and were analyzed (along with the extant molluscan fauna of Meade County, Kansas) to look for changes in molluscan diversity through time. The fossil assemblages were analyzed for two reasons: I) to test the hypothesis that diversity (measured as taxonomic richness, dominance, turnover and habitat type) did not change through the five million years of this study (HOI); and 2) to test the hypothesis that there is no relationship between the taxonomic composition of a molluscan assemblage and the local climate (H02)' The latter hypothesis was tested by re-examining prior molluscan paleoecological studies, by analyzing the molluscan assemblages collected for this study and through ordination analysis. The results of the first analysis clearly showed that the taxonomic structure of the assemblages changed through time (HOI was falsified), but the changes appeared to be random fluctuations. Richness ranged from 7 taxa (in assemblages AGO, FAL, RYA, SPA) to 30 (C03B) with r2 = 0.237. Dominance ranged from 1.40 (SPA) to 9.66 (RTA) with r2 = 0.113. The Habitat Ratio ranged from -0.43 (FAL) to 1.00 (X I E) with r2 = 0.029 and Turnover ranged from 0.17 (AGO, F AL) to 1.00 (RNT), when each assemblage was compared to RNT (r2 = 0.024). No discernable long-tern trend in taxonomic diversity or community composition was observed. The results of the second analysis indicate that molluscs, as a group, are not as useful as climate indicators as previously supposed. Today, both land and freshwater mollusc species are broadly geographically distributed and thus are found in a variety of locations with different local climates and vegetation. Previously, Miller (1975, 1976) grouped molluscan species with similar environmental tolerances into four units called Climate Groups and used them to infer the past climatic conditions of a region. However, only a weak climate signal was detected (ANOY A found significant differences among the Groups for minimum temperature [p < 0.0001], maximum temperature [p < 0.0001] and annual precipitation [p = 0.0082]), after the previous results were reanalyzed using the methods of this study. The fossil molluscan assemblages collected for this study displayed no climate signal (ANOY A found no significant differences among the assemblages for minimum temperature [p = 0.0714] or precipitation [p =0.691] but did find a difference in maximum temperature [p = 0.0207]. A subsequent Tukey-Kramer HSD test failed to find significant differences among assemblages). Finally, the results from ordination analysis of my data did not show a strong relationship between the climate variables used in this study (minimum and maximum temperatures, annual precipitation) and the current geographic distribution of molluscan taxa. One should proceed with caution if molluscs are to be used to interpret climate. The results of this study do not provide a strong endorsement for using molluscs as paleoclimate indicators. This finding is counter to the prevailing wisdom among paleontologists. More work needs to be done in the area of molluscan biogeography and physiology to see if the conclusions herein hold up and to better understand molluscan biology in general.
Steffen, Brian William 1975-, "Testing the use of molluscs to infer climate with special reference to the late Cenozoic molluscs of the Meade Basin, southwest Kansas." (2010). Electronic Theses and Dissertations. Paper 1376.