Informatics Publishing Limited

Kaustubh Thirumalai ¹, Arvind Singh ², R. Ramesh ²

Affiliations
1 Jackson School of Geosciences, University of Texas, Austin, TX - 78713, US
2 Physical Research Laboratory, Ahmedabad - 380 009, IN

Abstract

MATLAB^TM is a powerful, easy to use, software package suitable for many mathematical operations, which finds plenty of scientific applications. One such application is the fitting of trend lines for a given data set so as to interpret the relationship of the variance of the parameters involved. We provide here a code in MATLAB^TM that performs the weighted linear regression with (correlated or uncorrelated) errors in bivariate data which can handle 'force-fit' regression as well.

Keywords

Weighted Linear Regression, Correlated Errors, Force Fit, Trend Line, Bivariate Regression.

Full Text

     

Kaustubh Thirumalai ¹, Steven C. Clemens ²

Affiliations
1 Department of Geosciences, University of Arizona, 1040 E, 4th Street Tucson, AZ 85721, US
2 Department of Earth, Environmental, and Planetary Sciences, Brown University, 324 Brook Street, Providence, RI 02912, US

Abstract

Planktic foraminiferal geochemistry has yielded extensive insights into Cenozoic climate change over timescales ranging from centuries to millions of years. Additionally, recent studies have targeted reconstructing sub-centennial climate signals, including past seasonality, using the stable oxygen isotopic composition (δ¹⁸O) in individual tests of planktic foraminifera. Here we assess the signal involved in reconstructing past Indian monsoon strength and seasonality using bulk-foraminiferal and individual foraminiferal analysis (IFA). We focus on three areas offshore India and study the regional sensitivity of foraminiferal carbonate to monsoonal runoff using a forward-model that includes newly re-evaluated relationships between local salinity and seawater-δ¹⁸O. We then evaluate the utility of foraminiferal-δ¹⁸O across these regions in reconstructing monsoon variability. We find that the foraminiferal-δ ¹⁸O signal is dominated by processes that control seawater-δ¹⁸O variations in the Bay of Bengal versus temperature variations in the Arabian Sea and discuss the implications for the IFA-δ¹⁸O signal. Our findings support that both bulk and individual foraminiferal records developed offshore India can provide skilful reconstructions of past monsoon rainfall variability. We conclude that statistical analyses such as that provided in this work can offer useful blueprints to interpreting the relationship between monsoon rainfall and foraminiferal geochemistry.

Keywords

Forward-Modelling, Individual Foraminiferal Analysis, Palaeomonsoon, Planktic Foraminifera.

Full Text