Rapid,non-destructive and simultaneous method such as diffuse reflectance infrared spectroscopy provides potentially useful alternative to time-consuming chemical methods of soil or sediment elemental analysis.To assess the utility of near-infrared reflectance spectroscopy(NIRS) for the determination of elemental concentrations in the ornithogenic sediments and penguin palaeoecological reconstruction,four sediment cores(Y2,Y2-4,Y4,AD3) on the Ardley Island in the maritime Antarctic were analyzed by both chemical and spectroscopic methods.All these sediment cores have been proven to be influenced by penguin droppings in our previous studies.Stepwise multiple linear regression and principal component regression approaches were used to develop calibration models for predicting the concentrations of nine bio-elements including P,Ca,Cu,Zn,Se,Sr,Ba,F and S. The results showed that the correlation coefficients between optimal spectra-predicted and chemically analyzed concentrations of these nine bio-elements were quite high(more than 0.90),and the predicated concentration-versus-depth profiles were very consistent with those from actual chemical analysis.Furthermore,in order to infer the palaeoecological significance of spectra data,principal component analysis(PCA) was used to reconstruct the fluctuation of historical penguin population.Compared with the change pattern of penguin size inferred from the concentration variations of nine bio-elements,they had a very similar trend.Overall,this study demonstrated that using reflectance spectroscopy to infer palaeoecological information recorded in Antarctic ornithogenic sediments is feasible.
