Feasibility of using near-infrared reflectance spectroscopy for the analysis of C, N, P, and diatoms in lake sediments

Abstract

The present study explored whether rapid, non-destructive near-infrared reflectance spectroscopy (NIRS) could complement conventional paleolimnological and chemical analyses of sediment cores for greater efficiency and cost-effectiveness. The study used a 47-cm long freeze-core from the deepest point in Lake Arendsee, Mecklenburg Plain in northern Germany taken in 1993 to elucidate eutrophication history and to identify the pre-impact algal communities in this system. The core had been analyzed for total C, CO32-, N, P, and diatoms. Thirty-four of the 47 1-cm thick core sections were scanned by NIRS and calibrations were developed for total C, CO32-, N, P, N:P, total diatoms, and three dominant diatom species with different sedimentary profiles (Stephanodiscus binatus, Cyclotella rossii, and Fragilaria crotonensis). Total C ranged from 167-194 mg g-1 dry weight (d.w.), CO32- from 31.3-66.4 mg g-1 d. w., N from 9.9-17.4 mg g-1 d. w., and P from 0.7-6.0 mg g-1 d. w. Calibrations developed using multiple linear regression between NIR-predicted values and chemically-measured values were excellent for P (r2 > 0.99), good for C, N, and N:P (r2 > 0.93), and satisfactory for CO32-(r2 > 0.8). Calibrations for total diatoms and for individual species were highly statistically significant (r2 between 0.54 and 0.69). Although the calibrations are not useful for reliable predictions of the content of diatoms in the samples, the results indicate that NIRS detects spectral properties associated with diatoms or lake conditions when they were present, and that further work is warranted to attempt to improve the results. The study demonstrated that rapid, non-destructive, simultaneous analysis of total C, CO32-, N, P, and N:P in sediment cores is feasible.