Empirical models for describing recent sedimentation rates in lakes distributed across broad spatial scales

Abstract

Over the last 20 years there has been a surge of interest in paleolimnology and as a result a large accumulation of lake sedimentation records. This emerging archive has allowed us to develop empirical models to describe which variables explain significant variation in sedimentation rates over the past ∼150 years across large spatial scales. We hypothesized that latitude would be a significant explanatory variable of profundal zone lake sedimentation rates across a temperate to polar gradient. We further hypothesized that along a more longitudinally-constrained dataset (i.e. east coast of North America), latitude would explain a greater proportion of the variance. To test these hypotheses, we collated data from 125 natural, average-sized lakes (with surface area <500 km2) by recording authors’ estimates of sedimentation rates (measured as mm/year) or by digitizing recent sediment profiles and calculating sedimentation rates over the past ∼150 years. We found that, at both scales, latitude was the strongest predictor of lake sedimentation rates (full dataset: r2 = 0.28, P = 0.001, n = 125; east coast dataset: r2 = 0.58, P < 0.001, n = 43). By conducting a multiple linear regression analysis, we found that 70% of the variance in sedimentation rates from the east coast transect was explained by latitude and elevation alone. This latter model is of sufficient strength that it is a robust predictive tool. Given that climate and land-use strongly co-vary with latitude and that both of these factors have previously been shown to influence lake sedimentation rates, it appears that latitude is a surrogate measure for climate and land-use changes. We also show support for land-use as an important variable influencing sedimentation rates by demonstrating large increases in recent versus Holocene accumulation rates. These results indicate that it is possible to make generalizations about sedimentation rates across broad spatial scales with even limited geographic data.