Chironomid-based reconstruction of 500-year water-level changes in Daihai Lake, northern China

Abstract

In recent decades, lakes in arid and semi-arid regions of northern China have been rapidly shrinking, threatening the area’s ecological security. However, the impacts of climate change and human activities on lake shrinkage remains unclear due to short instrumental records and a lack of long-term lake-level change records. To address this issue, we reconstructed the water depth changes of Daihai Lake, a typical lake experiencing rapid shrinkage in northern China, over the last 500 years using lake-level-sensitive chironomid assemblages. We established a regional quantitative transfer function of chironomid water-depth using 54 surface sediments from Daihai Lake and Bosten Lake in northern China. The partial least squares (PLS) model yielded a coefficient of determination (R2jack) of 0.85, a maximum bias of 1.75 m, and a root mean square error prediction (RMSEP) of 1.58 m demonstrating the reliability of inferred water depth. Several diagnostic methods were used to verify the results, and they were compared with a 60-years instrumental record. The reconstructed data revealed that the lake level was low during 1500–1780 CE, gradually increasing to its peak in the 1960s and then rapidly decreasing by approximately 10 m in the last few decades. By comparing the results with regional historical documents, palynology and stalagmite records, and meteorological data, we concluded that climate change was the main driver of lake level dynamics before the 1960s, while human activities dominated the rapid shrinkage of the lake in recent decades. Our results provide a valuable limnological dataset for the water management in Daihai Lake and a reliable method for reconstruct the lake-level changes.