Within the scope of SIRRO (Siberian River Run off) expeditions of RV Akademik Boris Petrov were carried out in the Kara Sea during August and September 1999, 2000 and 2001. One major aim of the project is characterisation and quantification of terrigenous sediment primarily supplied by Ob and Yenisei rivers. We used a high resolution CHIRP echosounding system for selected profiles and additionally the ships sonar was used to obtain data of Holocene sediment thickness. The main objective of this study is the identification and characterisation of sediment echo types in conjunction with geophysical and sedimentological investigations of the collected gravity cores.During the expedition the geological work concentrated on a detailed sediment profiling survey by means of an ELAC sediment echograph (operating frequency of 12 kHz, impulse length of 1 ms, and 200 W transmitting power) as well as a GeoChirp sytem (2-8 kHz), and the sampling of bottom sediments using Multicorer, Giant Box Corer, and Gravity Corer for ground truthing.The history and extent of the Quaternary glaciations in Eurasia is controverse. Within this context, the largest remaining uncertainty is how far the Kara Sea ice sheet extended towards the east during the LGM (Svendsen et al., 1999; Polyak et al., 2000, 2002; Stein et al. 2002). The key question related to this problem is: how did the discharge of the siberian rivers respond/interact to an proximal ice sheet?Based on echosounding results we are able to reconstruct the palaeo drainage network of Ob and Yenisei rivers. Both rivers incised into the recent shelf, leaving filled and unfilled river channels and river canyons/valleys behind. The rivers were in the braided/meandering stage depending on the variable shelf gradient. Asymmetrical channel levee complexes with incision depths of 60 meters and more developed. The rivers formed channel levee complexes which in some places border to glacial dominated morphology, which implies fluvial deflection by an ice shield. This finding denots the non existence of an ice sheet on large areas of the Kara Sea shelf. Furthermore mapping of sediment thickness and character reveals no evidence for an ice dammed lake, as postulated by some workers.Sidorchuk et al. (2001) found macromeanders formed during the LGM until 14000 years BP, fluvial features an order of magnitude bigger than recent fluvial dimensions. Although the mean precipitation was lower than today, they related these features to an increased discharge which main cause is in their opinion the existence of permafrost, whose degradation (as recent) leads to the development of increased groundwater flow. Therefore it seems reasonable to relate fluvial features on the Kara Sea shelf, developed during sea level lowstand, to a pronounced riverine discharge as proposed by Sidorchuk et al. (2001).