The heterogeneity of the subfossil chironomid deposition (in terms of assemblages and paleo-diet) has to be studied to assess the reliability of paleolimnological interpretations (i.e., spatial representativeness). We collected 31 cores in the deepest part of Lake Grand Maclu (Jura, France) with the aim of assessing the spatial structures at two spatial scales: a sediment core surface (31.7 cm2) and 4 m2 in the deepest zone of the lake. The results showed that chironomid assemblages exhibited a uniform distribution at the two spatial scales studied. Moreover, a sediment core surface was very reliable for assessing the biodiversity (Simpson’s index) of the deepest part with more than 95% of the overall diversity sampled on average. However, δ13C values of Chironomus anthracinus-type remains revealed a spatial pattern in the contribution of carbon sources to its biomass. We suggest that a heterogeneous distribution of microbial biomass (such as methane-oxidizing bacteria) could induce a spatial pattern (patch) in food availability for chironomid larvae. We recommend increasing the number of head capsules analyzed in spectrometry to reduce this potential bias due to patterns and to improve upscaling results from individual downcore records to the entire deepwater basin. © 2015, Springer International Publishing Switzerland.