ABSTRACT: Foraminifera (Rhizaria) and testate amoebae (Rhizaria and Amoebozoa) are single-celled organisms with marine and continental affinities. They are regarded as valuable bioindicators in transitional areas such as tidal and salt marshes and have been widely used for (paleo)-environmental characterization. Along-lasting debate regularly occurs on the use of living vs. dead fauna to accurately represent modern environments. Moreover, environmentally based benthic foraminiferal and testate amoebaean studies need a reliable sampling strategy to capture the spatial variability, particularly in transitional environments where patchiness complicates the data interpretations. The objective of the present study is therefore to define the minimum required number of replicates for capturing the variability of either living (environment) or dead (paleoenvironment) benthic foraminifera and testate amoebae. To address this question, 49 samples (i.e., replicates) were selected from a square meter zone in the tidal flat-salt marsh transitional zone along the Canche Estuary (Northern France). The range of faunal spatial variability was measured using geostatistical tools. The minimal number of samples for capturing the patchiness was determined using bootstrap resampling procedure. We provide evidence that for both living and dead fauna more than 26 samples (and even higher for some species) are needed to correctly evaluate the patchiness. Indeed, the living and dead fauna do not follow spatial homogenous trends, and this might bias paleoenvironmental interpretations. The commonly-suggested number of three replicates might not be enough to characterize the fauna in such heterogeneous environments.