Estimates of contemporary effective population size (N e) can provide valuable information for genetic conservation and monitoring, pinpointing populations at higher risk of genetic erosion, decreased fitness, maladaptation and, ultimately, demographic decline. There are however potential limitations in the application of commonly employed genetic estimators of contemporary N e to widespread forest tree populations. Genetic isolation by distance within populations, small and spatially restricted samples, among-population gene flow and overlapping generations are factors that can potentially affect the accuracy of marker-based estimates of contemporary N e , depending on the demographic scenario. In particular, we illustrate the uncertainty faced by forest researchers and managers when interpreting contemporary N e estimates obtained for continuously distributed tree populations with large census size N. To that end, we use previously published genotypic data of 21 Pinus pinaster populations, including distinct sampling schemes, together with a widely used method based on linkage disequilibrium patterns observed in a single (one-time) population sample. We hypothesize that spatially restricted sampling might be the main putative factor behind the apparently low N e /N estimates obtained for the large and continuously-distributed populations studied here. Because of its statistically-inherent assessment difficulty, we call for caution when interpreting marker-based estimates of contemporary N e for monitoring widely-distributed tree populations or small genetic conservation units embedded within large continuous tree populations.