Anatomically segregated apical and basal dendrites of pyramidal neurons receive functionally distinct inputs, but it is unknown if this results in compartment-level functional diversity. Hippocampal place cells exhibit representational drift (Ziv et al., 2013), but this has only been characterized with somatic recordings. While dendritic activity is implicated in the formation of stable place fields and shaping hippocampal activity, in vivo studies of hippocampal dendritic activity have largely been confined to basal dendrites (Sheffield and Dombeck, 2015) or small cross-sections of apical dendrites (Cichon and Gan, 2015, Otor et al., 2022). Here we imaged calcium signals from apical dendrites, soma, and basal dendrites of large populations of pyramidal neurons in area CA3 of mouse hippocampus during head-fixed navigation. This visually stunning preparation allows simultaneous imaging of the entire dendritic tree of CA3 pyramidal neurons, offering an unparalleled opportunity to study spatial and contextual coding properties in dendrites. However, existing segmentation tools (Pachitariu et al., 2016, Giovannucci et al., 2019) are not suitable for extraction of densely labeled dendrites. Hence, we developed computational techniques in the same family as constrained non-negative matrix factorization (CNMF) (Pnevmatikakis et al., 2016, Giovannucci et al., 2019), with improved initialization and refinement steps to identify dendrites in highly overlapping preparations. We name the collection of tools “dendritic non-negative matrix factorization,” (d-NMF), but note that the technique is morphology agnostic and works with dendritic, somatic, or axonal regions of interest. Using these tools, we identified robust spatial tuning in apical and basal dendrites, similar to soma, but with higher activity rates. Across days, apical dendrites’ representations of space were more stable than basal dendrites, resulting in better decoding of the animal’s position. Thus we demonstrate different rates of representational drift at a sub-cellular level, potentially reflecting functionally distinct input streams leading to different dendritic computations in CA3.