The oceans’ biological pump plays a major role in the carbon cycle by transferring 5-10 Pg carbon annually out of the upper ocean. A relationship formulated >30 years ago – the ‘Martin curve’ – continues to be used to parameterize particle flux attenuation for the global ocean in models. Despite the almost universal adoption of this empirically-derived relationship, we have made little progress in understanding the drivers of C flux attenuation in over three decades.  The fundamental problem with obtaining a mechanistic understanding of the C flux attenuation curve is that it represents multiple processes that concurrently attenuate particle flux (zooplankton flux feeding, diurnal vertical migration, and microbial remineralization of settling particles). Disentangling these processes is further confounded by the likelihood that each process will have different patterns with depth, that may also vary regionally. Hence the way ahead to “deconstructing the Martin curve” is to separate these processes and their drivers. We present a novel in situ approach (called C-RESPIRE) that first intercepts sinking particles (covered in attached microbes) in the upper Twilight Zone (~100-300 m depth), immediately isolates the particles from grazers, and then incubates them at depth during the same deployment. This approach, across multiple deployment depths, provides vertical profiles of rates of microbial remineralization of particle flux.