Marine benthic invertebrate populations found in estuarine or coastal habitats often exchange larvae. However, the dynamics of larval exchange are poorly understood because of difficulties in (1) making synoptic assessments of horizontal and vertical larval distribution patterns over large areas for extended periods of time and (2) determining the origins of field-sampled larvae. This study examines how temporal changes in the vertical and horizontal distribution of crab larvae (i.e., Pachygrapsus crassipes and Lophopanopeus spp.) affect larval transport. Larval concentration and water velocity data were collected concurrently and were used to estimate larval exchange between regions of San Diego Bay (SDB) and between SDB and nearshore coastal waters. A larval fingerprinting technique was used to distinguish SDB and non-SDB spawned, stage I P. crassipes zoeae to quantify larval exchange between SDB and nearshore coastal waters. First order estimates of larval exchange over a tidal cycle between inner and outer regions of SDB and between the bay and nearshore coastal habitats corroborate a net transport of stage I P. crassipes zoeae from SDB as inferred from larval behavior. The estimated net larval exchange of Lophopanopeus spp. zoeae was into SDB, suggesting retention within the bay through larval development. Trace elemental fingerprinting of stage I P. crassipes zoeae revealed bi-directional exchange between SDB and the nearshore coastal environment when the predominant transport predicted from zoeal swimming behavior was out of San Diego Bay. Approximately 5% of stage I P. crassipes zoeae sampled in the mid region of SDB originated from outside SDB, while 26% of zoeae sampled at the entrance originated from outside SDB. Combined use of trace elemental fingerprinting and synoptic field sampling techniques will help improve our understanding of larval transport and ultimately the population dynamics of nearshore species.