Saeed Roshan
University of Miami
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Saeed Roshan.
Global Biogeochemical Cycles | 2015
Saeed Roshan; Jingfeng Wu
Dissolved zinc (dZn) concentration was determined in the North Atlantic during the U.S. GEOTRACES 2010 and 2011 cruise (GOETRACES GA03). A relatively poor linear correlation (R2 = 0.756) was observed between dZn and silicic acid (Si), the slope of which was 0.0577 nM/µmol/kg. We attribute the relatively poor dZn-Si correlation to the following processes: (a) differential regeneration of zinc relative to silicic acid, (b) mixing of multiple water masses that have different Zn/Si, and (c) zinc sources such as sedimentary or hydrothermal. To quantitatively distinguish these possibilities, we use the results of Optimum Multi-Parameter Water Mass Analysis by Jenkins et al. (2015) to model the zinc distribution below 500 m. We hypothesized two scenarios: conservative mixing and regenerative mixing. The first scenario (conservative) could be modeled to results in a correlation with observations with a R2 = 0.846. In the second scenario, we took a Si-related regeneration into account, which could model the observations with a R2 = 0.867. Through this regenerative mixing scenario, we estimated a Zn/Si = 0.0548 nM/µmol/kg that may be more realistic than linear regression slope due to accounting for process b. However, this did not improve the model substantially (R2 = 0.867 versus0.846), which may indicate the insignificant effect of remineralization on the zinc distribution in this region. The relative weakness in the model-observation correlation (R2~0.85 for both scenarios) implies that processes (a) and (c) may be plausible. Furthermore, dZn in the upper 500 m exhibited a very poor correlation with apparent oxygen utilization, suggesting a minimal role for the organic matter-associated remineralization process.
Nature Communications | 2017
Saeed Roshan; Tim DeVries
Biologically fixed carbon is transferred from the surface to deep ocean as sinking particles or dissolved organic carbon (DOC). DOC is estimated to account for ~20% of global export production, but the degree to which this varies regionally has not been assessed at a global scale. Here we present the first observationally based global-scale assessment of DOC production and export, obtained by combining an artificial neural network estimate of the global DOC distribution, and a data-constrained ocean circulation model. Our results demonstrate that the efficiency of DOC production and export varies more than threefold across oceanographic regions. DOC production and export display a pronounced peak in the oligotrophic subtropical oceans, where DOC accounts for roughly half of the total organic carbon export. These stratified nutrient-depleted regions are expected to expand with future warming, amplifying the role of DOC in the biological pump, and magnifying the need to improve DOC cycling in climate models.The degree of regional variability in marine dissolved organic carbon (DOC) export production is poorly constrained on a global scale. Here, the authors combine an artificial neural network and a data-constrained ocean circulation model to show that the efficiency of DOC export varies 3-fold across regions.
Global Biogeochemical Cycles | 2015
Saeed Roshan; Jingfeng Wu
Cadmium (Cd) incorporated into the benthic microfossils has been widely used in reconstructing the past water circulation in the North Atlantic. This requires a major control by conservative mixing over regeneration process on the seawater cadmium distribution in the North Atlantic. Through coupling the recently reported Cd data at depths below 300 m with quantitative water mass analysis along the GA03 transect, we tested two models for cadmium cycling within the North Atlantic: conservative mixing alone and conservative mixing plus regeneration which is termed as regenerative mixing. The results show that the regenerative mixing model reproduces the observations (slope = 0.99 and R2 = 0.97) much better than the conservative mixing model (slope = 0.99 and R2 = 0.88). The regenerative mixing model was applied to estimate the amount of dissolved cadmium regenerated “within” the North Atlantic. This regionally regenerated cadmium contributes to ~10% and >50% of the total cadmium in the North Atlantic at depths >1000 m and 300–1000 m, respectively, indicating that the microfossil-reconstructed seawater Cd can be attributed to the mixing from the South Atlantic to the North Atlantic with high accuracy for depths below 1000 m, while within 300–1000 m cadmium is more controlled by the regeneration process than the mixing process, complicating the attribution of the microfossil-reconstructed seawater Cd to the changes in water mass geometry within this 300–1000 m depth range. A regionally and vertically constant Cd/PO43− regeneration ratio ~262 pM/µmol/kg was derived from the regenerative mixing model which is comparable with ratios estimated in the recent studies for the North Atlantic.
Deep-sea Research Part Ii-topical Studies in Oceanography | 2015
Mariko Hatta; Christopher I. Measures; Jingfeng Wu; Saeed Roshan; Jessica N. Fitzsimmons; Peter N. Sedwick; Pete Morton
Deep-sea Research Part Ii-topical Studies in Oceanography | 2015
Jessica N. Fitzsimmons; Gonzalo Carrasco; Jingfeng Wu; Saeed Roshan; Mariko Hatta; Christopher I. Measures; Tim M. Conway; Seth G. John; Edward A. Boyle
Marine Chemistry | 2016
Saeed Roshan; Jingfeng Wu; William J. Jenkins
Marine Chemistry | 2014
Jingfeng Wu; Saeed Roshan; Gedun Chen
Deep-sea Research Part Ii-topical Studies in Oceanography | 2015
Jingfeng Wu; Saeed Roshan
Marine Chemistry | 2015
Saeed Roshan; Jingfeng Wu
Global Biogeochemical Cycles | 2017
Saeed Roshan; Jingfeng Wu; Tim DeVries