Results of nitrate, ammonium and urea uptakes for both the seasons at nine stations are shown in Table 1. Using the average Redfield ratio (C/N) of 6.6, we convert the total nitrogen uptake to carbon uptake; the total productivity in the BOB varied from 90 to 870 mgC m−2 d−1 during Sep–Oct. 2002, with an average value of 316 mg C m−2 d−1. The total productivity during Apr–May 2003 varied from 154 to 975 mg C m−2 d−1 with an average of 575 mg C m−2 d−1. Also, the new production during Apr–May 2003 (overall average ∼433 with coastal average of 552 and open ocean average of 284 mg C m−2 d−1) was higher than that during Sep–Oct 2002 (overall average ∼318 with coastal stations averaging ∼281 and open ocean ∼364 mg C m−2 d−1). These values are comparable to the new production off India, 400 ± 160 mg C m−2 d−1 reported for the Arabian Sea [Sambrotto, 2001]. There is a very significant correlation between new (y) and total production (x): y = 0.88 x − 0.89 (coefficient of determination, r2 = 0.99) in BOB. The relationship remains significant (r2 = 0.96) with a similar slope, when ammonium concentration from zooplankton regeneration is considered for uptake calculation during Sep–Oct 2002 (Figure 2). Data for the Arabian Sea [Watts and Owens, 1999] also show such a correlation, but with some scatter: y = 0.33 x − 0.36 (r2 = 0.86) (Figure 2). However, this relationship becomes somewhat better (y = 0.46 x − 0.43; r2 = 0.93) when conservative estimates (i.e., no ambient ammonium) were considered. During this calculation the average ammonium concentration was taken as 0.13 μM [Woodward et al., 1999]. The significance of this linear relationship lies in the quantification of new production from satellite data on total productivity. In both the seas, the x-intercept, i.e., minimum amount of regenerated production in the total absence of extraneous nitrate supply is ∼1 mmol N m−2 d−1 (equivalently ∼80 mg C m−2 d−1).
 The possible sources of nutrient for higher new production in the Bay could be (i) the supply of nutrients from the adjacent land due to the river discharge [Ittekkot et al., 1991]. However, to quantify the riverine contribution to the nutrient pool of the BOB river borne nutrient data is needed. The nitrogen flux into the BOB was indirectly calculated by Kumar et al.  based on the global flux of fresh water (37.4*1012 m3 yr−1 [Martin and Whitfield, 1983]) and nitrogen (50*1012 gN yr−1 [Duce et al., 1991]) by world rivers to ocean, to be 2.17*1012gN yr−1. If the fresh water from the Indian rivers is assumed to spread over total area of BOB (2.2*1012 m2), the above influx works out to be 2.68 mgN m−2 d−1. (ii) The contribution to the nitrogen pool of BOB through wet deposition of nitrogenous aerosols is reported to be around 1.53 mg N m−2d−1 [Kumar et al., 1996] using an annual precipitation rate of 2 m over BOB [Tomczak and Godfrey, 1994]. Thus, the total (river + atmosphere) external input of nitrogen to the Bay is ∼4.21 mg N m−2 d−1. However, a higher estimate for the external nitrogenous nutrients to BOB was given by Schafer et al. , where the fluvial input (particulate and dissolved) was estimated to be 5.2*1012 gN yr−1, amounting to 6.36 mgN m−2 d−1. Total atmospheric nitrogen input was 6.83 mgN m−2 d−1, the same order of magnitude as the fluvial input; thus the total extraneous nitrogen (fluvial + atmospheric) to the Bay is ∼13.19 mgN m−2 d−1 (∼0.94 mmol N m−2 d−1). Our results for the two seasons show an average new production of 4.7 mmol N m−2 d−1; therefore, the estimated nitrogen inputs from rivers and atmosphere can at best supply only ∼20% of the total required nitrogenous inputs. The rest (80%) of the required nitrogen has to come from the supply of nitrate from deeper waters. These nutrients could be brought to the surface by the cyclones or high speed winds, very frequent in the BOB during post monsoon (Sep–Dec), in varying amounts, depending on location, intensity and residence time. Data from India Meteorological Department show that 25 and 15% of the total cyclones in BOB occur during Sep–Oct and Apr–May respectively [Das, 1995]. The shallow nitracline in BOB [Madhupratap et al., 2003] might have allowed the nitrate to come up, especially when aided by churning due to cyclonic winds in the post monsoon season. Also, the stratification due to freshwater discharge reduces towards the south and is too weak to prevent nutrients from coming up [Vinaychandran et al., 2002]. During this study, wind speed of ∼20 m/s was encountered over the Bay (Qscat wind speed data; http://www.ssmi.com/qscat/qscat_browse.html). The higher average new production at coastal stations during Apr–May may be due to East India Coastal Current, a poleward current along the western boundary of BOB, active at north of 10°N, bringing cooler, more saline water with nutrients to the surface [Shetye et al., 1993]. This western boundary current of anticyclonic subtropical gyre, best developed during March–April, decays only by June, which includes the study period. Also, the typical vertical profiles of nitrate uptake (Figure 3) during Apr–May indicate significant uptake rates below 40 m, where nutrients are available in plenty (average ∼7 μM) along with light (euphotic zone ∼80 m). Such a condition could lead to photosynthetic rates several fold higher than that at sea surface [Pollehne et al., 1993]. A unique possibility of nutrient supply is sub-marine ground water discharge. Dowling et al.  found an average nitrate concentration of 43 μM in the Bengal basin ground water, the main source for sub-marine ground water discharge into BOB. Further studies are required to quantify this.