Adding nitrate and phosphate separately or together in the Central Indian Ocean: a nutrient enrichment experiment

Nutrient enrichment experiments were carried out in the Central Indian Ocean during the Chinese First Around-the world Research Cruise, adding nitrate, phosphate, or a mixture of both of them to surface seawater. The concentration of nitrate, nitrite, ammonia, and phosphate were analyzed spectrophotometrically, the chlorophyll-a con5 centration with fluorescence analysis, and the temperature variation during the experiment recorded. Addition of nitrate resulted in rapid growth of phytoplankton concomitant with depletion of nitrate in the water samples. No apparent variation occurred in chlorophyll-a concentration when phosphate was added. Combining nitrate and phosphate proved to be best to promote phytoplankton bloom, and nitrate was depleted 10 prior to phosphate. After nitrate was consumed, a substantial amount of phytoplankton survived on the supplied phosphate. No correlation was found between the nitrate to phosphate ratio and chlorophyll-a or phytoplankton growth rate. We also found no correlation between water temperature and chlorophyll-a or phytoplankton growth rate. We conclude that neither nitrate to phosphate ratio nor water temperature control the 15 growth of phytoplankton.


Introduction
A wide variety of nutrients are essential for phytoplankton growth in the oceans, including macronutrients such as nitrogen, phosphorus and silicon, and micronutrients such as iron and zinc. Nutrients limitation of phytoplankton has been reported extensively 20 in different sea areas. Among those nutrients, nitrogen and phosphorus play a particularly important role in limiting biological productivity as evidenced by their often near complete exhaustion in surface waters (Gruber, 2004). The question arises if nitrate or phosphate or both are the primary nutrients controlling phytoplankton production? Different answers have been given in nutrient enrichment experiments depending on the phytoplankton growth in the oligotrophic tropical Pacific Ocean (Thomas, 1966(Thomas, , 1967(Thomas, , 1969(Thomas, , 1970, South Pacific subtropical region Dufour and Berland, 1999), South China Sea (Chen et al., 2004), Kaneohe Bay of Hawaii (Larned, 1998) and Cape Bolinao of NW Philippines (Terrados, 1999). However, phosphorus limited growth occurs in the northwest Mediterraneann (Thingstad et al., 1998), the East China 5 Sea (Wong et al., 1998), and Bohai Sea (Zou and Zhang, 2001). Nitrogen and phosphorus may all well limit phytoplankton production in Daya Bay (Wang et al., 2007;Zhu et al., 2008), Taiwan Strait (Wang et al., 2008), and specific areas in the Yellow and East China Sea (Liu et al., 2004).
Nitrogen or phosphorus control on phytoplankton growth varies per sea. In the In-10 dian Ocean, there has been some research on the nutrient limitation of phytoplankton. Phytoplankton growth could be stimulated by the addition of either NH + 4 or NO − 3 and co-limited by both Fe and macronutrients in the central area of the northwestern Indian Ocean (Takeda et al., 1995). NH + 4 is also the major nitrogenous nutrient used by phytoplankton in the western Indian Ocean (Mengesha et al., 1999). To in- 15 vestigate phytoplankton nutrient control in the Central Indian Ocean, we carried out a nutrient enrichment experiment by adding nitrate, phosphate, or both to surface sea water. Water temperature, the concentration of nitrate, nitrite, ammonia, and phosphate, and chlorophyll-a concentration were monitored during the experiment. The influence of water temperature, nutrient depletion, and N:P on phytoplankton growth 20 was presented.

Materials and methods
Our nutrient enrichment experiment was conducted in the central Indian Ocean during the Chinese First Around-the world Research Cruise in December 2005. The cruise ran along the Mid-Indian ridge to explore hydrothermal vents. About 12 full-depth CTD and 25 12 multi-sampler stations were completed in the Indian Ocean. Data were collected both by continuously recording instruments such as CTD and deep-towed devices, and 2651 individual sample analysis for nutrient, chlorophyll-a, CH 4 , biodiversity, etc. The primary productivity was influenced by the prevailing northeast monsoon at that time. Surface seawater for nutrient enrichment experiment was collected from 50.17 • E, 37.81 • S.
The experimental facility set on the bow consists of four 500-L barrels each with a diameter of ca. 100 cm, a set of cooler and a temperature controller (Tang et al., 5 2009). The side wall of acrylic barrels is transparent, ensuring enough light into the barrels. Cooling water from the cooler could cycle within the rubber tubes, which are circled through each barrel. Thus, The cooler and temperature controller could adjust the experimental temperature close to that of the actual surface seawater.
All sampling and incubator equipment was washed three times with surface seawater  Table 1. Seawater in B1 was used as background without nutrient addition. Nitrate, phosphate or mixtures of them were added separately into the other three barrels (see in Table 2) as inorganic salts: KNO 3 , FeSO 4 ·7H 2 O, and 15 KH 2 PO 4 . Next the sea water in the barrels was stirred with a glass rod and left to stay for 17 d. The sea water was sampled every 12 or 24 h to analyze for nutrients spectrophotometrically, and for chlorophyll-a with a fluorescence method. The temperature of each of the barrels was measured when sampling.
3 Results and discussion 20

Phytoplankton growth
Chlorophyll-a concentration is an important parameter in reflecting phytoplankton growth. The highest average concentration of chlorophyll-a (Chl ave ) was present in B4 with the addition of nitrate and phosphate, whereas Chl ave in the background barrel (B1) approximated that of B3 with the addition of phosphate, with a value below of that maximum of chlorophyll-a concentration as a function of time is defined as the blooming spot of phytoplankton (BSP). BSP occurred on the 8th day of the experiment for B2 and B4 (see in Fig. 1a). The increase of chlorophyll-a concentration at BSP (Chl BSP ) in B4 was the most pronounced, being 2.56 mgm −3 higher than the initial value (Chl ini ).
Chl BSP in P -addition barrel (B2) increased 0.92 mgm −3 . However, no apparent increase 5 in chlorophyll-a concentration appeared in B1 and B3, i.e. the background barrel and the one to which only phosphate was added. Clearly, phosphate addition has little influence on phytoplankton growth. Nitrate addition strongly promoted phytoplankton bloom, which was also proved by the nutrient enrichment experiment in the northwestern Indian Ocean (Takeda et al., 1995). The pronounced maximum in B4 indicates that 10 addition of nitrate together with phosphate proved to be best in promoting phytoplankton growth. In addition, the average concentration of chlorophyll-a after BSP (Chl aBSP ) was much higher than Chl ini in both B2 and B4 (see in Fig. 1a and Table 3), indicating that a substantial concentration of phytoplankton managed to survive. The abundance of phytoplankton after BSP in B4 was higher than that in B2. 15 The phytoplankton growth rate (R) as a function of time is calculated as follows: where t is the incubation time and Chl t is the concentration of chlorophyll-a at time t. R in the four barrels is shown in Fig. 1b and Table 3. The average value of R (R ave ) was highest in B4. R ave in B1 and B3 was similar but much lower than that in B2 and B4. 20 The value of R in the four barrels increased rapidly and similarly at the beginning of the experiment, but decreased rapidly after the initial pulse in the background barrel (B1) and P-addition barrel (B3). Maxima in R appeared at BSP in B2 and B4, corresponding with a maximum in chlorophyll-a concentration. These results suggest that adding nitrate results in rapid growth of phytoplankton, whereas adding phosphate did not. 25 R BSP in B4 was 0.12 d −1 higher than that in B2. We conclude that adding nitrate and phosphate enhanced phytoplankton growth relative to adding nitrate only.

The influence of water temperature on phytoplankton growth
The average water temperature was ca. 21.5 • C during our experiment. The temperature variation is shown in Fig. 2. The temperature increased from 18 • C on the 4th day to 23.5 • C on the10th day: the most pronounced temperature trend in our experiment. Later on the water temperature decreased from 23 • C to 22 • C with an average of 5 22.5 • C. The most rapid increase in chlorophyll-a concentration and in R in B2 and B4 all occurred from the 6th to 8th day, which coincided with the period in which temperature increased. This indicates that increasing water temperature has some effect on phytoplankton growth. Previous studies (Eppley, 1972;Goldman et al., 1974;Yoder, 1979) have proved that temperature plays an important role in the growth of diatoms. 10 However, the community sensitivity to temperature may be far smaller than the species sensitivity because of the adaptation of different organisms to a specific temperature range (Sarmiento and Gruber, 2006). As shown in Table 4, there was no significant correlation between water temperature and chlorophyll-a concentration in the four barrels. Water temperature was negatively correlated with R in B1, B2, and B3, whereas no 15 correlation showed up in B4. We conclude that, at least in our experiment, temperature does not dominate the growth of phytoplankton community. Fig. 3 and Table 3. [NO 3 ] in B1 (no addition of nutrient) and B3 (P ad-20 dition) was low and show no obvious change trends, whereas that in B2 (N addition) and B4 (N and P addition) went in general down. Within the first 9 days, [NO 3 ] was reduced from 11.55 µM to 6.86 µM in B2 and from 12.63 µM to 0.35 µM in B4, concomitant with an increase in chlorophyll-a concentration. Especially from the 7th to 9th day in our experiment, the drop in [NO 3 ] from 8.46 µM to 0.35 µM in B4 occurred stantial amounts of phytoplankton. We conclude from these results that nitrate plays an important role in phytoplankton growth.

The variation in concentration of nitrate ([NO 3 ]) and phosphate ([PO 4 ]) in the barrels is shown in
The average of [PO 4 ] ([PO 4 ] ave ) in B1 was similar to that in B2, and they all showed a flat concentration profile with time. No obvious depletion of phosphate in B3 occurred with low nitrate and low chlorophyll-a levels (see Table 3). We infer that adding only 5 phosphate does not enhance growth of phytoplankton. In B4 with the addition of nitrate and phosphate, [PO 4 ] went down after the 6th day. It went down to 2.06 µM on the 9th day during phytoplankton bloom concomitant with a rapid decrease of [NO 3 ] to 0.35 µM. This proved that nitrate is used up prior to phosphate. [PO 4 ] continued to decrease subsequently to 1.36 µM at the end of the experiment. After BSP, a substantial amount 10 of phytoplankton was present in B4 (see in Fig. 1a) whereas [NO 3 ] was close to that in B1 (see in Fig. 3a). Our inference is that phytoplankton continued to live on phosphate.
R in all four barrels increased rapidly during the first day of our experiment (see in Fig. 1b). Neither [NO 3 ] nor [PO 4 ] showed significant and uniform variation (see in Fig. 3). This increase of R has little relationship with the variation of nutrient concentra- 15 tion in the experiment, as is also proved by the poor correlations (P >0.05) between R and [NO 3 ] or [PO 4 ]. When R in B4 went up from 0.28 d −1 on the 6th to 0.46 d −1 on the 8th, [NO 3 ] decreased from 8.69 µM to 3.74 µM, and [PO 4 ] from 2.65 µM to 2.27 µM. In B2, R increased 0.07 d −1 from the 6th to 8th day with slight reduction of 0.31 µM in [NO 3 ]. We infer that the increase of R around BSP is mainly caused by phytoplankton 20 bloom with the ingestion of nitrate, phosphate, or both.

The influence of N:P to phytoplankton growth
N:P represented the ratio of dissolved inorganic nitrogen (DIN: NO 3 +NO 2 +NH 3 ) to dissolved inorganic phosphorus (DIP: PO 4 ), shown in Fig. 4 and Table 3. Note the difference in scale for this ratio in Fig. 4. The average was highest in B2 and the low-25 est in B3. Previous studies (John and Flymn, 2000;Plinski and Jozwiak, 1999) have indicated that N:P ratio was an important factor to influence the growth of phytoplankton species. Because the optimal N:P promoting phytoplankton growth varied among 2655 different species, it is thought that N:P had much more impact on species than phytoplankton community. In B1 with low nitrate and low phosphate, N:P ranged from 3.61 to 12.24 without clear change trend, and no obvious phytoplankton bloom occurred. In B2 with high nitrate, a maximum in N:P appeared on the 8th day when phytoplankton was blooming. In B3 with high phosphate, the N:P trend line showed an obvious peak on the 7th day, while the chlorophyll-a concentration remained stable. In B4 with high nitrate and high phosphate, N:P reduced rapidly when phytoplankton was blooming. Thus, N:P in our experiment did not control the growth of phytoplankton community, as confirmed by the poor correlations between N:P and chlorophyll-a concentration or R (see in Table 4).

Conclusions
We conducted nutrient enrichment experiments by adding nitrate, phosphate or a mixture of them to surface seawater in the Central Indian Ocean. Temperature, chlorophylla concentration, and nutrient concentration were investigated. Several conclusions may be drawn from our experiment to wit: 15 i) The addition of nitrate stimulates phytoplankton bloom, but phosphate addition does not. Adding nitrate and phosphate together proved to be the best in promoting phytoplankton growth.
ii) With the addition of nitrate and phosphate, phytoplankton growth uses them in tandem. Nitrate is depleted prior to phosphate. After nitrate depletion, phytoplankton 20 continues to grow using phosphate.
iii) Neither temperature nor N:P controls the growth of phytoplankton community.   2661   Table 4. Result of correlation analysis between temperature and chl-a, temperature and R, N:P and chl-a, and N:P and R. T is water temperature; C is correlation coefficient; P is significance level; chl-a represents the concentration of chlorophyll-a.