Discussion

In terms of physical structures, the study site is characterized with strong seasonality. The deepening of MLD (Fig. 2.3B) in the cold seasons apparently served as a regular source of inorganic nutrients for plankton growth in the epilimnion. The extremely high N/P ratios (1,432±685 molN molP^-1; Fig. 2.4C) recorded in the epilimnion indicted that many plankton (phytoplankton and bacterioplankton) were subjected to P-deficiency, especially during the warm and stratified seasons, as revealed by higher readings of bulk APADIA and specific APADIA (Fig. 2.6C). However, both enzymatic readings fluctuated greatly during the stratified seasons, implying that episodic event (i.e. typhoon) might also affect the SRP concentrations at the dam-site (Fig. 2B), and thus the behaviors of bulk APADIA and specific APADIA (more discussion below).

Bulk APA is an enzymatic reaction. Its expression in the field is subjected to physical, chemical (i.e. substrate availability), and biological regulations. Table 2.1 indicated that the values of bulk APA in this system as a whole could be affected by the changes of physical (temperature and light intensity), chemical (SRP concentrations and availability, i.e. MLD), and biological (Chl a, the abundances of picocyanobacteria and bacteria) parameters. Multiple linear regression analysis (Table 2.2) indicated that MLD (Beta = -0.41) seemed affected the total variations (seasonal and inter-annual) of bulk APADIA more than the biomass of CYA (Beta = +0.30). Overall, it suggested that the changes of phosphate availability (as inferred from MLD) and pico-phytoplankton biomass were the two important factors in determining the seasonal and inter-annual variability of bulk APA in the epilimnion. After biomass normalization, MLD still was one of the essential factors responsible for the variation of specific APADIA.

Based on the negative correlation between phosphate concentrations and APA, many field and enclosure studies have concluded that phosphate supply could be one of the most important factors regulating APA (Chrost & Overbeck 1987, Siuda & Chrost 1987, Istvanovics et al. 1992, Zhou & Zhou 1997, Nausch 1998, Labry et al. 2005). High phosphate concentrations often repressed the synthesis rate of APase (Perry & Eppley 1981, Jamet et al. 1997, Kruskopf & Du Plessis 2004, Labry et al. 2005, Kim et al. 2007, Cao et al. 2010). An “induction-repression” mechanism of phosphate availability on APA has been proposed by Jansson et al (1988a). This study verified the negative relationship of SRPDIA (consider as phosphate) concentration on bulk APADIA and specific APADIA (Table 2.1). However, it is further identified that the “availability” of phosphate that is the changes of the mixed layer depth, is more appropriate and representative than “concentration” itself in explaining the variations of bulk and specific APA (Table 2.2). Several studies suggested that the stoichiometry of inorganic nutrient (i.e. N/P ratios) might affect the expression of APA in the field. For instance, Petterson (1985) found that specific APA in oligotrophic Lake Erken (with dissolved inorganic N/P ratio varied from >1,200:1 in Apr to 8:1 in Sep) increased 10 fold during P-limited period (May~Jun) but decreased to undetectable during N-limited season (Sep). A identical phenomenon was also be found in Chesapeake Bay (Fisher et al.

1992). However, the results of analyses of correlation (Table 2.1) and multiple linear regression analysis (Table 2.2) indicated that this is not the case for the study site.

Potential reason might be that DIN concentrations (14~87 μMN) and N/P ratios (1,432±685 mol N mol P^-1) were too high in this system, so that N-limitation could never occurred.

In addition to physical mixing processes and limiting-mineral availability, light intensity, through its effects on autotrophs, could be also important in regulating bulk

and specific APA. In fact, surface light intensity ranked 4^th among the five most suitable variables for bulk APA, and ranked 1^st among the four most suitable variables for specific APA (Table 2.2). Intuitively, light may enhance autotrophs’ C-fixation rate and results in a higher demand of non-carbon materials (e.g. phosphate) simultaneously. An elevation of APA under higher light intensities eventually would be expected. It is well known that the physiological responses of heterotrophs including bacteria, are light-independent. Light might still affect bacterial APA indirectly because of the mineral-competition between picocyanobacteria and heterotrophic bacteria (i.e.

osmotrophs) in many mineral-limited environments (Thingstad et al. 1993). In another word, it is suspected that light might have an additive (or even multiplicative) effect on either bulk or specific APA. Light effect on osmotrophs’ APA behaviors will be specifically examined in Chapter 5.

Typhoon is a summer-to-autumn episodic event in the northern Hemisphere. On average, more than 20 typhoons were formed in the tropical Pacific Ocean each year, and 6~7 of them passed through Taiwan (data source, Taiwan Central Weather Bureau, www.tcwb.gov.tw). During typhoon events, free phosphate and particle-attached phosphate would be transported from up-stream and the tributaries down to the study site by hyper-pycnal flow formed at the depths of 40~80 m (Chen et al. 2006). The magnitude of the sub-surface SRP maximum (Fig. 2.2B) formed in the mid-waters reflected the strength of typhoon, and served important phosphate source for plankton grown in the epilimnion (Tseng et al. 2010). The strength of summer typhoons and thus phosphate supply could affect the seasonal trend of bulk APA. This was justified by the results of Fig. 2.8 indicating that the temperature responses (i.e. the slopes) of bulk APADIA were quite different among the four sampling years. This implies that the potential impact of episodic events (typhoon and extreme precipitation) can’t be ignored

especially for systems located at typhoon prevailing areas. Recent studies indicated that the intensity (and frequency) of strong typhoon (Chan & Liu 2004, Webster et al. 2005, Wu et al. 2005) and extreme precipitation (Alexander et al. 2006, Kwon et al. 2007) might be enhanced under warming climate. Based on this line of reasoning, many mineral-limiting freshwater ecosystems in sub-tropical to tropical areas might become less P-deficit for plankton growth in summer.

In this system, DOP contributed >90% of total phosphate (TP = SRP +DOP; Fig.

2.5). DOP could serve as an additional source of P for plankton growth. The turn-over times of TP (=TP inventories/APA) estimated to be in the range of 0.2~11.8 d^-1, which were within the range (12~24 d^-1) reported by Labry et al (2005). The reported values of bulk APA (1~95 nM h^-1) of this study were comparable to the oligotrophic ecosystems (Table 2.5), such as the Red Sea (40~150 nM h^-1) and the Baltic Sea (40~160 nM h^-1).

Since bulk APA is a function of living biomass and specific APA (Bulk APA = biomass x specific APA), specific APA has been recognized as a better indicator for P-deficiency of plankton. In this study, specific APA was derived from the normalization of bulk APA by the biomasses of phytoplankton and bacteria, despite of the fact that all plankton groups or cells respond equally to P-stress (Rengefors et al. 2003, Lomas et al.

2004). Healey & Hendzel (1979), Pettersson (1980, 1985) and Gage & Gorham (1985) defined that a system would be in a status of “critical” and “severe” P-deficiency when the observed specific APA were in the range of 40~250 nmol mgC^-1 h^-1 and >250 nmol mgC^-1 h^-1, respectively (Table 2.6). In this system, average values of the specific APA during the cold-mixing seasons were 158±138 nmol mgC^-1 h^-1, implied that plankton were critically P-deficient. In warm-stratified seasons, plankton were facing severe P-deficiency since the averaged specific APA reached 391±207 nmol mgC^-1 h^-1.