are constants. Flushing times de- 

 termined from the full solution (t , 

 time to 90% of difference between 

 initial and asymptotic total amount 

 of Chlorophyll a in the estuary) were 

 found to be close to -In (0.1)/ \ 

 which is the 90 percent decay time 

 for A , the smallest magnitude 

 eigenvalue of A Solutions for speci- 

 fic boundary conditions were about \ 

 day less than t and in only 2 out of 

 25 cases did the difference slightly 

 exceed one day. Using this simpli- 

 fication, response times averaged 2.8 

 d during high flow and 5.3 d during 

 low flow. The minimum (0.8 d) oc- 

 curred during 1977 peak flow and the 

 maximum (6.8 d) during 1977 low flow. 

 Because of the relative stability of 

 estuarine net circulation with re- 

 spect Q f variations (Figure 4) , 

 flushing times remain short even 

 during low flow periods (c.f. Ketchum 

 1967). A well-mixed lower Hudson 

 Estuary (i.e. Q f <<< tidal flows) 

 would not be as stable. Mean water 

 residence times (=Estuary volume/Q f ) 

 would be longer, increasing from 

 7.6 d (high flow) to 26.0 d (low 

 flow). 



CHLOROPHYLL a FLUXES 



In light of the possible effects 

 of variability of Chlorophyll a on 

 scales of a tidal cycle (discussed 

 above) to several days (discussed 

 below) , we chose to examine the mean 

 Chlorophyll a fluxes during high and 

 low flow periods (roughly winter to 

 early-spring and late spring to 

 summer). Though this does not 

 resolve short term variations, such 

 as the occurrence and fate of parti- 

 cular phytoplankton blooms, a fairly 

 reliable picture emerges of Chloro- 

 phyll a fluxes in terms of the sea- 

 sonal variation of flow regime and 

 biomass of phytoplankton size frac- 

 tions (Malone 1977, Malone et al. 

 1980). 



Input fluxes to the estuary 

 reflected the occurrence of net- 

 plankton blooms in adjacent coastal 

 waters and advection into the estuary 

 under high flow and growth of nanno- 

 plankton within the estuary under low 

 flow conditions (Malone 1977, Malone 

 et al. 1980). During high flow the 

 main source of Chlorophyll a was at 

 the mouth of the estuary (MP-7) in 

 the lower layer (Figure 4), where 

 bottle samples had a mean of 64 

 percent Chllorophyll a in the net- 

 plankton fraction (range 49-90%). 

 The most important input during low 

 flow was the upstream boundary upper 

 layer where nannoplankton accounted 

 for 92 percent of Chlorophyll a on 

 the average (range 82-97%). A small- 

 er input of Chlorophyll a occurred 

 at MP -J (12 x 10 mg Chi a d vs. 

 35 x 10 mg Chi a d at MP 18), and, 

 unlike the high flow period, was not 

 dominated by net-plankton (mean % net 

 31%, range 2% to 84%). 



Fluxes within the estuary showed 

 that Chlorophyll a propagated longi- 

 tudinally and vertically away from 

 its sources (Figure 4). Chlorophyll 

 a was advected upstream during high 

 flow by a consistently larger lower 

 layer influx than downstream upper 

 layer flux. Vertical advection and 

 exchange transported Chlorophyll a 

 from the lower to upper layers. Thus 

 the flux of Chlorophyll a during high 

 flow followed the longitudinal and 

 vertical flux into the lower layer, a 

 consequence of vertical exchange 

 fluxes from sources in the upper 

 layer. South of MP net vertical 

 fluxes into the upper layer reflected 

 the lower layer input at MP -7. 



The high flow pattern of large 

 sinks in the upper layer and smaller 

 source fluxes in the lower layer 

 (except between MP and MP -3) is 

 best explained by the sinking of 

 phytoplankton within the estuary. An 

 average sinking rate of 3.5 m d 



179 



