Spatial Characterization of Water Quality in Florida Bay and Whitewater y Multivariate Analyses: Zones of Similar Influence

We apply an objective statistical analysis to a 6-yr, multiparameter dataset iti an c4l’ort lo tlctic~ibe the spatial dependence and inherent variation of water quality patterns in the lilorida Ray-Whitrwalc*r Hny area. Principal component analysis of 16 water quality parameters collected monthly over a 6-yr period resnlted in live prinr.ipaI COIIIponents (PC) that explained 71.8% of the variance of the original variables. The “organic” coIIIpoIIvIl1 (I’( :,) wils Clllllposed of TN, TON, APA, and TOC; the “inorganic N” component (PCIl) cuntainecl NO2 , NO:, , m~cl NII,‘, the “phytoplankton” component (PC,,,) was made up of turbidity, TP, and Chl a; DO and f~~mprra~urc~ wcrv i~~vvrtiely related (PC,); and salinity was the only parameter included in PC,. A cluster analysis of mean and SI) ol’ I’(: tswrt~w adted in the spatial aggregation of 50 fixed monitoring stations in Florida Bay and Whitewater Ilay into nix ZOIW~~ 01’ similar influence (ZSI) defined as Eastern Florida Bay. Core Florida Bay, Western Florida Bay, (:oot Ilay, 1111~ IIUIW Mu~~yrovc Fringe, and the Outer Mangrove Fringe. Marked differences in physical, chemical, snd biological charnrtcrlwtics awmg ZSI were ill-ustrated by this technique. Comparison of medians and variability of param-tcr vahan ~JNIWI~ %SI allowed large-scale generalizations as to underlying differences in water quality in these regions. Il’or ~xa~nph+ Kantc:rtl IGrida Bay had lower salinity7 TON, TOC, TP, and Chl a than the Core Bay as a function ol’ dil‘l‘cwwc:eri ~II I‘rc~nl~wa~cv inpuls and water residence time. Comparison of medians and variability within ZSI resulted in new I~ypotl~m~~s tm 10 lhe processes generating these internal patterns. For example, the Core Bay had very high TON, ‘I’()(:, a11r1 NI f., ’ ~~llll~l!ll~~d~~~~ilS but very low NO,~ 9 leadiig us to postulate the inhibition of nitrification via CO production by ‘I’()( i ph~hlysia. We believe that this simple, objective approach to spatial analysis of fixed-station monitoring data~h will arid twlcn~i~bi and managers in the interpretation of factors underlying the observed parameter dislribulion patterns. We alrc~ mprcl tbal this approach will be useful in focussing attention on specific spatial areas of concern and in gvnvraring IWW ideas COI hypothesis testing.