Underwater Spectral Energy Distribution and Seagrass Depth Limits along an Optical Water Quality Gradient Charles L. Gallegos, W. Judson Kenworthy, Patrick D. Biber, and Bret S. Wolfe ABSTRACT. We measured in situ inherent optical properties and seagrass maximum depth distribution in widely differing optical water types, including turbid green waters of the Indian River Lagoon (IRL, Florida, USA), a mix of turbid and clear waters in Pan- ama, and very clear waters in Belize. We used Hydrolight to model in situ spectral energy distributions and measured lea fabsorbance spectra (Thalassia testudinum )to distinguish between photosynthetically available radiation (PAR) and photosynthetically usable ra- diation (PUR). Attenuation coefficients for PAR and PUR were nearly indistinguishable in Belize and Panama and differed only slightly in the IRL. Grass grew to depths of pen- etration of 33% of PAR in the IRL, 14% in Panama, and approximately 5% in Belize, although we expect the value for Belize is an underestimate because conditions more turbid than are typical were prevailing at the time of the measurements. Corresponding percentages for PUR were 27%, 12%, and 5% for IRL, Panama, and Belize, respectively. These regional differences in light requirements were striking, and less than half of the difference could be attributed to latitudinal variations in incident light. We conclude that factors other than spectral energy distribution that covary with water clarity control site-specific ligh trequirements o fseagrasses .Possibilities include epiphytes and sediment quality. Charl eGLs.allego Ssm, ithsonia Ennvironmental INTRODUCTION Researc hCente r6,4 C7ontee Ws ha rRfoad E,dge- water ,Maryland 21037 ,USA .W .Judson Ken- Seagrasses are important primary producers that play a role in the stability, worth yC,ente fro Croasta Flisherie asn dHabitat nursery function, biogeochemical cycling, and trophodynamics of many coastal Research N, CCO SN, ation aOl cean iacn dAtmo- and estuarine ecosystems and as such are important for sustaining a broad spec- spher iAcdministration 1,0 P1iver Isslan dRoad, trum of organisms (Hemminga and Duarte, 2000). Seagrasses are potentially Beaufort N, orth Carolin a28516 U, SA P. atrick sensitive indicators of declining water quality because of their high light re- D B. iber D, epartmen to fCoasta Sl ciences T, he quirements (11%-37% surface irradiance) compared to those of other aquatic Universi to ySfouther Mn ississipp Gi,u Clfoa sRte- primary producers with much lower light requirements (<1%) (Dennison et searc hLaboratory 7,0 3Eas Bt eac hDrive O, cean al., 1993; Zimmerman, 2003). Seagrass communities have declined in coastal Springs M, ississipp 3i 9564 U, SA B. re tS W. olfe, Departme n oEtfnvironment aSlcienc eU,niver- regions worldwide (Orth et al., 2006), which is usually attributed to reductions sity of Virginia ,Clark Hall ,291 McCormick in water clarity brought about, at least initially, by accelerated eutrophication in Rd C.,harlottesville V,irgini a2290 4U, SA C.orre- the coastal zone (Krause-Jensen et al., 2008). spondi naguth oGCra.:lleg o(gsallegosc@si.edu). Management efforts aimed at preserving and restoring seagrass systems gen- Manuscrip treceived 2 3Jul y2008 a; ccepted 20 erally focus on improving water clarity (Batiuk et al., 2000; Kenworthy and Ap 2ri0l09. Haunert, 1991; Steward and Green, 2007), based on the high light requirements 360 ° SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES of seagrasses and the reduction in light penetration associ- lengths equally. By contrast, measurements of photosyn- ated with eutrophication (Ralph et al., 2007). Deciding thetically usable radiation (i.e., PUR; see Morel, 1978) on the extent of water quality improvements (or limit of weight quanta in proportion to the efficiency with which allowable deterioration) requires more detailed knowl- they are absorbed. There are no sensors for direct mea- edge of the wavelength-specific light requirements of sea- surement of PUR; it must be calculated from the under- grasses. Based on a survey of available literature, Carter water spectrum (measured or modeled) weighted by the et al. (2000) determined that mesohaline and polyhaline relative absorption spectrum of the plant of interest. submerged grass communities in Chesapeake Bay require Using a bio-optical model of light penetration in the a long-term average of 22% of surface irradiance at the mesohaline Chesapeake Bay, Gallegos (1994) determined deep edge of the grass meadow for survival. Gallegos and that the 22% surface PAR requirement for seagrasses oc- Kenworthy (1996) determined a similar requirement for curred at the same depth as the penetration of 16% of sur- mixed beds of Thalassia testudinum, Halodule wrightii, face PUR. The distinction is potentially important because and Syringodium filiforme in the Indian River Lagoon the penetration of PUR is more sensitive to the concentra- (IRL) near Ft. Pierce, Florida. In contrast, Steward et al. tion of phytoplankton chlorophyll (i.e., eutrophication) (2005) found 20% to be near the minimum for the IRL, than is the penetration of PAR, for the reason that phyto- while the average light requirement was 33% of annual plankton chlorophyll absorption selectively removes those incident irradiance, similar to the wide range (24%-37%) same wavelengths most efficiently used in photosynthesis reported for the southern Indian River Lagoon (Kenworthy by seagrass. Thus, by basing light requirements on PUR and Fonseca, 1996). More recently, Duarte et al. (2007) rather than on PAR, we would predict greater restoration analyzed 424 reports of seagrass colonization depths and benefit from chlorophyll reduction, and greater seagrass light attenuation and found generally higher light require- losses from chlorophyll increases, than by light require- ments for plant communities growing in shallow, turbid ments based on PAR (Gallegos, 1994). waters than in clear, deep waters. The authors suggested The objective of this work was to determine whether that large differences in light requirements between shal- the distinction between PAR and PUR requirements low- and deep-growing seagrasses may be partially attrib- could be determined from in situ depth distributions of uted to differences in the quality of light. Seagrasses may seagrass communities. The distinction cannot be drawn grow deeper in clear water because there is more high- from depth distributions at a single site such as Chesa- energy blue light available for photosynthesis, whereas peake Bay or the IRL, because within these systems the in shallow turbid water the shorter blue wavelengths are underwater spectrum is peaked in the green, and thus rapidly attenuated. there is insufficient spectral variability in available light The wavelength specificity of light absorption by to differentiate between depth limits based on PAR com- seagrasses has implications for setting water quality re- pared with PUR. The gradient of optical water quality quirements needed to protect or restore these plants in types across locations of the Smithsonian Marine Science eutrophic waters that are dominated by inefficient green Network, however, offers a potentially ideal scenario wavelengths. The absorption of light by the complement of for making this determination. All three of the domi- pigments (chlorophyll a and chlorophyll b) in seagrasses is nant seagrass species found in the IRL also occur in the highly wavelength selective, with absorption peaks in the tropical waters of Carrie Bow Cay, Belize, and Bocas del blue (centered around 450 nm) and red (centered around Toro, Panama. In optically clear waters, the underwater 670 nm) regions of the visible spectrum, and a broad ab- spectrum peaks in the blue, near an absorption peak of sorption minimum in the green between 500 and 600 nm chlorophyll a or b. In blue water, therefore, PUR pen- (Drake et al., 2003; Zimmerman, 2003). Wavelengths of etrates deeper than PAR, and plants should grow to rela- light that are poorly absorbed by the plant are relatively tively deeper depths in blue tropical waters if PUR rather inefficient at driving photosynthesis (Drake et al., 2003; than PAR is the determining factor. To investigate this Falkowski and Raven, 2007). distinction, we surveyed seagrass distributions and mea- Light requirements of seagrasses that have been de- sured inherent optical properties (IOPs), from which we termined to date (Batiuk et al., 2000; Kenworthy and calculated underwater light spectra at the deep edges of Fonseca, 1996) have been based on photosynthetically grass beds, to test the hypothesis that across the optical available radiation (PAR, 400-700 nm) because of the water quality gradient seagrass would grow to a consis- widespread availability of underwater quantum sensors. tent depth of penetration of PUR but a variable percent- PAR measurements weight quanta of all visible wave- age of PAR. NUMBER 38 e¢ 361 METHODS using a WETLabs ac-9 instrument with a 0.1 m path- STUD SYITES length, equipped with a pressure sensor to measure depth. A Seabird SBE-5T pump provided water flow to the ac-9 Station locations are shown in Figure 1. We occupied and a WETLabs MPAK unit that controlled pump and in- stations in the clear tropical waters off Carrie Bow Cay, struments and logged data. Belize (station Blue Ground Range, BGR), and in Bahia Measured absorption and beam attenuation coef- Almirante, Panama (station STRI [Smithsonian Tropical ficients were corrected for temperature according to the Research Institute]), a station receiving colored-water dis- manufacturer’s protocols. We corrected absorption coeffi- charge from a nearby creek in Panama (station SNO3), and cients for scattering errors (Kirk, 1992) by the Zaneveld et the more eutrophic waters of the Indian River Lagoon, al. (1994) algorithm that subtracts a fraction of measured Florida (ICW194; see Figure 1). Detailed characteristics scattering coefficient from absorption (Equation 1): of these sites are given by Lang (2009) in the Introduction to this volume. t-w (A) a ain (A) ra e(cray (A) i any (A)) (1) OPTIC APLROPERTIES where d;,,() is the scattering-corrected absorption coef- ficient less pure water absorption at wavelength i, a, is We measured in situ profiles of IOPs, the spectral ab- the measured non-water absorption coefficient subject to sorption and beam attenuation coefficients, at nine wave- scattering error, c;,, 1s the measured non-water beam at- lengths (412, 440, 488, 510, 532, 555, 650, and 715 nm) tenuation coefficient, and « is a coefficient that accounts 80°20'0"W 88°10'0"W Kilometers Kilometers OF aie Ain iW l6 27°30'0"N 0 B lGureou Rnadnge 16°50'0"N {Carr Bieo Cway 30°0'0"N 9°20'0"N 10°0'0"N Kilometers 0 250 500 1,000 1,500 90°0'0"W 70°0'0"W 82°20'0"W 82°0'0"W FIGURE 1. Locations of stations in the Indian River Lagoon, Florida (upper left), Belize (upper right), and Panama (lower right). Lower left pane lshows overview of Caribbean. Light gray shading in Belize panel indicate scora rlee hf abitat. 362 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES for overall errors with the reflective tube absorption meter From the simulations of spectral downwelling ir- of the ac-9 that result from a failure to collect all scattered radiance we calculated PAR according to its definition light (Kirk, 1992). In this work we verified the assumption (Equation 3): that non-water absorption at the longest ac-9 wavelength (715 nm) was not measurable in the laboratory (Tzortziou 700 700PAR(z)= { O(d,z)dr= | a (3) et al., 2006). Thus, we calculated ¢ by Equation 2: 400 400 A a,,(715) a where O is the quantum flux, E, is the spectral down- en_ e (5) =a, (75) welling irradiance in energy units, / is Planck’s constant, \ is the wavelength and y = 27c/) is the frequency of light, We measured the absorption spectrum of Thalassia and c is the speed of light in vacuum. PUR was calculated testudinum leaves in an integrating sphere (LICOR 1800- in an analogous manner, weighted by the plant absorption 12S) interfaced to an Ocean Optics USB2000 spectrom- spectrum, measured at the deep edge of the Belize site: eter. A clean segment of leaf was placed on a microscope slide over the opening to the sphere and illuminated with 700PUR(z) = | Q(A,z)aq,(d) dd (4) a fiberoptic microscope light source. Black tape on the 400 slide obscured the portion of the opening not covered by the leaf. Percent transmittance (%T) of the leaf was where G7), (\) is the absorption spectrum of T: testudinum calculated referenced to the slide and tape without a leaf normalized to its peak at 675 nm and to unit sum. For in place. Absorbance was calculated as —In(%T), and comparison of attenuation rates, PAR and PUR were both the spectrum was normalized to the value at the absorp- normalized to their values at the surface. tion peak at 675 nm. Measurements on eight leaves col- lected from the deep edge at the site in Belize were aver- SEAGRA SSSURVEYS aged. Similar measurements made in Panama had similar results. At each sampling site a pair of scuba divers entered the water to visually confirm the seagrass bed (T. testudi- RADIATIV ETRANSFE RMODELING num) deep edge, defined as the visible transition between vegetated and unvegetated bottom. Once the physical To calculate spectral diffuse attenuation coefficients boundaries of the meadow edges were identified under- and underwater light spectra, we used the commercially water, the divers laid out two 10 m long transects par- available radiative transfer model, Hydrolight 4.2, which allel to the edge of the seagrass bed. At 1.0 m intervals is extensively documented by Mobley (1994). User in- along each transect, the divers visually estimated seagrass put consists of specifications for IOPs, boundary condi- cover in a 0.25 m* quadrat using the Braun-Blanquet scale tions, and assumptions on inelastic scattering processes. (1965). The Braun-Blanquet cover abundance scale is a vi- We used the pure-water absorption coefficients of Pope sual assessment technique for estimating the canopy cover. and Fry (1997) and pure-water scattering coefficients for Values are 0.1 = solitary shoot, with small cover; 0.5 = freshwater from Buiteveld et al. (1994). We used in situ few shoots, with small cover; 1 = numerous, but less than estimates of absorption, attenuation, and scattering coef- 5% cover, 2 = 5%-25% cover, 3 = 25%-50% cover, 4 = ficients binned at 0.5 m intervals. Following Tzortziou et 50%-75% cover, and 5 = more than 75% cover. al. (2006), we used the Fournier—Forand scattering-phase At the same location each diver counted the number function, the shape of which was shown by Mobley et al. of seagrass short shoots in either a 0.25 m? or 0.0625 m? (2002) to be well specified by the backscattering ratio. We quadrat, depending on the shoot density. Short shoot omitted inelastic scattering processes because our interest counts were multiplied by the appropriate scaling factor is in downwelling irradiance, and these processes primar- and averaged for the 10 quadrats to obtain an estimate ily affect only calculations of upwelling radiance. For in- of the number of short shoots per square meter. For com- cident irradiance and the distribution of total irradiance parison of deep edge seagrass characteristics, we also sur- between direct and sky irradiance we used the built-in veyed relatively shallow sites at the Blue Ground Range RADTRAN routine for the time, location, and estimate of station in Belize (2.4 m) and the STRI station in Panama approximate cloud cover. (1.8 m). At SNO3 in Panama we only surveyed at the deep NUMBER 38 © 363 edge. Deep edge data for the IRL are from annual sur- by a transition from T. testudinum to unvegetated sedi- veys by the South Florida Water Management District ment. Braun-Blanquet scores ranged from 0 to 3, indicat- (http://my.sfwmd.gov/gisapps/sfwmdxwebdc/dataview ing cover values less than 50%. .asp?query=unq_id=1797). Seagrass depth limits in the IRL at the site where op- tical measurements were made in 2001 were reported as 0.92 m for beds described as continuous and dense, with a RESULTS lower limit of 50% to 60% cover. SEAGRAS SDEPTH LIMITS OPTIC APLROPERTIES At the Blue Ground Range station in Belize, the deep edge of the Thalassia testudinum meadow was located at A wide range of optical properties was observed 10-11 m. The deep edge was a distinct transition from among the four sites (Figure 2a). Based on absorption a sparse cover of T: testudinum to unvegetated, fine car- spectra, Belize had the clearest water while the most tur- bonate mud. Recently germinated seedlings of the small bid water occurred in the IRL. The two sites in Panama opportunistic species Halophila decipiens were observed were intermediate. The rank order of sites was different just outside of the deep edge of the T. testudinum meadow. for scattering coefficients (Figure 2b), with scattering coef- Braun-Blanquet cover values ranged from 0.5 (a few in- ficients at the Panama shallow site (SN03) being the high- dividual short shoots) to 1 (<5%). Thalassia testudinum est and the Panama deep site (STRI) the lowest. short shoot densities ranged from 0 to 48 shoots m7’, averaging 22.4 shoots m. At the shallow Blue Ground ABSORPTIO SNPECTRUM Range transect, T. testudinum Braun-Blanquet scores ranged from 3 to 4, indicating that cover generally varied Normalized absorption by T. testudinum was simi- from 25% to 75%, while densities ranged from 176 to lar to measurements by other investigators (Zimmerman, 416 shoots m7, averaging 310 shoots m~’. At the shal- 2003), having peaks in the red wavelengths (~680 nm), a low station T: testudinum was 14 times more dense than broad maximum at blue wavelengths (400-490 nm), and at the deep edge. No other seagrass species were observed a trough at green wavelengths (~525-625 nm) (Figure 3, at this station. solid line). This spectrum was used to calculate PUR from At the STRI station in Panama we located the deep simulated downwelling spectral irradiance according to edge of the T: testudinum at 8.5 m. The transition edge Equation 4. However, even at the local minimum at 555 of the T: testudinum meadow was distinct; however, there nm, measured absorption was still 37% of the red peak. was considerably more H. decipiens just downslope of the On considering that T: testudinum has no chlorophyll pig- edge than there was at the Blue Ground Station in Belize. ments that absorb green wavelengths (Zimmerman, 2003), Thalassia testudinum short shoot densities ranged from we also constructed a hypothetical photosynthetic action 0 to 56 shoots m~’, averaging 18 shoots m~, similar to spectrum based on chlorophyll absorption alone, consisting the deep edge at the Blue Ground Range Station in Be- of Gaussian curves with peaks at 410, 430, 455, 642, and lize. Braun-Blanquet values ranged from 0 to 1, indicating 680 nm for an alternate calculation of PUR (see Figure 3, that cover was generally less than 5%. We also observed dashed line). The hypothetical action spectrum is expected three quadrats with a relatively sparse cover of Halodule to produce the maximal separation between PAR and PUR, wrightii. At the shallow STRI station (1.8 m), T. testudi- especially in turbid green water, because the trough in the num densities ranged from 160 to 528 shoots m7 with hypothetical chlorophyll absorption spectrum at green an average of 465, 25 times the density at the deep edge wavelengths is much more pronounced compared with the and more dense than the shallow station at Blue Ground measured absorption spectrum, which includes an unquan- Range in Belize. Braun-Blanquet values ranged from 3 to tified contribution by photosynthetic carotenoids. This hy- 4, similar to the shallow station at Blue Ground Range pothetical chlorophyll-based action spectrum serves as a (BGR) in Belize. site-independent sensitivity test for the greatest possible At the SNO3 site in Panama, the deep edge of the T: difference between PAR and PUR for a higher plant. We testudinum bed was located at 2.4 m. Short shoot densi- did not measure absorption spectra in the IRL, so they are ties ranged from 0 to 288 m *, with a mean value of 114. unknown. The hypothetical spectrum allows a comparison The deep edge of the T. testudinum meadow was marked among sites in the absence of measurements at all sites. 364 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES a .Surface Absorption Spectra Ground Range station in Belize, diffuse attenuation coef- ficients for PAR and PURm were indistinguishable, while AOA — BGR, Belize that for PURh was only 7% higher than for PAR (Table 1). 1.8 ‘ ----STRI, Panama i aes arc eal i saie a Musee eo SNO3, Panama The largest differences among the three attenuation coef- a 16 Ne) ieee opal ICW194, Florida ficients occurred at the IRL. The relative differences be- San Water tween attenuation coefficients for PAR and PURm (13%) = 40) and between PAR and PURh (31%) were similar for the [o=O IRL and SNO3 site in Panama, although the absolute coef-o Oje) ficients were smaller at SNO3 (Table 1). 21 The percentages of surface light remaining at the deep 2 edges of the seagrass beds varied widely among the loca- f1o7)xe}2) tions, from about 5% at the Blue Ground Range site in< Belize to about 30% at the IRL (see Table 1). The percent- ages based on PUR were, as expected, lower than those based on PAR, but the differences among sites was still large (Table 1). Because of the extremely large differences Waveleng (tnhm) among sites in the percentage of light at the seagrass bed deep edge, the calculation of PUR did not yield a consis- b .Surface Scattering Spectra tent value across sites. The overall range was, however, somewhat smaller for PUR than for PAR (Table 1). Spec- tra of downwelling irradiance at the deep edges calculated by Hydrolight are shown in Figure 5. The overall frac- tion of surface irradiance remaining at the deep edges at -1 the different locations follows the percentages in Table 1. Qualitative differences in the spectra of light remaining at (m C_o)Secfafitctieernintg Measured - - - - Hypothetical Waveleng (tnhm) absorptionNor(m=()oa>li)zed FIGURE 2 .Surface water absorption spectra (a) and surface water scattering spectra (b )a tsite ssampled in Belize ,Panama ,and Florida. PAR AND PUR PROFILES 400 450 500 550 600 650 700 Waveleng (tnhm) Profiles of normalized downwelling PAR and PUR based on the measured absorption spectrum (PURm) and FIGURE 3 .Normalized absorption spectra used fo rcalculating pho- PUR based on the hypothetical action spectrum (PURh) tosyntheticall yusabl eradiation b, ase do nabsorptio nspectrum mea- are shown for the stations having the least and the most sured on Thalassia testudinum leave s(solid line) ,and a hypothetical separation between PAR and PUR in Figure 4. The dif- action spectrum derived by assuming only ligh tabsorbed by chloro- fuse attenuation coefficients for each of the three quanti- phylls a and b drive photosynthesis in Thalassia (dashed line). ties are reported for all stations in Table 1. At the Blue NUMBER 38 e¢ 365 the deep edges also occur. Because of absorption by water, virtually no light is present at wavelengths greater than TABLE 1. Depths of seagrass deep edge (Zax) and attenuation 600 nm at the BGR location in Belize and very little at coefficients for photosynthetically active radiation (Kpap) and STRI in Panama. Increasing amounts of red wavelengths photosynthetically usable radiation (PUR )weighted by measured are present at the SNO3 and IRL sites as a result of the absorption spectrum o fThalassia testudinum leaves (Kpypm) or shallower depths of the deep edges. The peaks of the in weighted by a hypothetica laction spectrum (Kpypn ;see Figure 2). Percentage o fsurface ligh tpenetrating to the seagrass deep edge situ spectra shift progressively toward green wavelengths i sgiven in parentheses. along the progression from BGR to IRL, and the greatest similarities are at 400 to 410 nm, where the percentage of surface irradiance remaining ranges from 2% to 6%. Vlawew Kpar Kpurm Kpurh Site 2 (m) (m7) (m~') (m~') Perce nsturfac ierradiance BGR, Belize 10 0.293 0.293 0.314 0 20 40 60 80 100 (5.2%) (5.4%) (4.2%)STRI, Panama 8.5 0.232 0.247 0.304 (13.6%) (12.0%) (7.4%) SNO3, Panama 2.4 0.836 0.945 1.098 (14.1%) (11.0%) (7.7%) IRL, Florida 0.92 1.157 1.301 1.52 (32.7%) (27.1%) (21.8%) 4BG RB,lu Geroun Rdang eS;TR SIm, ithsonia Tnropic aRlesearc Ihnstitut eS;N03, Panam caree sktatio nIR; LIn, dia Rniv eLragoon. E s Q® [ok DISCUSSION 144. a. Belize At all three study sites we were able to locate a distinct deep edge of the Thalassia testudinum meadows, charac- 16 terized by a transition from moderate and sparsely veg- Perce nsturfac ierradiance etated seagrass to either unvegetated substrate or patches 0 20 40 60 80 100 of the smaller, low light adapted seagrass Halophila de- cipiens. Where we were able to sample shallower sites in Belize and Panama, there were substantially higher densi- ties of T. testudinum. The presence of H. decipiens at the Blue Ground Range (BGR) station in Belize and the STRI site in Panama further confirmed that we were sampling at light-limiting edges of the T. testudinum distribution. E Halophila decipiens is a small, ruderal species of seagrass s[oe commonly found growing in deep or turbid water and has Q® lower light requirements than T. testudinum (Kenwor- thy, 2000; Gallegos and Kenworthy, 1996; Kenworthy et al., 1989). The presence of H. decipiens at these two bF.lorida stations was a good indication of light-limiting conditions for Thalassia. Although we did not record H. decipiens at SNO3 in Panama, a thorough visual examination by divers FIGURE 4 .Vertica lprofiles of photosynthetically active radiation at deeper depths than the observed T. testudinum distribu- (PAR, solid line), and photosynthetically usable radiation (PUR) tion confirmed there were no seagrasses growing beyond based on measured absorption spectrum (PURm ;dashed line )and 2.4 m depth. hypothetica laction spectrum (PURh ;dotted line) in (a) Belize and Attenuation coefficients for PAR and PUR were nearly (b) the Indian River Lagoon (IRL), Florida. Profiles were normal- indistinguishable in Belize and Panama and differed only ized to the irradiance incident at the surface (100%). slightly in the IRL. Based on these one-time profiles, we calculated that seagrass grew to depths of penetration of 366 °¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES 50 percentages calculated, and repeating this study during an- other season could yield different percentages. Nonetheless, assuming that the light requirements for seagrasses at Belize are similar to those in Panama, the regional differences in light requirements between the IRL ‘= oNwOoOoO and the two tropical sites remain striking. Qualitatively, the differences are consistent with the observations of PAR (sZu_r_f_Pa)ceercent Duarte et al. (2007) that seagrasses growing in shallow, turbid waters (e.g., IRL) have higher light requirements than those growing in clear, deep water (Panama, Belize). Calculation of PUR closed the gap only slightly, leading =) us to conclude that factors other than spectral energy dis- tribution contribute substantially to site-specific light re- quirements of seagrasses, especially at the deep edges. An extended growing season in the more tropical locations of Waveleng t(nhm) Belize and Panama could possibly account for some of the a a a difference. The tropical sites receive about 7% more inci- FIGURE 5 .Spectra of photosynthetically active radiation (PAR) at dent radiation annually than the IRL site, most of which the depth o fthe seagrass deep edge (Zax) in Belize (BGR ,solid line), occurs during winter months (November through Febru- Panama (STRI, dashed line, and SN03, dotted line), and Florida ary) when temperatures are also more favorable in the (IR L[ICW194] d, ot-dashed line). tropics. Other possible differences between sites include leaf-shading epiphytes, sediment quality (e.g., grain size or organic matter content), and possible periods of low oxygen in thermally stratified deeper waters. These latter 33% of PAR in the IRL, 14% in Panama, and approxi- factors have management implications because they are all mately 5% in Belize. Corresponding percentages for PUR affected by coastal eutrophication. Improved understand- were 27%, 12%, and 5% for IRL, Panama, and Belize, ing of the factors accounting for site-specific differences in respectively. The accuracy of these estimates depends on seagrass light requirements is, therefore, urgently needed. the degree to which the profiles were measured under conditions that are typical for their respective growing ACKNOWLEDGMENTS seasons. We are fairly certain this was not the case in Belize, where strong northerly winds, atypical for the We gratefully acknowledge the Smithsonian Marine Sci- season, blew for several days before and on the day of ence Network for Pilot Project funds to conduct these stud- sampling. Horizontally sighted Secchi disk visibility at a ies. We thank Sam Benson for diving assistance in Belize and seagrass bed near Twin Cays was 5.5 m during the time Amy Lewis for help with optical data reduction and analy- of our measurements, compared with annual means of sis. This is contribution no. 844 of the Caribbean Coral Reef 10.1 m (+0.38 m SE) for 2004 and 8.9 m (+0.25 m SE) Ecosystem Program (CCRE), supported in part by the Hunt- for 2005 (see Koltes and Opishinski, 2009: fig. 6, this erdon Oceanographic Research Fund, and Smithsonian Ma- volume). If the water column were more strongly stirred rine Station at Fort Pierce contribution no. 787. with higher than typical concentrations of particulate matter, then our estimates for Belize would be biased low, as we suspect they are. The estimated PAR light require- LITERATURE CITED ments for the IRL are, however, based on more frequent visits and are in agreement with other published estimates Batiuk ,R .A. ,P .Bergstrom ,M .Kemp ,E .Koch ,L .Murray ,J .C .Steven- (Kenworthy and Fonseca, 1996; Steward et al., 2005). The son ,R. Bartleson ,V. Carter ,N. B. Rybicki ,J. M. Landwehr ,C. L.Gallegos ,L .Karrh ,M .Naylor ,D .Wilcox ,K .A .Moore ,S .Ailstock, limitation of our approach was the inability to determine an dM T. eichberg 2. 000 C. hesapeak eBa ySubmerge dAquati cVeg- the integral of light requirements for the whole growing etatio nWate Qr ualit ayn dHabitat-Base dRequirement asn dResto- season from only a few days of measurements. Because of ratio nTargets A :Secon dTechnica slynthesis U. S. E. nvironmental this limitation, it is unlikely that the observed depth distri- Protectio Angenc Cy,hesapeak Bea Pyrogram.Braun-Blanquet J,1. 965 P. lan tSociology T: h eStud yo fPlan tCommuni- bution of the seagrasses is fully captured by PAR and PUR tie Lso. ndo nH:affner. NUMBER 38 ¢ 367 Buiteveld, H., J. H. M. Hakvoort, and M. Donze. 1994. “The Optical Kirk ,J .T .O .1992 .Monte Carlo Modeling of the Performance of a Re- Properties of Pure Water.” In Ocean Optics XII, ed. J. F. Jaffe, flectiv eTub eAbsorptio nMeter A. pplie dOptics 3, 1:6463-6468. pp 1. 74-183 B. ellingham W, ash .S: ociet yo fPhoto-Optica Ilnstru- Koltes ,K .H. ,and T .B .Opishinski .2009 .“Patterns of Water Quality ment aEtniognineers. and Movemen tin the Vicinity o fCarrie Bow Cay ,Belize.” In Pro- Carter ,V. ,N .B .Rybicki ,J .M .Landwehr ,and M .Naylor .2000 .“Light ceeding so tfh eSmithsonian Marin eScienc eSymposium e, d M. A. . Requirement sfo rSAV Surviva al nd Growth. ”In Chesapeake Bay Lang ,I .G .Macintyre ,and K .Ritzler ,pp .379-390 .Smithsonian Submerge dAquat iVcegetatio nWate Qr ualit yan dHabitat-Based Contribution sto the Marine Sciences N, o .38 .Washington D, .C.: Requirement san dRestoratio nTargets A :Secon dTechnica Sl yn- Smithsoni aInstituti oSnchola rPlyress. thesis ,ed. R. A. Batiuk ,P. Bergstrom ,W. M. Kemp ,E. W. Koch, Krause-Jensen, D., S. Sagert, H. Schubert, and C. Bostrom. 2008. L .Murray ,J .C .Stevenson ,R .Bartleson ,V .Carter ,N .B .Rybicki, Empirica lRelationship sLinking Distribution and Abundance J. M. Landwehr ,C. L. Gallegos ,L. Karrh ,M. Naylor ,D. J. Wilcox, o fMarine Vegetation to Eutrophication .Ecologica Ilndicators, K. A. Moore ,S. Ailstock ,and M. Teichberg ,pp. 4-15. U. S. Envi- 8:515-529. ronment aPlrotectio Angenc yC,hesapeak Bea Pyrogram. Lang ,M .A .2009 .“Introduction to the Smithsonian Marine Science Dennison, W. C., R. J. Orth, K. A. Moore, J. C. Stevenson, V. Carter, Network. ”I nProceeding so tfh eSmithsonia nMarin eScienc eSym- S .Kollar ,P .W .Bergstrom ,and R .A .Batiuk .1993 .Assessing Water posium ,ed. M. A .Lang ,I. G .Macintyre ,and K. Riitzler ,pp. 1-9. Qualit wy it hSubmerse dAquat iVcegetation B.ioScienc e4,3:86-94. Smithsonian Contribution sto th eMarin eSciences N, o 3. 8 W. ash- Drake ,L .A. ,F .C .Dobbs ,and R .C .Zimmerman .2003 .Effects of Epi- ington D, .C S.:mithsonia nInstitutio nScholar lPyress. phyt eLoad on Optica Pl ropertie sand Photosyntheti cPotentia ol f Mobley ,C .D .1994 .Light and Water .Radiative Transfer in Natural th eSeagrasse sThalassi atestudinum Bank se xK6nig and Zostera Water sS.a nDiego A:cadem iPcress. marin aL L.imnolog yan dOceanography 4,8:456-463. Mobley ,C .D. ,L .K .Sundman ,and E .Boss .2002 .Phase Function Ef- Duarte C, .M. N, .Marba D, .Krause-Jensen a, nd M .Sanchez-Camacho. fect son Oceani cLigh tFields A. pplied Optics 4, 1:1035-1050. 2007 T. estin gth ePredictiv ePowe ro Sf eagras sDept hLim iMt odels. Morel ,A .1978 .Available ,Usable ,and Stored Radian tEnergy in Rela- Estuarie asn dCoast s3,0:652-656. tio nt oMarin ePhotosynthesi sD. eep-Se aResearch 2,5:673-688. Falkowski ,P .G. ,and J .A .Raven .2007 .Aquatic Photosynthesis ,2nd ed. Orth, R. J., T. J. B. Carruthers, W. C. Dennison, C. M. Duarte, J. W. Princeton N, J ..P: rinceto nUniversit yPress. Fourqurean, K. I. Heck Jr, A. R. Hughes, G. A. Kendrick, W. J. Gallegos ,C .L .1994 .Refining Habita tRequirements o fSubmersed Kenworthy ,S .Olyarnik ,FE .T .Short ,M .Waycott ,and S .L .Wil- Aquati cVegetation R: ol eo Of ptica Ml odels E.stuaries 1,7:198-219. liams 2. 006 A. Globa Cl risi sfo rSeagras sEcosystems B. ioScience, Gallegos ,C .L. ,and W .J .Kenworthy .1996 .Seagrass Depth Limits in 56:987-996. the Indian River Lagoon (Florida ,U.S.A.) :Application o fan Op- Pope ,R .M. ,and E .S .Fry .1997 .Absorption Spectrum (380-700 )nm of tica lWate rQuality Model .Estuarine Coasta land Shel fScience, Pur eWater I. II.ntegratin gCavit yMeasurements A. pplie dOptics, 42:267-288. 36:8710-8723. Hemminga ,M .A. ,and C .M .Duarte .2000 .Seagrass Ecology .Cam- Ralph, P. J., M. J. Durako, S. Enriquez, C. J. Collier, and M. A. Doblin. bridg Ce:ambridg Uenivers iPtyress. 2007 I.mpac to fLigh tLimitation on Seagrasses J.ourna ol fExperi- Kenworthy ,W .J .2000 .The Role o fSexua lReproduction in Maintain- menta Ml arin eBiolog yan dEcology 3, 50:176-193. in gPopulation so Hf alophil adecipiens I:mplication sfo rth eBiodi- Steward ,J.S. ,and W.C .Green .2007 .Setting Load Limits fo rNutrients versit yan dConservatio no Tf ropica Sleagras sEcosystems P. acific an dSuspende dSolid sBase dupo nSeagras sDepth-Lim iTtargets. Conservatio Bniolog 5y:,251-259. Estuarie asn dCoast s3,0:657-670. Kenworthy, W. J., C. A. Currin, M. S. Fonseca, and G. Smith. 1989. Steward ,J .S. ,R .W .Virnstein ,L .J .Morris ,and E .F .Lowe .2005S .Setting Production D, ecomposition a,n dHeterotrophi cUtilizatio no tfhe Seagras sDepth C, overage a, nd Ligh tTarget sfo rthe Indian River Seagrass ,Halophila decipiens ,in a Submarine Canyon .Marine Lagoo nSystem F,lorida E.stuarie s2,8:923-935. Ecolog Pyrogres Sserie s5,1:277-290. Tzortziou, M., J. R. Herman, C. L. Gallegos, P. J. Neale, A. Subrama- Kenworthy ,W .J. ,and M .S .Fonseca .1996 .Ligh tRequirements o fSea- niam, L. W. Harding Jr., and Z. Ahmad. 2006. Bio-optics of the grasse Hs alodul ewright iain dSyringodium filiform eDerive dfrom Chesapeak eBa yfrom Measurement san dRadiativ eTransfe Cr lo- th eRelationshi pBetwee nDiffus eLigh At ttenuatio nan dMaximum sure E. stuarin eCoasta al n dShe lSf cience 6, 8:348-362. Dept Dhistributio nE.stuarie s1,9:740-750. Zaneveld ,J .R .V. ,J .C .Kitchen ,and C .Moore .1994 .“The Scattering Er- Kenworthy W, J. .a, nd D E. H. aunert 1. 991 T. he Ligh tRequirement so fSea- ro rCorrectio no Rf eflecting-Tub eAbsorptio nMeters. ”pp 4. 4-55. grasses P: roceeding so afWorksho pt oExamin eth eCapabilit yo Wf a- In Ocean Optic sXII ,ed .S .Ackleson .Bellingham ,Wash. :SPIE. te Qr ualit yCriteria S, tandard san dMonitorin gProgram st oProtect Zimmerman ,R .C .2003 .A Biooptica lMode lo fIrradiance Distribution Seagrasse Ns.ation aOlcean aicn Adtmospher Aicdministratio Tnech- an dPhotosynthesi si nSeagras sCanopies L.imnolog yan dOcean- nic aMl emorandum NMFS-SEFC-28 7W. ashingto nD,.C N.:OAA. ograp 4h8y:,568-585. Gallegos, Charles L. et al. 2009. "Underwater Spectral Energy Distribution and Seagrass Depth Limits along an Optical Water Quality Gradient." Proceedings of the Smithsonian Marine Science Symposium 38, 359–367. View This Item Online: https://www.biodiversitylibrary.org/item/131385 Permalink: https://www.biodiversitylibrary.org/partpdf/387366 Holding Institution Smithsonian Libraries and Archives Sponsored by Biodiversity Heritage Library Copyright & Reuse Copyright Status: In Copyright. Digitized with the permission of the rights holder License: http://creativecommons.org/licenses/by-nc-sa/3.0/ Rights: https://www.biodiversitylibrary.org/permissions/ This document was created from content at the Biodiversity Heritage Library, the world's largest open access digital library for biodiversity literature and archives. Visit BHL at https://www.biodiversitylibrary.org. This file was generated 31 May 2024 at 05:46 UTC