This article was published in an Elsevier journal. The attached copy is furnished to the author for non-commercial research and education use, including for instruction at the author’s institution, sharing with colleagues and providing to institution administration. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy
Journal of Experimental Marine Biology and Ecology 353 (2007) 115 – 125 www.elsevier.com/locate/jembe
Comparing shoot population dynamics methods on Posidonia oceanica meadows José Miguel González-Correa a,b,⁎, Yolanda Fernández-Torquemada b , José Luis Sánchez-Lizaso b a
Centro de Investigación Marina de Santa Pola, CIMAR, Ayto. de Santa Pola y Universidad de Alicante. Torre d'Enmig s/n, Cabo de Santa Pola, Alicante, Spain b Departamento de Ciencias del Mar y Biología Aplicada, Universidad de Alicante. POB 99, E-03080-Alicante, Spain Received 11 May 2007; accepted 19 September 2007
Abstract Prediction capacity of three main shoot population dynamics methods (age structure, net shoot recruitment per plagiotropic rhizome and shoot census) have been tested for a period of four years (2002–2006) on a Posidonia oceanica meadow. Accuracy of each method was checked by comparing measured and predicted densities at the end of the study period. Predicted densities came from the evolution of initial densities (measured in 2002) by a basic exponential model of population growth. The exponential model used the different net shoot recruitment rate estimates by each population dynamics method on three depths (upper, medium and lower limit) and three localities at each depth. Predictions performed by shoot census and net shoot recruitment per plagiotropic rhizome methods matched with measured densities at the end of the study period. Conversely, age structure method underestimated shoot densities at each depth, indicating an unreal decrease of shoot population in the meadow. © 2007 Elsevier B.V. All rights reserved. Keywords: Mortality rate; Net shoot recruitment rate; Population demography; Posidonia oceanica; Shoot census; Shoot population dynamics
1. Introduction Posidonia oceanica is a seagrass endemic to the Mediterranean Sea which is considered to be the climax community on soft substrata in the shallow waters of the Mediterranean (Pérès and Picard, 1964). It plays an important multifunctional role in the coastal ecosystems in relation to productivity, habitat, hydrodynamic
⁎ Corresponding author. Departamento de Ciencias del Mar y Biología Aplicada, Universidad de Alicante. POB 99, E-03080Alicante, Spain. Tel.: +34 965 903400x2916; fax: +34 965 903815. E-mail address:
[email protected] (J.M. González-Correa). 0022-0981/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jembe.2007.09.008
protection and the provision of resources for invertebrate and fish communities (Pergent and PergentMartini, 1991; Francour, 1997; Sánchez-Jerez et al., 1999; Hemminga and Duarte, 2000; Harriague et al., 2006). P. oceanica meadows are formed by the regular addition of a basic structural module consisting of a piece of rhizome (internode) and a node bearing a cluster of leaves (shoots) and roots (Harper and Bell, 1979; White, 1979). Shoot population dynamics reflect the balance between recruitment and mortality of shoots; therefore, the meadow may be considered as a closed population (without immigration and emigration). Net population growth rate per individual shoot (r) may be calculated as the difference between the per
Author's personal copy
116
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
capita birth rate (gross recruitment, Rgross) and the death rate (mortality, M): r ¼ Rgross M P. oceanica shows a low Rgross as compared to other seagrasses (Marbà and Duarte, 1998); a factor which intrinsically handicaps the study of its population dynamics. Indirect and direct methods have been developed to estimate Rgross and M rates. Indirect methods provide retrospective assessments of meadow dynamics over time scales of up to years or decades over the past (Fourqurean et al., 2003). They are featured by low sampling effort; indeed, only unique
random sampling is required. Sampled rhizomes are dated by reconstruction techniques (plastochrone interval or lepidochronological analysis), which allow the application of two indirect methods: one based on age distribution of the living vertical (hereafter orthotropic) shoots of the population (Duarte et al., 1994; Peterson and Fourqurean, 2001) and the other on the net shoot recruitment per horizontal (hereafter plagiotropic) rhizomes (González-Correa et al., 2005, 2007). The age structure method has been used for a broad range of plant types, including seagrasses, palms and mangroves (Duarte et al., 1994, 1999; Peterson and Fourqurean, 2001) since at least 1973 (Patriquin, 1973). But, its validity has been disputed in recent years
Fig. 1. Study area; localities and depths situation.
Author's personal copy
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
because of the previous assumptions of equilibrium in the population (r = 0, therefore Rgross = M; Ebert et al., 2002, 2003; Fourqurean et al., 2003), which should restrain its use on unstable meadows. Conversely, the net shoot recruitment per plagiotropic rhizome method has been recently proposed and used only with P. oceanica (González-Correa et al., 2005, 2007). This method changes the focus from orthotropic to plagiotropic rhizomes, which are considered responsible for vegetative growth of meadow. Shoot census has been proposed as the most reliable way to asses shoot demographic status of seagrass populations (Short and Duarte, 2001). It has been performed thorough cable tie tagging of shoots, which allows the survey of Rgross, M and therefore r shoot rates on fixed plots for a limited period of time (Marbà et al., 2005). This method requires regular visits, requiring increased dive time and greater economic resources. Population dynamics methods are an important tool in the management of seagrass meadows. They are used both to reflect the recent past, as well as, to forecast the near future. Indeed, its outcomes support important and costly decisions related to management such as evaluation impact analysis (Durako, 1994; Peterson and Fourqurean, 2001), surveillance of putative impacts
117
(Fernández-Torquemada et al., 2005) or the creation of new marine protected areas (Sánchez Lizaso, 1993; Marbà et al., 2002). However, the accuracy and possible bias of the different methods remain unknown. The aim of this study was to compare the three main population dynamic methods (age structure, net shoot recruitment per plagiotropic rhizome and shoot census). To check the accuracy of methods, we compared predicted and measured densities four years after the study started. Predicted densities emanated from the evolution of initial densities (measured in 2002) by applying a basic exponential model of population growth. The exponential model used the different net shoot recruitment rates estimated by each population dynamics method. 2. Materials and methods 2.1. Study area and sampling design Study was carried out between June 2002 and June 2006 on a P. oceanica meadow located at Jávea (Alicante, SE Spain, SW Mediterranean Sea; Fig. 1). According to a survey specifically conducted for this study, P. oceanica meadow is a healthy meadow with a
Fig. 2. Representation of indirect method used to date principal plagiotropic rhizomes. A; primary plagiotropic; B secondary plagiotropic; C: secondary orthotropic; D: tertiary branches.
Author's personal copy
118
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
low percentage of death “matte” (b 5%). It ranges from 3 to 15 meters in depth and the height of “matte” reached 1–1.5 m. Shoot density ranges from ∼ 1200 shoots m− 2 at the upper limit, ∼ 530 shoots m− 2 at middle depth and ∼ 220 shoots m− 2 at the lower limit. These average densities agreed with standard values proposed by
Fig. 3 (continued ).
density-depth function for healthy meadow in PergentMartini et al. (1994). The meadow is rooted on a heterogeneous rocky bottom at the upper limit while at the lower limit it grows predominately on a coarse sandy bottom. Three localities, separated from 1.5 to 3 km, were randomly placed at upper (3 m), medium (7 m) and lower limits (15 m in depth) of meadow. Shoot density was measured at the start (in 2002) and at the end (in 2006) of the study period by four random plots (40 × 40 cm) at each locality and depth. At each locality, net shoot recruitment (r) was estimated through three different seagrass population dynamics methods: 2.1.1. Shoot census (SC) Four fixed plots (40 × 40 cm), framed by sticks and cable, separated from 1 to 10 m were randomly placed within each locality. At each plot, 20–30 shoots were tagged by cable ties. An annual visit was performed in order to count new, death and remaining shoots from June 2002 to June 2006. Net shoot recruitment per shoot (r) was calculated each year as the difference between recruited shoots per shoot (Rgross) and dead shoots per shoot (M). Inter-annual average of r was calculated for each locality.
Fig. 3. Gross recruitment and mortality rate estimated by SC method at 1) 3 m, 2) 7 m and 3) 15 m in depth. Data are means (± SE), n = 12.
2.1.2. Net shoot recruitment per plagiotropic rhizome (NSRP) In this method, principal plagiotropic rhizomes were considered as the basic unit of expansion in the meadow.
Author's personal copy
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125 Table 1 Net shoot recruitment estimated by SC and AS methods and net plagiotropic rhizomes estimated by NSRP method for each depth. Data are means (±SE), n = 3 Methods
Depth 3 m
Depth 7 m
Depth 15 m
SC NRSP AS
0.04 ± 0.05 0.15 ± 0.06 − 0.12 ± 0.02
0.03 ± 0.02 0.08 ± 0.07 − 0.15 ± 0.06
− 0.02 ± 0.04 − 0.02 ± 0.03 − 0.19 ± 0.03
Density of principal plagiotropic rhizomes was estimated at the start of sampling (June, 2002) by four random plots (40 × 40 cm). In addition, ten principal plagiotropic
119
rhizomes were harvested in June 2006. Each rhizome bore a variable number of orthotropic and secondary plagiotropic rhizomes. Plagiotropic rhizomes age was estimated by lepidochronology technique (Pergent and Pergent-Martini, 1991). This technique is based on thickness cycle of sheaths attached to rhizomes; identifying an annual period as equating to the length of rhizome as measured between two sheaths of minimum thickness. Minimum thickness sheaths are obvious in orthotropic rhizomes but not in plagiotropic rhizomes (Pergent, 1987). So, the age of a plagiotropic piece of rhizome was calculated as the age expressed by
Fig. 4. 1) Principal plagiotropic rhizomes per m2, 2) new plagiotropic rhizomes per principal plagiotropic rhizome and year, 3) death principal plagiotropic rhizomes per principal plagiotropic rhizome and year and 4) net recruitment of shoots per principal plagiotropic rhizome and year at the three depths of study. Data are means (±SE), and n = 12 at Fig. 4-1 and n = 30 at Fig. 4-2, -3 and -4.
Author's personal copy
120
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
Fig. 5. Age structure of shoot population at 1) 3 m, 2) 7 m and 3) 15 m.
subtracting between the youngest from the oldest orthotropic rhizome to which it was connected (Fig. 2). This dating method was not sensitive to plagiotropic rhizomes younger than 1 year; therefore, only rhizomes older than 3 years were selected in order to minimize such imprecision. Annual Rgross, M and r rates of principal plagiotropic rhizomes were estimated. Branching of new plagiotropic rhizomes on principal plagiotropic rhizome was considerate as Rgross rate. M rate was estimated by the following algorithm, M ¼ Rm=ðRt AÞ where Rm corresponds to dead plagiotropic rhizomes; Rt is the total of rhizome harvested and A, is the average age of harvested rhizomes. In addition, the average of net shoot recruitment per principal plagiotropic rhizome was calculated for each locality.
2.1.3. Age structure (AS) The dynamics of P. oceanica was inferred from the age structure of thirty alive shoots randomly harvested in June 2006 according to Peterson and Fourqurean (2001). Shoot age (in years) was estimated as the total number of leaves produced during the life-span of a shoot (that is the sum of standing leaves plus leaf sheaths and scars = number of PIs) divided by the average number of leaves produced by an orthotropic rhizome annually. This average was calculated applying lepidochronological technique to five orthotropic rhizomes. Shoot mortality rate (M, yr− 1) was derived from the exponential decline in the number of living shoots (N0) with time (t, in years) as, Nt ¼ N0 eM t where N0 is the number of shoots with age equal to the
Author's personal copy
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
121
Fig. 6. Shoot mortality rate estimated from slope of semilogarithmic regression at 1) 3 m; 2) 7 m and 3) 15 m in depth. Data are means (±SE), n = 90.
mode, and Nt is the number of shoots older than the modal age at time t. M was estimated using a semilogarithmic linear regression model, where constant mortality over shoot age classes and years was assumed. The annual shoot recruitment rate (R, yr− 1) was calculated as, k P
Rgross ¼
eð1n nj þMtj Þ
where Nt is the shoot density at the time t, N0 is the initial shoot density (measured at 2002), r is the average interannual of net shoot recruitment, which was calculated through the three different methods. Finally t, corresponds to the time expressed in years. In the NSRP method, r is related to net recruitment of principal plagiotropic rhizomes. Therefore the result (Nt) was converted to shoot
j¼1
N th
where n is the number of living shoots in the j age class, t is the age of the jth age class, k is the oldest age class b 1 yr− 1 old, M is the mortality rate per capita, and N is the total number of living shoots in all age classes. To calculate shoot density evolution per year, the net shoot recruitment obtained by each method was applied to the following basic population dynamic equation, Nt ¼ N0 ert
Table 2 Analysis of variance (ANOVA) results comparing initial and final measured densities at three depths (upper, medium and lower limit) and at three localities for each depth Source of variation
F-versus
d.f.
MS
F
T D L (T × D) T×D Residual
L (TXD) L (TXD) Residual L (TXD)
1 2 12 2 54
29454 5083856 61814 46564 30445
0.48 n.s. 82.24 ⁎⁎⁎ 2.03 ⁎ 0.75 n.s.
⁎⁎⁎p b 0.001; n.s. not significant.
Author's personal copy
122
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
Before analysis, Cochran's test was used to test density variance homogeneity. Where significant differences were found the data was ln(x + 1) transformed. When the factors of interest showed significant differences a posteriori pair-wise comparisons of means were performed using the Student–Newman–Keuls (SNK) test. 4. Results 4.1. Shoot census (SC) method M and Rgross rates did not show any significant pattern (p N 0.05, Fig. 3) within the study period. Both rates remained within similar range, from ∼ 0.020 to 0.16 shoots (shoot year)− 1, and as a consequence, interannual average of r rate was close to zero with a large variability at each depth (Table 1). Fig. 7. Representation of predicted by the three methods and measured shoot densities at 3 m, 7 m and 15 m from 2002 to 2006. Data are means (±SE), n = 4.
density by multiplying it by the average of net shoot recruitment per principal plagiotropic. 3. Data analysis An orthogonal analysis of variance design (ANOVA) was employed to compare the final densities estimated by NSRP, SC, AS and data measured at 2006. Linear model for this analysis was: Xijn ¼ c þ Mi þ Dj þ Mi Dj þ ResidualnðijÞ where μ is the overall mean, Mi is the effect of the ith method and Dj is the effect of depth. Residualn(ij) is the error term that estimates variability among samples. Mi and Dj are fixed factors. Initial and final measured shoot density were compared with an orthogonal model that included a temporal factor Ti, related to moment of sampling (2002– 2006) and a spatial factor Lk. The linear model was: Xijn ¼ c þ Ti þ Dj þ Lk Ti Dj þ Ti Dj þ Residualnðk ðijÞÞ
where Ti was the effect of ith year and Lk corresponds to effect of kth locality. Table 3 Analysis of variance (ANOVA) results comparing final densities estimated by SC, NSRP, AS methods and direct measures at upper, medium and lower limit Source of variation
F-versus
d.f.
MS
F
M D M×D Residual
Residual Residual Residual
3 2 6 24
1.087 9.700 0.039 0.098
11.03⁎⁎⁎ 99.00⁎⁎⁎ 0.40n.s.
SNK test: SC = NSRP = Me N AS. ⁎⁎⁎p b 0.001; n.s. not significant.
4.2. Net shoot recruitment per plagiotropic rhizome (NSRP) method Principal plagiotropic density was approximately 2fold greater at 3 than at 7 and 15 m depth (Fig. 4-1). The average age of pieces of principal plagiotropic rhizomes harvested were 3.7, 4.8 and 5.9 years old at 3, 7 and 15 m depth respectively. These pieces showed Rgross and M rates with a similar range consistently lower than 6% (Fig. 4-2 and -3). Therefore, average of net principal plagiotropic rhizome recruitment remains close to zero (Table 1) with a large variability. In addition an expected decreasing pattern of net shoot recruitment per principal plagiotropic rhizome was detected in relation to increasing depths (Fig. 4-4). 4.3. Age structure (AS) method The age structure showed a modal age of 5 years at 3 and 15 m depth and 6 years at 7 m. Modal age shoot cohort was followed by an exponential decline due to shoot mortality (Duarte et al., 1994; Fig. 5). From this age structure, Rgross rate calculated was 0.069 ± 0.037, 0.076 ± 0.046 and 0.043 ± 0.001 at 3, 7 and 15 m in depth
Table 4 Percentage differences between predicted densities by NSRP, SC and AS methods vs measured densities for each depth at final of study period. Data are means (±SE), n = 3 Methods
Depth 3 m
Depth 7 m
Depth 15 m
SC NSRP AS
40.6 ± 40.18 36.0 ± 17.75 − 30.6 ± 9.30
11.8 ± 13.48 13.2 ± 21.48 − 44.7 ± 7.86
− 12.4 ± 15.65 2.4 ± 5.22 − 56.8 ± 4.88
Author's personal copy
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
respectively. Whereas, M rate, calculated from the slope of a semilogarithmic regression, ranged from ∼ 0.2 to 0.34 (Fig. 6). Consequently, an obvious negative trend of net shoot recruitment rate can be deducted by this method at each depth (Table 1). Measured densities in 2002 and 2006 were not significantly different (Table 2); in addition, predicted densities in 2006 by SC and NSRP matched with measured densities at each depth. But, predicted densities by AS method were significantly lower than measured and predicted by the other two methods (Fig. 7, Table 3). Indeed, predicted densities by AS method ranged from ∼ 31 to 53% lower than measured ones in 2006 (Table 4). 5. Discussion Predictions made by SC and NSRP methods on a steady meadow matched with measured densities at the end of the study period. Conversely, predictions from the AS method underestimated shoot densities at each depth, indicating an unreal decrease of shoot population in the meadow. Predicted densities by SC and NSRP methods showed the highest inaccuracy compared with measured densities at 3 m in depth (Table 4). This deviation could be explained by the natural density variability (Panayotidis et al., 1981; Balestri et al., 2003; Leriche et al., 2006), especially increased by the heterogeneity of sea bottom at the upper limit. The AS method has been intensively used in the last years (Marbà et al., 1996, 2002; Peterson and Fourqurean, 2001). However, it has been recently questioned because of a failure to check previous assumptions of the method: the age distribution has to be stable (r = 0 and R = M) as well as the age independence of R and M rates (Ebert et al., 2002, 2003; Fourqurean et al., 2003). In this study, both previous assumptions were checked by SC method, obtaining that r ∼ 0 and that R and M rates did not show a significant trend over the study period. On the other hand, AS method assumes that age distribution of harvested shoots is representative of the whole shoot population. A logic age structure of whatever natural population would fit with an exponential curve: a maximum number of young individuals followed of an exponential decrease with time. But, the age distribution of living harvested shoots in AS method is often characterised by few very young shoots, and exponentially declining numbers with increasing shoot age (Duarte et al., 1994). The separation of modal cohort from the youngest cohort has been reported from zero to even eight years for P. oceanica (Marbà et al., 1996).
123
Here, modal cohort was separated five to six years according depth. This misadjustment has been explained by a severe reduction in shoot recruitment over a number of years before the sampling date (Marbà et al., 1996). However, this explanation does not fit with the results of SC method in our study case. Evidently, it implies that age structure of harvested shoots was biased, which in turn led to underestimation of Rgross and overestimation of M rates. Collection procedures of shoots has already been pointed out by Jensen et al. (1996) as being responsible for a demographic bias, which together with problems in the sampling design, was linked to erroneous predicted density decline of Thalassia testudinum populations in Florida Bay (Florida, USA). Thus, predictions of shoot population dynamics reported in the literature that used this method should be taken in to account with caution. For correct application, this method would require that the harvested shoot age structure fits with the whole shoot age structure. To do this, additional sampling techniques would need to be developed. For P. oceanica meadows, it could be carried out by the extraction of a big random square of “matta” (i.e. 1 m2) and the subsequent ageing of all its shoots. However, this would involve an arduous and largely destructive process. SC method facilitated a detailed and direct survey of the evolution of shoot population. Its application requires repeated visits, substantial increase in dive-time and economic resources (Marbà et al., 2005). This work should be carried out by trained researchers because it is a tedious and accurate job performed under hostile conditions. In addition, repeated handling of shoots, as well as, an incorrect fitting of cable ties too close to the meristematic apex could produce additional mortality. In this study no significant increase of M rate due to handling has been observed in comparison with other methods used. SC method showed an important range of variability (Fig. 3). This may be due to the fact that shoots were tagged in groups of neighbouring shoots, framed under similar environmental conditions, and with similar stress levels. The position of sampling squares on the meadow (within the mass or on the edge, over sand or rocky substratum,…) determine that the shoots belonging to the same plot have bigger or smaller M and Rgross rates. To reflect correctly the global trend of shoot population dynamics using this method it is recommended to design a multi-scale sampling method to reflect the variability at different spatial scales. The NSRP method is based on the assessment of meristematic apex of plagiotropic rhizomes to yield new shoots, and hence precludes previous assumptions inherent in the demographic method. Meristematic
Author's personal copy
124
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125
apex on plagiotropic rhizomes is mainly responsible for vegetative plant growth (Tomlinson, 1974); therefore, this method assessesed the support of dynamics population shoots directly. But, outcomes are constrained to the medium-term because of the impossibility of extracting very long plagiotropic rhizomes from the meadow. Plagiotropic rhizomes harvesting deserves special attention. Divers have to be sure first, to perform random harvesting that includes living and dead plagiotropic rhizomes. Secondly, it is important to completely pull out the plagiotropic rhizome from the oldest shoots to the final apex, in order to be able to determine if it is alive or dead. These cautions must be taken in to account in order to not underestimate M rate of plagiotropic rhizomes. Here, the similar outcomes obtained by SC method and measured densities vs. NSRP method, requires that M rates are not underestimated. Finally, plagiotropic rhizome harvesting is a very aggressive technique, and it should be only used with caution, never on meadows with notable regressions or with a limited surface. In conclusion, SC and NSRP methods are shown to be more accurate in the medium- term than AS method; which tends to underestimate recruitment and overestimate mortality. References Balestri, E., Cinelli, F., Lardicci, C., 2003. Spatial variation in Posidonia oceanica structural, morphological and dynamics features in a northwestern Mediterranean coastal area: a multi-scale analysis. Mar. Ecol. Prog. Ser. 250, 51–60. Duarte, C.M., Marbà, N., Agawin, N., Cebrian, J., Enríquez, S., Fortes, M.D., Gallegos, M.E., Merino, M., Olesen, B., Sand Jensen, K., Uri, J., Vermat, J., 1994. Reconstruction of seagrass dynamics: age determinations and associated tools for the seagrass ecologist. Mar. Ecol. Prog. Ser. 107, 195–209. Duarte, C.M., Thampanya, U., Terrados, J., Gaertz-Hansen, O., Fortes, M.D., 1999. The determination of the age and growth of SE Asian mangrove seedlings from internodal counts. Mangroves Salt Marshes 3 (4), 251–257. Durako, M.J., 1994. Seagrass die-off in Florida Bay (USA) changes in shoot demographic characteristics and population dynamics in Thalassia testudinum. Mar. Ecol. Prog. Ser. 110, 59–66. Ebert, T.A., Williams, S.L., Ewanchuk, P.J., 2002. Mortality estimates from age distributions: Critique of a method used to study seagrass dynamics. Limnol. Oceanogr. 47 (2), 600–603. Ebert, T.A., Williams, S.L., Ewanchuk, P.J., 2003. Rejoinder to Fourqurean et al. Limnol. Oceanogr. 48 (5), 2074–2075. Fernández-Torquemada, Y., Sánchez-Lizaso, J.L., González-Correa, J.M., 2005. Preliminary results of the monitoring of the brine discharge produced by the SWRO desalination plant of Alicante (SE Spain). Desalination 182, 395–402. Fourqurean, J.W., Marbà, N., Duarte, C.M., 2003. Elucidating seagrass population dynamics: Theory, contrasts, and practice. Limnol. Oceanogr. 48 (5), 2070–2074. Francour, P., 1997. Fish assemblages of Posidonia oceanica beds at Port-Cross (France, NW Mediterranean): assessment of composi-
tion and long-term fluctuations by visual census. P.S.Z.N.: Mar. Ecol. 18 (2), 157–173. González-Correa, J.M., Bayle, J.T., Sánchez-Lizaso, J.L., Valle, C., Sánchez-Jerez, P., Ruiz, J.M., 2005. Recovery of deep Posidonia oceanica meadows degraded by trawling. J. Exp. Mar. Biol. Ecol. 320, 65–76. González-Correa, J.M., Bayle, J.T., Sánchez-Jerez, P., Valle, C., 2007. Posidonia oceanica meadows are not declining globally. Analysis of population dynamics in marine protected areas of the Mediterranean Sea. Mar. Ecol. Prog. Ser. 336, 111–119. Harper, J.L., Bell, A.D., 1979. The population dynamics of growth form in organisms with modular construction. In: Anderson, R.M., Turner, B.D., Taylor, L.R. (Eds.), Population dynamics. Symp. Br. Ecol. Soc., 20. Blackwell Scientific Publications, Oxford, pp. 29–52. Harriague, A.C., Bianchi, C.N., Albertelli, G., 2006. Soft-bottom macrobenthic community composition and biomass in a Posidonia oceanica meadow in the Ligurian Sea (NW Mediterranean). Estuar. Coast. Shelf Sci. 70, 251–258. Hemminga, M.A., Duarte, C.M., 2000. Seagrass Ecology. Cambridge University Press, Cambridge, p. 298. Jensen, S.L., Robbins, B.D., Bell, S.S., 1996. Predicting population decline: seagrass demographics and the reconstructive technique. Mar. Ecol. Prog. Ser. 136, 267–276. Leriche, A., Pasqualini, V., Boudouresque, C.F., Bernard, G., Bonhomme, P., Clabaut, P., Denis, J., 2006. Spatial, temporal and structural variations of a Posidonia oceanica seagrass meadow facing human activities. Aquat. Bot. 84, 287–293. Marbà, N., Duarte, C.M., 1998. Rhizome elongation and seagrass clonal growth. Mar. Ecol. Prog. Ser. 174, 269–280. Marbà, N., Duarte, C.M., Cebrian, J., Gallegos, M.E., Olesen, B., Sand-Jensen, K., 1996. Growth and population dynamics of Posidonia oceanica on the Spanish Mediterranean coast: elucidating seagrass decline. Mar. Ecol. Prog. Ser. 137, 203–213. Marbà, N., Duarte, C.M., Holmer, M., Martínez, R., Basterretxea, G., Orfila, A., Jordi, A., Tintoré, J., 2002. Effectiveness of protection of seagrass (Posidonia oceanica) populations in Cabrera National Park (Spain). Environ. Conserv. 29 (4), 509–518. Marbà, N., Duarte, C.M., Díaz-Almela, E., Terrados, J., Alvarez, E., Martínez, R., Santiago, R., Gacia, E., Grau, A.M., 2005. Direct evidence of imbalanced seagrass (Posidonia oceanica) shoot population dynamics in the Spanish Mediterranean. Estuaries 28, 53–62. Panayotidis, P., Boudouresque, C.F., Marcot-Coqueugniot, J., 1981. Microstructure de l'herbier de Posidonia oceanica (L.) Delile. Bot. Mar. 24, 115–124. Patriquin, D.G., 1973. Estimation of growth rate, production and age of the marine agiosperm Thalassia testudinum: seasonal and interanual variability. Aquat. Bot. 47, 1–11. Pérès, J.M., Picard, J., 1964. Nouveau manuel de bionomie benthonique de la Méditerranée. Recueil des Travaux de la Station Marine d´Endoume 31, 1–137. Pergent, G., 1987. Recherches lépidocronologiques chez Posidonia oceanica (Potamogetonaceae). Fluctuations des paramètres anatomiques et morphologiques des écailles des rhizomes. PhD thesis, University Aix–Marseille II, Marseille. Pergent, G., Pergent-Martini, C., 1991. Leaf renewal cycle and primary production of Posidonia oceanica in the bay of Lacco Ameno (Ischia, Italy) using lepidochronological analysis. Aquat. Bot. 42, 49–66. Pergent-Martini, C., Rico-Raimondino, V., Pergent, G., 1994. Primary production of Posidonia oceanica in the Mediterranean Basin. Mar. Biol. 120, 9–15.
Author's personal copy
J.M. González-Correa et al. / Journal of Experimental Marine Biology and Ecology 353 (2007) 115–125 Peterson, J.B., Fourqurean, J.W., 2001. Large-scale patterns in seagrass (Thalassia testudinum) demographics in south Florida. Limnol. Oceanogr. 46 (5), 1077–1090. Sánchez Lizaso, J.L., 1993. Estudio de la pradera de Posidonia oceanica (L.) Delile de la reserva marina de Tabarca (Alicante): fenología y producción primaria. PhD thesis, Universidad de Alicante, Alicante, Spain. Sánchez-Jerez, P., Barberá-Cebrián, C., Ramos-Esplá, A.A., 1999. Comparison of the epifauna spatial distribution in Posidonia oceanica, Cymodocea nodosa and unvegetated bottoms: importance of meadow edges. Acta Oecol. 20, 391–405.
125
Short, F.T., Duarte, C.M., 2001. Methods for the measurement of seagrass growth and reproduction. In: Short, F.T., Coles, R.G. (Eds.), Global seagrass research methods. Elsevier, Amsterdam, pp. 155–182. Tomlinson, P.B., 1974. Vegetative morphology and meristem dependence the foundation of productivity in seagrasses. Aquaculture 4, 107–130. White, J., 1979. The plant as a metapopulation. Ann. Rev. Ecolog. Syst. 10, 1098–1145.
[SS]
