Factors affecting seaweed biogeographical and ecological trends

Factors affecting seaweed biogeographical and ecological trends

Factors affecting seaweed biogeographical and ecological trends along the N am ibian coast H R Engledow1 & JJ Bolton 'Phycology Research Group, Depart...

413KB Sizes 0 Downloads 3 Views

Recommend Documents

Ecological Factors Affecting Coral Reef Assemblages Off Hurghada
Three shallow reef sites were investigated during four seasons (Autumn, 2003, Winter,. Spring and Summer, 2004) off the

FACTORS AFFECTING CAPACITY
Psychosocial Factors Affecting Capacity –Psychosocial factors include one's surroundings, way of life & things that af

FACTORS AFFECTING FUTURE RESULTS
petrochemical prices and by changes in margins on refined products. Oil, gas, petrochemical, and product prices and marg

Factors affecting prostitution - Ruhama
explains that in relation to other working class women's incomes in Canada, women in prostitution appear to be earning h

International LSD Prevalence—Factors Affecting Proliferation and
In 1968–1969 the Windsor, California lab of Nick Sand and Tim Scully ... after serving 3-1/3 years due to community se

Factors Affecting International Mergers and Acquisitions - IFAMA
The purpose of this paper is to investigate the factors that explain outward and inward merger and acquisition (M/A) act

Molecular and microstructural factors affecting mechanical properties
General deformation for a polymer: stress-strain curve as observed in one material as a function of temperature of rate

identification and understanding of factors affecting - ScholarSpace
Thanks to the Keala family on Maui for providing me with a loving foundation of ..... heat loss occurs when the ambient

Environmental Factors and Production Characteristics Affecting the
LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, ple

Factors affecting seaport capacity - MSSANZ
Malcom McLean in 1956 (Cudahy, 2006). Containerization brings not only opportunities, but also restrictions to the marit

Factors affecting seaweed biogeographical and ecological trends along the N am ibian coast H R Engledow1 & JJ Bolton 'Phycology Research Group, Department o f Biology, Ghent University, Krijgslaan 281/S8, 9000 Ghent, Belgium Botany Department, University o f Cape Town, Private Bag, Rondehosch, 7701 South Africa 1E-mail: henry. [email protected] he K ey words: Namibia, biogeography, sand

Abstract Namibia forms the northern part of the Benguela marine province in south-western Africa. A splitting of the province into two sub-provinces has been proposed on zoogeographical evidence. This split is the result of decreasing species diversity in the northern regions, and is therefore not a taxonomic division. A number of factors have been found to affect seaweed biogeography on the Namibian coast, e.g. temperature factors, habitat availability and heterogeneity, sand inundation, etc. These factors also have an effect the local occur­ rence and abundance of certain species. Biomass sampling quadrats were taken at regular intervals up the shore and the species composition determined. Southern Namibia is separated from northern and central Namibia, due to the dominance of different species. Southern and northern Namibia had a greater horizontal spread on the primary axis of the DCA analysis, reflecting their greater habitat availability and heterogeneity. The shores of central Namibia, in contrast, exhibit minimal variation between shores. The difference between southern and central Namibia is an ecological shift brought about by an increase in temperature in the northern regions and a relatively high degree of sand inundation. The latter has result­ ed in the dominance of sand-tolerant species. As a result, the sub-province is an ecological entity rather than a floristic one.

Introduction Namibia comprises the northern half of the Benguela marine province (Engledow et al., 1992; Bolton & Anderson, 1997), which stretches from Moçamêdes (southern Angola, 15° 10’ S) to Cape Agulhas (southern-most tip of Africa). This region has also been referred to as the south-western African marine province (Lawson et al., 1990; Lüning, 1990; Field & Griffiths, 1991), or as the ‘West Cape’ by Hommersand (1986). Stephenson (1948), who looked at the distribution of 318 common intertidal species (a third of these species being algae) around the South African coast, referred to this region as the cold temperate west coast marine province. Emanuel et al. (1992) pro­ posed splitting the Benguela province into two dis­ tinct biogeographical units based on the results of

a zoogeographical multivariate analysis of inverte­ brate species presence/absence data. They separated a cool temperate southwest coast (Namaqua province) from a cool temperate northwest coast (Namib province), with Lüderitz marking the bor­ der between these two provinces. A similar pattern was observed by Engledow et al. (1992), using sim­ ilar species presence/absence data for seaweeds, but the interpretation of the results differed. The sepa­ ration of central and northern Namibia from the rest of the west coast was largely due to a decrease in species numbers in these regions and not a floris­ tic change. The Benguela marine province may be divided up as follows: (1) a southern coast transitional region (False Bay — Cape Agulhas), between the Benguela and Agulhas marine provinces; 285

SEAWEED B IO G E O G R A PH IC A L A N D E C O L O G IC A L T R E N D S IN NAMIBIA

Methods

1992. A total of seventeen sites was chosen in southern (SNAM), central (CNAM) and northern (NNAM) Namibia (Fig.l); nine of which were sampled over two seasons (i.e. 1990 & 1992; Table 1). Two to three transect lines were located at each site from spring low-water mark to spring highwater mark. All material was removed from 25 cm X 25 cm quadrats at 1 m intervals on each transect, sorted and the biomass of the various species meas­ ured. Details of the method of collection are set out in Engledow et al (1992). To establish site related­ ness, the average biomass of each species was esti­ mated for each shore. This was done by the sum­ mation of the biomass values for each species with­ in all the quadrats within a particular site, divided by the total number of quadrats at that site. The data were then subjected to a detrended correspon­ dence analysis (DCA - Ter Braak, 1991) to reveal site relatedness.

The data used in this analysis comprise seaweed biomass data collected in spring 1990 and autumn

Results

(2) a characteristic cool5 water adapted south­ western coast region (southern Namibia southwestern Cape); which may be further split into two sub-provinces in the future, but the evidence at this stage is not conclusive; and (3) a depauperate warm5water adapted north-west coast region (northern Namibia — central Namibia). Here, we take a closer look at the biogeographi­ cal trends on the Namibian coast and the factors responsible for them. Not only are broad scale factors such as sea surface temperature important for observed distribution patterns, but also smaller scale or local factors e.g. sand inundation, habitat avail­ ability and heterogeneity. These factors may also have an effect on which species become dominant.

As can be seen from Figure 2, the Namibian shores separate into the three main regions defined. The ANGOLA Table 1. Coding of sites sampled along the Namibian coast in sequence from south to north. Sites in bold were sampled in 1990 (Spring) & 1992 (Autumn).

C unene T^iver M o u th r NNAM

F alse C ape F r i a ^

NAMIBIA

Möwe Bay'

Sw akopm undi Walvis Bay

Region

Code

Site

SNAM

ELB GSB KBF ES B DZP SKI L&A LGS V IN MY4 MY8 MOW TOS TOR TER RPT FCF

Elizabeth Bay Grossebucht Kleinbogenfels Essy Bay D iaz P oint Shark Island Rocks near Agate Beach Langstrand V iñeta Beach Myl 4 M yl 8 M öw e Bay Toscanini Torra Bay Terrace Bay Rocky Point False Cape Fria

Tropic of

CÑÁM

C apricorn

Lüderitz SNAM

O range>Z ^^^^

SOUTH \ a f r ic a

CNAM

C ape Town I r ^ _ False B a y ^ - ^ C ape A gulhas

N N AM

River M outh^ L nw c

swc

Figure 1. Map of Namibia showing the areas sampled (hatched bands) and regions delineated (See Table 1 for order of sites with­ in the bands).

286

SEAWEED B IO G E O G R A P H IC A L A N D E C O L O G IC A L T R E N D S IN NAMIBIA

325 •FCF

275 -

•TOR

SNAM oÎ 2 2 5 -

NNAM ^ .esb. g s b > IPZ' GSB2«.KBF . SK, 5 L&A #L&A2 1DPZ2

125

V

.T O R ? "

*MY82

•RPT

•ELB

-

*T E R

•TER2

•VIN2

CNAM

•VIN, •LGS2

75 25 -

■MY4

-25 -50

50

1 50 250 AXIS 1 (Eigen value 0.76)

350

separated on Axis 2. The central Namibian sites are closely vertically aligned, showing very little differ­ ence with respect to Axis 1, but separate on Axis 2. The Skeleton coast sites are more spread out and diverse. Toscanini on the Skeleton coast has a high degree of similarity with Myl 8 from Swakopmund. Most of the sites that were sampled in both years show very little difference between them, with the exception of Torra Bay. The Skeleton coast is spread out on both axes.

southern Namibian sites show a close affinity to one another, with the exception of Elizabeth Bay. The southern sites are also clearly separated from the central and northern regions. The sites in southern and northern Namibia are spread hori­ zontally, parallel to Axis 1, showing that they lie on some gradient. The central Namibian sites form an almost cohesive group, with the Myl 8 (1992) site forming part of the Skeleton coast group. The only other outlier is Myl 4, which closely aligns with False Cape Fria on the Axis 1, although they are

swc Warm Temperate Agulhas Flora

NW

SNAM

450

Figure 2. DCA of average bio­ mass for the various sites sam­ pled during 1990 and 1992. (Letter abbreviations followed by the number ‘2 ’ refer to the samples collected in Autumn 1992, all other samples were collected in Spring 1990).

CNA

NNA Tropical Flora

Cool Temperate Benguela Flora

Too Cold < - ■ -

Level

T hickness = Temperature Range S lop e = Relative Temperature

Biogeographic Constraints

■ ! Î A

Level Temperature Constraints Seasonal < ----------

Hypothetical S p e cies Temperature tolerance range

, Non S easonal Stochastic Warm

Level Habitat Availability & Heterogeneity

Level Local Factors Affecting S p ecies

I Sand Inundation, I Scouring & Hot, Dry East

Figure 3. Schematic diagram modeling the major factors affecting seaweed species rich­ ness along the Benguela marine province.

287

SEAWEED B IO G E O G R A PH IC A L A N D EC O L O G IC A L T R E N D S IN NAMIBIA

Discussion A number of factors are responsible for the observed decline of species in a south-north direc­ tion. Hommersand (1986) attributed this partly to lack of collected material, and this is partially true, but sufficient material has been collected in this study to validate the general trend. Lawson et al. (1990) proposed a decrease in wind stress and the related water circulation patterns, but they do not elaborate on how this would affect species richness. A similar pattern of decreasing species richness in a northward direction has been observed on the west coast of South America (Santelices, 1980). This was attributed to the cool influence of the Peru current and temperature increases in the northern reaches due to the southward incursions of sub­ tropical surface water or stochastic water (El Niño). The biogeographical distribution and ecological composition of species may be explained by look­ ing at a hierarchical system of factors or scales. For example, sea surface temperature will have a domi­ nant role to play in determining the presence of many species, while the effects of sand inundation have a more local effect on distribution and domi­ nance. The biogeographical patterns, species diver­ sity and ecological composition observed on the Namibian coast may be explained by the following model based on Engledow et al. (1992) (Fig. 3): Level 1: there is a sudden transition in sea sur­ face temperature between Namibia and Angola. As a result of this rapid transition, few tropical sea­ weed species can establish in Namibia and vice versa, unless they are eurythermal. The temperature changes from 18—21°C (Nature Conservation sta­ tion at Möwe Bay; Fig. 1) in northern Namibia, to 24°C in the north of Moçamêdes in southern Angola based on the summer isotherm (Lawson et al., 1975). This is the result of the confluence of the cool northward flowing Benguela current and the warm southward flowing Angola current. The likelihood of stenothermal species migrating between these two regions is minimal due to the temperature differences. Also, the general sea cur­ rent direction on the Namibian coast is northerly, which would probably reduce the likelihood of sea­ weeds migrating south. As a result, the northern boundary of the Namibian seaweed flora is relatively 288

well defined. Level 2: Not only is there a temperature discon­ tinuity at the northern extreme, but there is an increase in temperature at Walvis Bay (Bolton, 1986). Mean annual temperatures of the southern regions of the Benguela province are approximate­ ly 12—13°C with a range of 11.5—14°C in mean monthly temperature (Bolton, 1986), while in the Walvis Bay area, the annual mean is approximately 16°C with a range of 12-18.4°C (Molloy, 1990). Bolton (1996) hypothesized that a relatively nar­ row temperature range and a long geological peri­ od of prevailing warm temperate conditions would result in high diversity due to the coexistence of both stenothermic and eurythermie species. This change in temperature in central Namibia has probably resulted in halting the spread of more ‘cool’ water-adapted stenothermal species from venturing further north. This would account for the sudden decline in species richness in the central and northern reaches of Namibia. The shift in tem­ perature range may also influence species domi­ nance. Möwe Bay in Northern Namibia is topo­ graphically similar to sites in southern Namibia. The species that dominate this site are however dif­ ferent from those in the south, eg geniculate coralline algae, Centroceras clavulatum (C. Agardh) Montagne, Caulacanthus ustulatus (Turner) Kützing. These seaweeds are found further south, but are less dominant. They re-emerge as a more dom inant component on the warmer South African south coast (the Agulhas marine province). Unlike the northern reaches of the Benguela marine province, the southern part of the province abuts a warm temperate province (the Agulhas marine province). As a result of warmer tempera­ tures, some species from the Agulhas province are able to survive in southern part of the Benguela marine province for several reasons. The latter area is very heterogeneous with respect to its physical environment, having a diversity of microhabitats or sites (for example, not all areas experience extensive upwelling, e.g. Table Bay, Cape Town). Another reason may be that the species that do live in these southern parts are eurythermal or at their temper­ ature limits, and being near a propagule source (viz False Bay, which has a diverse flora due to its wide

SEAWEED B IO G E O G R A PH IC A L A N D EC O L O G IC A L T R E N D S IN NAMIBIA

temperature gradient) are able to recolonize an area if it becomes locally extinct. The Namibian coast is subject to upwelling of cool central water, but, unlike the southern part of the flora, it is non-seasonal (Shannon, 1985). This is particularly prevalent in Lüderitz, and, to a less­ er extent, further north. The seasonal temperature fluctuations are therefore more defined in the southern reaches. The coastal waters also receive counter-current tongues of warm water from the Angolan current, as well as warm water events caused by the El Niño phenomenon. These may increase the sea surface temperature up to 21 °C, while parts of Lüderitz lagoon may reach tempera­ tures of 28°C for periods of one to three weeks. These warm water events may serve as a limiting factor to the spread or establishment of various algae that could potentially occur here. A similar pattern has been noticed on the temperate coast of Pacific South America (Santelices, 1980). The Namibian coast is a less stable situation than that found on the west coast of South Africa, which has a much narrower temperature range and seasonal variation (Bolton, 1986). As a result, seaweeds along this coast are exposed to cool water for most of the year, with occasional warm water events that are tropical in nature. The sea temperature is there­ fore generally too cold to allow the establish­ ment/survival of tropical species and occasionally too warm for relatively ‘stenothermal’ warm tem­ perate species, thus limiting the geographical spread of South African west coast and Angolan species. Level 3: There is a decrease in habitat hetero­ geneity on a micro and macro scale as one moves northwards. Lüderitz, in southern Namibia, com­ prises a lagoon, a number of bays, numerous inlets and extensive rocky outcrops ranging from shel­ tered to extremely exposed. As one moves north of Lüderitz, there is a tailing off of the above, and the coast becomes relatively straight with Sandwich Harbour and Walvis Bay being the only inlets in the central reaches of the country (Branch & Griffiths, 1988). The Skeleton coast consists of long sandy beaches with small rocky outcrops, which are often completely submerged during neap tides. There are, however, a few relatively big rocky

outcroups that are not completely covered by spring high water eg Möwe Bay, Rocky Point, False Cape Fria and Cape Fria. As a result, the central and northern parts of Namibia are exposed to severe wave action with very few or no sheltered habitats. Likewise, on a micro-scale, there is a plethora of habitats in the south as opposed to the north. The rocky outcrops in southern Namibia usually extend above the spring high water mark and are more diverse with respect to texture, type, morphology and aspect; having numerous rock pools (at all tidal levels), gullies of various dimen­ sions, rocks of various sizes, etc. This diversity is present, to a lesser extent, further north. Thus there are a greater variety of habitats available in the south, allowing for the establishment of a wider range of species. The above argument may also be used to explain the ‘difference’ between Bolton & Anderson’s (1997) sub-provinces, viz. Namaqua and Southwestern Cape. The latter region, as a per­ centage of the total coastline, has greater habitat heterogeneity and substrate availability than the former, which may give rise to the observed greater species richness in the south. As mentioned earlier, however, this species difference may not be as pro­ nounced as it appears, even though the drop in species number in the northwestern Cape is prob­ ably a real one. Southern Namibia is reasonably habitat heterogeneous and it has a relatively sub­ stantial rocky shoreline (although not as much as southwestern Cape). Another component which has not been considered is the offshore islands of southern Namibia, which represent further habitat heterogeneity and which have not been extensively researched (with respect to seaweeds or intertidal communities). There is not only a decrease in habitat hetero­ geneity with a decrease in latitude, but also a decrease in habitat availability. In the central and northern regions of Namibia, the rocky out-crops are small, few and separated by long expanses of sandy beaches. The rocky shores would therefore act as small islands, and the sand a barrier to dis­ persal, ie not a suitable substrate for the attachment of seaweeds. Level 4: Other local factors may also affect sea­ 289

SEAWEED B IO G E O G R A PH IC A L A N D EC O L O G IC A L T R E N D S IN NAMIBIA

weed distribution. Sand/silt inundation of rocky shores is more prevalent in the central and north­ ern reaches of the Namibian coast. One would expect to find sand-tolerant species (eg Mazzaella capensis (J.Agardh) Fredericq — Bolton & Joska, 1993) and the exclusion of sand-intolerant species (.Aeodes orbitosa (Suhr) Schmitz —Bolton & Levitt, 1992). Mazzaella capensisâ able to ‘resprouf from its crustose base after a sand inundation event, while Aeodes orbitosa has no such persisting struc­ ture. Sand inundation, together with increased temperature conditions and possible related increases in herbivory, could account for the pre­ dominance of turf-forming seaweeds (eg Chylocladia capensis Harvey, Ophidocladus simpli­ ciusculus (Crouan & Crouan) Falkenberg in Engler & Prantl, Cladophora spp., Caulacanthus ustulatus (Turner) Kützing, Polysiphonia spp., Tayloriella tenebrosa (Harvey) Kylin) in the central and north­ ern parts of the coast. The potential exists for mass destruction of sea­ weeds and other intertidal organisms during east wind conditions on the coast of Namibia. This wind is hot and dry, and blows off the desert onto the rocky shore. If these winds coincide with low tide, they may have potentially devastating effects on the intertidal organisms. The bleaching of many intertidal seaweeds has been noted during such an event in Swakopmund (central Namibia). The severe desiccating effect of a similar wind in south­ ern California was noted by Seapy & Littler (1982). The effect of this wind may have a pro­ found effect on the intertidal diversity and species richness in that it may cause ‘local extinction of species. This, coupled with the fact that potential propagule sources, viz. close unaffected rocky shore, are relatively rare in the central and northern Namibian regions, means that recolonization of these shores is severely restricted. The reduced species richness of the central and northern reaches of Namibia can be attributed to the above. Temperature in conjunction with habi­ tat (including wave action) and propagule source would set the range of species that could potential­ ly survive in the given region. This range is further influenced by the species ability to tolerate events such as sand-inundation, warm water intrusion 290

and low tide east wind events. As was seen in Figure 2, southern Namibia is separated from northern and central Namibia. This is due to the dominance of different species in these regions. As was described above, there is a rise in temperature in the central region of Namibia, which would account for some of the observed shift. Also, southern and central Namibia were both found to be spread out on the primary axis. This may be explained by the increased habitat availability and heterogeneity found in these regions; while central Namibia lies virtually on one line on the primary axis and separated on the sec­ ondary axis. The latter is due to the fact that the species composition on these shores is very similar, and only the biomass values for the various species vary, giving the observed dispersion on the second­ ary axis. Central and northern Namibian sites are however more closely aligned to one another than those of southern Namibian. The Namibian coast may thus be divided into two main areas: (1) a southern region, which is similar to commu­ nities of South African west coast shores; and (2) a central and northern region. The latter region has a predominance of turfforming, ‘resprouting’ (those species with a persist­ ent crustose base) and opportunistic algal species. These two regions represent an ecological shift and not a floristic change driven by the factors outlined above, in particular: temperature, degree of sand inundation, and possibly wave exposure. A number of species are excluded from the central and north­ ern regions of Namibia on the basis of the rise in temperature. Other species are then able to become dominant because: (1) there is the absence of dominant/late successional species further south; (2) they are better adapted to the increased tem­ perature; and or (3) they are able to cope with a moderate to high degree of sand inundation on most shores. The changes between their various sites show a prevailing conditions. As species may be implicated

the three regions and species response to the a result, a number of as being environmental

SEAWEED B IO G E O G R A PH IC A L A N D EC O L O G IC A L T R E N D S IN NAMIBIA

indicators for particular conditions along this coastline.

Acknowledgements The authors would like to thank the National Research Foundation, Departm ent of Environment and Tourism and the University of Cape Town for support. We would also like to thank the Namibian department of Sea Fisheries for their cooperation with respect to permits and use of facilities in Lüderitz and Swakopmund.

References Bolton JJ (1986). Marine phytogeography of the Benguela upwelling region on the west coast o f southern Africa: a tem­ perature dependent approach. Bot. Mar. 29 : 251—256. Bolton JJ (1996). Patterns of species diversity and endemism in comparable temperate brown algal floras. Hydrobiologia 326 / 327 : 173-178. Bolton JJ & Anderson RJ (1997). Chapter 15: Marine vegetation. In: Vegetation of Southern Africa. Cowling RM, Richardson DM & Pierce SM (Eds), Cambridge University Press. 348—370 pp. Bolton JJ & Joska MAP (1993). Population studies in a South African carrageenophyte Iridaea capensis (Gigartinaceae, Rhodophyta). Hydrobiologia 260/261: 191—195. Bolton JJ & Levitt GJ (1992). South African west coast carrageenophytes. In: Proceedings o f the First International Workshop on Sustainable Seaweed Resource Development in Sub-Saharan Africa. Mshigeni KE, Bolton JJ, Critchley A & Kiangi G (Eds), March 22—29, 1992, Windhoek, Namibia. 37—49 pp. Branch GM & Griffiths CL (1988). The Benguela ecosystem Part V The coastal zone. Oceanogr. Mar. Biol. Annu. Rev. 26: 395-486. Emanuel BP, Bustemante RH, Branch GM, Eekhout S & Odendaal FJ (1992). A Zoogeographie and functional approach to the selection of marine reserves on the west coast of South Africa. S. Afr. J. Mar. Sei. 12: 341-354.

Engledow HR, Bolton JJ & Stegenga H (1992). The biogeogra­ phy of the seaweed flora of Namibia. In: Proceedings of the First International Workshop on Sustainable Seaweed Resource Development in Sub-saharan Africa. Mshigeni KE, Bolton JJ, Critchley A & Kiangi G (Eds), March 22—29, 1992, Windhoek, Namibia. 117—130 pp. Field JG & Griffiths CL (1991). Littoral and sublittoral ecosys­ tems of southern Africa. In: Intertidal and Littoral Ecosystems. Mathieson AC & Nienhuis PH (Eds), Elsevier, Amsterdam. 323—346 pp. Hommersand M H (1986). The biogeography of the South African marine red algae: a model. Bot. Mar. 29 : 257—270. Lawson GW, John DM & Price JH (1975). The marine algal flora of Angola: its distribution and affinities. Bot. J. Lin. Soc. 70 : 307-324. Lawson GW, Simons RH & Isaacs WE (1990). The marine algal flora of Namibia: Its distributions and affinities. B. Brit. MusfBot.) 20 : 153-168. Lüning K (1990). Seaweeds: Their environment, biogeography, and ecophysiology. C. Yarish & H. Kirkman (Eds). Wiley-Interscience Publication, New York. 527 pp. Molloy F (1990). Utilized and potentially utilizable seaweeds of the Namibian coast: Biogeography and accessibility. Hydrobiologia 204 / 205 : 293—299. Russell G (1991). Vertical distribution. In: Intertidal and littoral ecosystems. Mathieson AC & Nienhuis PH (Eds). Elsevier, Amsterdam. 43—65 pp. Santelices B (1980). Phytogeographic characterization of the tem­ perate coast of Pacific South America. Phycologia 19: 1—12. Seapy R. & Litder MM (1982). Population and species diversity fluctuations in a rocky intertidal community relative to severe aerial exposure and sediment burial. Mar. Biol. 71 : 87—96. Shannon LV (1985). The Benguela Ecosystem Part 1. Evolution of the Benguela, Physical features and processes. Oceanogr. Mar. Biol. Annu. Rev. 23 : 105—182. Stephenson TA (1948). The constitution of the intertidal fauna and flora of South Africa. Part III. Ann. Natal Mus. 11: 207-324. Ter Braak CFJ (1991). Canoco: A Fortran Programme for Canonical community ordination by partial detrended canoni­ cal correspondence analysis and redundant analysis (ver. 3.12). Vageningen Agricultural Mathematical Group.

291