Aquatic Botany New2014

Aquatic Botany 116 (2014) 93–102 Contents lists available at ScienceDirect Aquatic Botany journal homepage: www.elsevier.com/locate/aquabot Importance of seedling recruitment for regeneration and maintaining genetic diversity of Cyperus papyrus during drawdown in Lake Naivasha, Kenya Taita Terer a,b,∗ , A. Muthama Muasya c , Sarah Higgins d , John J. Gaudet e , Ludwig Triest a a Plant Biology and Nature Management, Vrije Universiteit Brussel (VUB), Pleinlaan 2, B-1050 Brussels, Belgium Wetland and Marine Section, Centre for Biodiversity Department, National Museums of Kenya, P.O. Box 40658-00100, Nairobi, Kenya c Department of Biological Sciences, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa d Lake Naivasha Riparian Association, P.O. Box 358, Naivasha 20117, Kenya e Ecology Consultant, 5825 Bent Twig Rd., McLean, VA 22101, USA b a r t i c l e i n f o Article history: Received 18 December 2012 Received in revised form 4 February 2014 Accepted 14 February 2014 Available online 26 February 2014 Keywords: Cyperus papyrus Drawdown Microsatellite Macrophytes Hydrochory Lake Naivasha a b s t r a c t Drawdown-flooding cycles occur commonly in many water bodies and influence plant succession in many ways including the expansion of macrophytes through seedling recruitment. We investigated a regeneration event of Cyperus papyrus in Lake Naivasha and documented seed production, seedling recruitment, zonation progression, dispersal and establishment in relation to evidence from genetic relatedness between seedlings and mature stands using microsatellite loci. Seed estimate counts in five papyrus umbels reached high values between 98,000 and 337,000. The drawdown drying phase led to desiccation accompanied by cracking of mudflat soils, and the oxidation and demise of littoral aquatic plants. Reflooding led to distinct zones of young papyrus and hygrophilous ephemerals. In the final flooding phase when the lake reached normal water levels, hygrophilous ephemerals died off while papyrus survived. Young floating papyrus mats dispersed through wind and wave action joined existing mature stands or spread into formerly unoccupied shoreline areas. Microsatellite analysis of seedlings in the drawdown zone and a neighboring stand of mature and juvenile shoots reveals high overall gene diversity (Ho = 0.476, He = 0.576, Ae = 2.8) reflecting an underlying sexual reproduction. A large overlap of genotypes was found in seedling and parent stands indicating a single gene pool. A total of 40 alleles were observed across 3 life stages (clonal juvenile, mature and seedlings), however, more private alleles and higher allelic diversity were detected in seedlings than in the parent individuals, showing their contribution to an increase in the local gene pool. Fine-scaled spatial genetic structuring was detected at about 100 m distance in parent stands for both juvenile and mature life stages indicating a potential influence caused by local seed rain. © 2014 Elsevier B.V. All rights reserved. 1. Introduction The Maasai historical local name, Enaiposha (meaning ‘receding water’) is an indication that throughout history Lake Naivasha has undergone large-scale water level fluctuations. Receding water resulting in drawdown (a drop of lake level), later followed by flooding, is a common feature of tropical wetlands and shallow lakes ∗ Corresponding author at: Wetland and Marine Section, Centre for Biodiversity Department, National Museums of Kenya, P.O. Box 40658-00100, Nairobi, Kenya. Tel.: +254 3742161/4; fax: +254 2 3741424. E-mail addresses: taaita@yahoo.com, tterer@museums.or.ke (T. Terer), muthama.muasya@uct.ac.za (A.M. Muasya), kijabe@africaonline.co.ke (S. Higgins), JJGaudet@aol.com (J.J. Gaudet), ltriest@vub.ac.be (L. Triest). http://dx.doi.org/10.1016/j.aquabot.2014.02.008 0304-3770/© 2014 Elsevier B.V. All rights reserved. and has elicited the attention of many biologists (e.g. Gunn, 1973; Howard-Williams, 1975; Gaudet, 1977; Harper, 1992; Verschuren et al., 2000; van der Valk, 2005; Bond et al., 2008; Terer, 2011). For wetland plants that have evolved to live in alternately wet and dry habitats, drawdown can be a very positive event that promotes regeneration by seedling recruitment (Gaudet, 1977; van der Valk and Davis, 1978; Welling et al., 1988). Cyperus papyrus (Cyperaceae), the focus of the current study, is known to be adapted to normal climatic cycles of drying-flooding, and exhibits both sexual reproduction and unlimited clonal propagation (Gaudet, 1977; Terer, 2011; Triest et al., 2013). The impact of drawdown-flooding cycles on hydrobiology especially in plant community succession is well studied (e.g. Gaudet, 1977; Wilcox and Meeker, 1991; Harper, 1992; Van Geest et al., 2005a,b), however, few studies relate such cycles to the role they 94 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 play in genetic diversity of aquatic species by seedling recruitment (Shibayama and Kadono, 2007; Uesugi et al., 2007). Available ecological information on Cyperus papyrus shows that it has a persistent seed bank (Gaudet, 1977; Boar, 2006). One important factor that could be expected to influence the extent of C. papyrus establishment is intensity of the seed rain onto the lakeshore sediment seed bank, yet this information is unavailable. Identification of factors that determine the recruitment rates in aquatic plants and longevity of seed banks are fundamental to the understanding of population dynamics especially those in recovery after long drought events (Smith and Kadlec, 1983; Brock and Rogers, 1998; de Winton et al., 2000; Shibayama and Kadono, 2007). Maintenance of high genetic diversity and high gene flow exhibited by C. papyrus in Lake Naivasha (Triest et al., 2013) imply that such diversity and flow can only occur when seeds/propagules are deposited on suitable microsites and are allowed to germinate and establish. In this process recruitment by seeds (hence successful sexual reproduction vis-a-vis clonal reproduction) is an indispensable mechanism to the long-term success of plant species (Santamaria, 2002). It is arguable that clonal propagation is not a substitute for sexual reproduction but merely prolongs the time to extinction when sex is absent for plant species (Honnay and Bossuyt, 2005; Silvertown, 2008). Making inferences on the life history survival strategies of C. papyrus can best be done during long drought events such as those experienced in the last ten centuries in eastern Africa (Verschuren et al., 2000). It is expected that as wetlands recede during such drought episodes, existing papyrus populations will be faced with severe water shortage so that some populations go extinct. In such circumstances, it is only the existence of a seed bank that will allow recuperation of the population after rewetting. In the event that these seed banks and seedlings are destroyed by human activities, the recovery of a population would have to rely on dispersal from other undisturbed populations. In 1960s papyrus dominated the littoral zone of Lake Naivasha, but reduced 10-fold (from ca. 50 km2 to ca. 5 km2 , Hickley et al., 2004) by 2005. The once continuous papyrus population is now highly fragmented due to processes driven by climate cycles and anthropogenic activities such as agricultural reclamation and grazing by domestic livestock (Hickley et al., 2004; Gichuki et al., 2006; Morrison and Harper, 2009), all of which impact negatively on regeneration by seedlings. An understanding of the connection between drawdown-flooding phenomena and papyrus population dynamics is crucial in putting in place conservation measures to reverse the loss of papyrus, which is known to provide the lake with an efficient and natural buffer against sediment and nutrient loading (Boar et al., 1999). Papyrus also provides for carbon sequestration, breeding grounds for fish and a habitat for threatened birds (Mnaya and Wolanski, 2002; Jones and Humphries, 2002; Terer et al., 2006; Owino and Ryan, 2006; Saunders et al., 2007; IUCN, 2010; Harper et al., 2011). In addition, papyrus is of great use to riparian people as a source of handicraft materials, and as fuel, fodder for livestock, medicine and roofing materials as well as socio-cultural uses (Jones, 1983; Gichuki et al., 2001; Simpson and Inglis, 2001; Terer, 2011; Terer et al., 2012a). Seedling recruitment, dispersal and the subsequent areal expansion of the papyrus swamp in Lake Naivasha has been reported in the past (Gaudet, 1977; Harper, 1992; Boar, 2006). The current study explores the importance of drawdown-flooding cycles in regeneration of C. papyrus in relation to seed production, seed bank, seedling recruitment and the link between dispersal and maintenance of genetic diversity. The aim of this study is to investigate the role of drawdownflooding in the regeneration of C. papyrus and the contribution of seedlings in maintaining lake population gene pools. Specifically, we intend to: (1) provide information on seed production of C. papyrus; (2) describe the most recent drawdown-flooding progression including plant zonation; (3) document the natural regeneration of C. papyrus by seedling recruitment, and dispersal; and (4) provide genetic information on the seedlings vs. parent populations with a view of demonstrating the importance of seedling recruitment in enriching and maintaining lake population gene pools. 2. Methods and materials 2.1. Study site Our study was carried out in Lake Naivasha (Kenya), a freshwater body on the floor of Eastern Great Rift Valley, occupying an endorheic basin with no surface outflow but underground seepage (Gaudet and Melack, 2006). The work was done between February 2008 and January 2012 in two papyrus shoreline swamps in the southern part of the Lake: Fisherman’s Camp (0◦ 49′ S 36◦ 20′ E, altitude 1902 m) and Kamere Public Beach (0◦ 48′ S 36◦ 19′ E, altitude 1901 m). The Fisherman’s Camp site was chosen because of its accessibility and undisturbed status on private land utilized for ecotourism (Terer et al., 2012b). The Kamere site, which is open to public access, was chosen because of its known population dynamics of local extinction and recovery (Terer, 2011). It should be noted that the Kamere site was devoid of papyrus during the drawdown period, thus it was not sampled for genetic analysis, although observations were made on the later colonization of that site by papyrus. 2.2. Seed production counts and data analysis We estimated seed production from only five randomly cut mature umbels from the Fisherman’s Camp site. The small sample size was chosen to avoid negative impacts on population size and deemed representatives. However the umbels were comparable in size, shape and form to most umbels at this and other sites on the lake. The cut umbels were packed tightly in polythene bags to avoid losing mature seeds. We counted the total number of rays (Fig. 1a) contained in an umbel, and then randomly selected 10 rays for further investigation from each umbel. From the selected 10 rays, we counted the number of spikes (Fig. 1b and c), spikelets (Fig. 1d), and the number of seeds (Fig. 1e) recovered from 50 spikelets from each umbel (total of 250 spikelets). The spikelet was randomly selected along the spike from top to bottom to account for their differences in length and hence the number of seeds contained in them. Fig. 1f and g shows the papyrus stand at this site and a representative papyrus umbel. The total number of seeds (ST ) for each umbel was estimated by average number of seeds recovered from 50 spikelets (SM ) multiplied by the average number of spikelets counted from 10 rays (SPM ), the obtained value was then multiplied by the total number of rays recorded for each umbel (RT ), simply expressed as: ST = (SM ∗ SPM ) RT . Images of general morphology of seeds and their position in a spikelet were taken with a Scion camera Model CFW-1310C mounted on an Olympus Stereoscope (Scion Corp.). 2.3. Drawdown succession and seedling density Repeat photography was used to document the drawdown area at Fisherman’s Camp starting with the dry phase (August 2009) until when it was reflooded (September 2010). At this site we also captured periodic regeneration of C. papyrus seedlings and the succession of aquatic and hygrophilous species (i.e., short-lived plants that die after re-flooding: Gaudet, 1977) until the C. papyrus seedlings had grown to maturity in September 2010. Observation T. Terer et al. / Aquatic Botany 116 (2014) 93–102 95 Fig. 1. Illustrated parts of papyrus umbel: a – ray, b – spike, c – enlarged part of spike and spikelet, d and e – spikelet showing details of glume arragments, with each bisexual flower (bearing nutlet/seed) enclosed by a glume, f – landward side of Fisherman’s site, Lake Naivasha and g – the seeded Cyerus papyrus umbel. The figure is not drawn to scale. records on the fate of ephemerals, floating papyrus islands inside the lake after reflooding and area colonized by papyrus were made at the Kamere site during a drawdown cycle (February 2008–August 2009) and after reflooding (August 2011 and January 2012). These observations have been used to understand C. papyrus drawdown-flooding dynamics and interpret genetic data. In addition, observations on the flowering rates of papyrus at Fisherman’s Camp were noted between 2007 and 2010 during a 6-monthly study and during 3–6 monthly monitoring of progression of drawdown and re-flooding (2008–2012) in order to monitor the effects of harvesting on papyrus (Terer et al., 2012b). To estimate the density of young papyrus on drawdown zones, we randomly counted a total of 9 plots each 1 m2 based on three distinct plant zonations (3 plots of 1 m2 in each zone, i.e. Asteraceae, Sphaeranthus and Sedge zones), running from parent C. papyrus in the drying shoreline toward the receding lake edge. Regeneration was presented as mean seedling counts per 1 m2 in each of the three zones. The dominant hygrophilous ephemerals and aquatic plants were either identified in the field or otherwise collected for identification at the East African Herbarium, Nairobi. A detailed list of drawdown plants in Lake Naivasha is given by Gaudet (1977), Harper (1992) and Morrison (2013). 2.4. Sampling and method for genetic analysis 2.4.1. Sampling, DNA extraction and PCR conditions At the Fisherman’s Camp site, 181 Cyperus papyrus samples for DNA analysis were collected from both the established parent stand (66 mature and 58 juvenile culms/stems) and the seedlings (57 culms) in drawdown mudflat sites situated 50–100 m distances opposite each other. The parent stand was sampled in December 2009 and seedlings in January 2010. The samples from the parent stand were collected along a 1500 m transect for both the mature and clonal juvenile stems at a 10 m interval. The seedlings were randomly collected in the drawdown zones opposite the established stand. DNA Genomic DNA extraction was performed on dry material stored on dry silica gel (15–20 mg) using the E.Z.N.A SP plant DNA Mini Kit (Omega bio-tek). Polymorphism was assayed on each DNA sample at microsatellite loci isolated from C. papyrus and selected from a microsatellite-enriched genomic library developed in the APNA-VUB laboratory following an enrichment procedure with Dynabeads (Glenn and Schable, 2005). Eight polymorphic microsatellite loci (Cypap1, Cypap3, Cypap4, Cypap5, Cypap10, Cypap13S, Cypap13F and Cypap14) showed no linkage disequilibrium and gave interpretable results. GenBank accession numbers and description of allele sizes provided online by Triest, Sierens & Terer “Multiplexing new microsatellites for papyrus (C. papyrus)” were published as Permanent Genetic Resources added to Molecular Ecology Resources Database 1 December 2012–31 January 2013 (MERPDC, 2013). Polymerase chain reaction (PCR) amplification conditions were: 1.7 ␮l 10× PCR buffer, 2.72 ␮l of each 0.2 mM dNTP, 1.36 ␮l of 2 mM MgCl2 , 0.68 ␮l 10× of the forward and reverse primer each (Hex-labeled), 0.68 ␮l of bovine serum albumin (BSA), 0.17 ␮l of 5 U Taq DNA polymerase and 2 ␮l of genomic DNA, in a total volume of 17 ␮l. The PCR amplification were performed in a thermal cycler (MJ research PTC-200 and BioRad MyCycler) and started with 3 min at 95 ◦ C, followed by 30 cycles of 45 s at 94 ◦ C, 1 min at 54 ◦ C, 2 min at 72 ◦ C and a final extension at 72 ◦ C for 5 min. The amplification products were denatured for 5 min at 95 ◦ C and separated on 8% denaturing polyacrylamide gels (acryl-bisacrylamide 19:1, 7 mM urea) with fluorescence detection (Gel Scan 2000, Corbett research) and visualized with One-Dscan software (Scanalytics). Amplicon length was estimated using the Genescan 350 Tamra Size Standard (Applied Biosystems). 2.4.2. Genetic diversity analysis After excluding four repeated multilocus genotypes (MLGs) from the data set, we calculated genetic diversity measures on 177 individuals using both fstat 2.9.3 (Goudet, 1995) and GenAlex 6.41 (Peakall and Smouse, 2006). A global test for genotypic disequilibrium based on 2800 permutations (fstat) and tests for linkage within each life stage (GenAlex) were performed. Deviations from Hardy-Weinberg equilibrium (HWE) per locus and per life stage were tested using a Chi-Square (GenAlex). 96 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 Table 1 Estimated seed production of five umbels of Cyperus papyrus taken from Fisherman’s Camp in Lake Naivasha, Kenya. Umbel 1 2 3 4 5 No. Rays (RT ) Subsample (n = 10, 50) of rays (N) and spikelet respectively N No. Spikes Average no. spikelet/ray (SPM ) Average no. seeds/spikelet (SM ) 186 210 410 415 576 49 46 42 56 48 105.4 132.6 142.6 193.3 117.2 5 5 5 4 5 At locus level, we estimated the number of alleles (A), effective number of alleles (Ae), observed heterozygosity (Ho), expected heterozygosity (He) and temporal gene flow (Nm) using GenAlex (Peakall and Smouse, 2006). Weir and Cockerham (1984) estimation of FIT (Cap F), FIS (small f) and FST (Theta) were calculated with fstat(Goudet, 1995). At life stage level (mature, juvenile and seedlings), the number of alleles (A), effective number of alleles (Ae), observed heterozygosity (Ho), expected heterozygosity (He) were calculated in GenAlex (Peakall and Smouse, 2006), while the within site inbreeding coefficient (FIS ) for all 8 loci and for 6 loci (when omitting two unlinked loci that show strong heterozygote deficiency in each site) were calculated in fstat(Goudet, 1995). Inbreeding values for the total Fisherman’s Camp transect are given by small f and jackknifed over loci using 18,000 randomisations in fstat(Goudet, 1995). 2.4.3. Genetic structure analysis Genetic structure was assessed with a two-level analysis of molecular variance (AMOVA) using allelic distance for calculating FST in GenAlex (Michalakis and Excoffier, 1996). Pairwise FST values and inferred gene flow (Nm) values were calculated between all pairs of life stages and tested for differentiation using 1000 permutations. A principal coordinate analysis (PCoA) of three life stages and a population assignment test (leave-one-out option) was based on co-dominant genotypic distances in GenAlex (Goudet, 1995). Further, a spatial autocorrelation based on allelic distances (1000 permutations) was conducted for the mature and juvenile stems, considering five distance classes at their end point with an equal number of individuals for each class, using GenAlex (Goudet, 1995). 3. Results 3.1. Seed (Nutlet) production of papyrus At the Fisherman’s Camp site C. papyrus exhibited high seed production capacity. The five umbels that were investigated produced Mean no. seeds/ray 527 662 713 773 586 Total no. seeds/umbel (ST ) 98,022 139,020 292,330 320,878 337,536 between 98,000 and 337,000 seeds (Table 1). At this site, we estimated one-third of the stems at flowering stage at any given time during our 3–6 monthly visits between year 2007 and 2012, an indication of enormous and constant seed rain onto the lakeshore and swamp. The massive seed production was a result of the many rays (186–576, Table 1) that make up the papyrus umbel. A ray consisted of 4–8 (mode = 5) spikes (Fig. 2a) each with 7–42 spikelets (mode = 27, Fig. 2b). The seeds were borne on distichously arranged flowers in the spikelet, with average counts between four and five (mode = 5 seeds, Fig. 2b–d). 3.2. Drawdown succession, papyrus seedling recruitment and density Regeneration by seedling recruitment occurred following a drought episode in 2008/2009, leading to a recession of lake water by ca. 150 m at Fisherman’s Camp. This created drawdown mudflat zones where seedling recruitment of C. papyrus and hygrophilous ephemeral established (Fig. 3a–g). Fig. 3a, b, d and g shows the main phases of drawdown-flooding period between August 2009 and September 2010. Note the position of fragmentation gap (FG) of 30 m length with a solid arrow in Fig 3a, b and d separating two continous parent papyrus stands. The first phase of the drawdown process involved exposure of receded drawdown mudflats, and drying up accompanied by cracking and oxidation of the soil (Fig. 3a). In Fig. 3a, mostly stranded littoral aquatic plants consisting of Ludwigia spp., Eichhornia crassipes (Mart.) Solms, Hydrocotyle spp., Cyperus spp. were growing at edge of parent papyrus stand and were sparsely present on dry mud. Further drying reduced the Eichhornia crassipes cover drastically. The lake rose because of river discharge after the first seasonal rains (Phase 2) and flooding the almost bare drawdown area (Fig. 3a), which evolved by December 2009 into 2 major zones; the lake-edge zone (zone 1, Fig. 3b) and parent papyrus-drawdown edge (zone 2: Sphaeranthus + other Asteraceae zone, Fig. 3b and c). Fig. 2. Parts of Cyperus papyrus images showing a – a spike with several spikelet (3–5 mm in length), b – open–up spikelet to show the position of seed (nutlet) which were alternately arranged, c – nutlet/seed ca. 50×, scale 0.2 mm and d – two distinct sides of the nutlets (grooved/flattened and ovate sides). T. Terer et al. / Aquatic Botany 116 (2014) 93–102 97 Fig. 3. (a–g) Photos showing drawdown – flooding progression, zonation, succession and regeneration of Cyperus papyrus and hygrophilous ephemerals by seedling recruitment in drawdown area created by receded Lake Naivasha, Fisherman’s Camp site (Photos: Terer T.). A fragmentation gap (FG) separating two continuous parent papyrus stand is shown by a solid arrow in Fig. 3a, b and d. (a) Drawdown desicated mud – flat zone with cracked soils in late stage of drawdown [August. 2009], (b) Initial reflooding phase showing 2 main zones lake – edge zone (zone 1/seedling zone) and drawdown – parent papyrus (zone 2/hygrophilous zone) [December 2009]. (c) A zoom – in of hygrophilous species succession. Note Nicotiana glauca Graham shown with a white arrow and sheep grazing in outer part of zone 2 [December 2009]. (d) Progression of reflooding showing distinct plant zonation from lake – edge toward land [January 2010]. (e) A zoom – in of papyrus young plants (seedlings) growing among the hygrophilous ephemerals [January 2010] pointed with black arrow. (f) Three – times enlarged photo showing uprooted papyrus young plant already with several clonal shoots [January 2010]. (g) Mature grown papyrus from seedlings forming a fringing swamp parallel to the parent stand. Note that all the hygrophilous emphemerals died off, leaving only papyrus [September 2010]. 98 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 Table 2 Genetic diversity measures and characteristics of each microsatellite locus of the Fisherman’s Camp Cyperus papyrus population. Microsatellite locus Mean sample size A Ae Ho He F Capf Smallf Theta Nm Cypap1 Cypap3 Cypap4 Cypap5 Cypap10 Cypap13fa Cypap13sa Cypap14 58 58 58 58 58 58 58 58 4.0 6.3 6.7 5.0 4.3 4.0 3.3 2.0 2.8 3.5 4.1 3.2 2.6 2.7 2.1 1.1 0.699 0.729 0.717 0.634 0.602 0.086 0.282 0.059 0.647 0.712 0.758 0.684 0.613 0.627 0.510 0.056 −0.080 −0.022 0.055 0.072 0.016 0.862 0.451 −0.031 −0.077 −0.020 0.067 0.072 0.030 0.865 0.449 −0.019 0.072 0.020 0.062 0.076 0.031 0.866 0.451 −0.042 −0.005 0.000 0.005 −0.005 −0.001 −0.009 −0.002 0.022 118 40 24 91 70 37 28 12 Overall SE 58 0.7 4.4 0.3 2.8 0.19 0.476 0.057 0.576 0.044 0.165 0.066 0.179 – 0.180 – −0.002 – 53 13 Mean number of alleles (A), effective number of alleles (Ae), observed heterozygosity (Ho), expected heterozygosity (He), at the level of three life stages (temporal effects), Weir and Cockerham (1984) estimation of FIT (CapF), FIS (smallf) and FST (Theta), gene flow (Nm). a Two unlinked loci amplified from similar primer regions but showing high fixation levels. As reflooding progressed in January 2010 (Phase 3), plant zonation became distinct from the lake-landwards, notably: Sedge zone (with papyrus seedlings), Sphaeranthus zone, Asteraceae zone, parent C. papyrus and Acacia zones (Fig. 3d). The Sedge and Sphaeranthus zones were dominated by C. papyrus and Sphaeranthus species respectively, while the Asteraceae zone, which stretched to the parent papyrus zone consisted of many plants of the Asteraceae family (e.g. Bidens pilosa L., Tagetes minuta L., Conzya spp., Sphaeranthus spp.), mixed with C. papyrus, Cynodon spp., Eichhornia crassipes (Mart.) Solms (water hyacinth) and Ludwigia stolonifera (Guill. & Perr.) P.H. Raven (Fig. 3e). The zones with grasses formed good grazing areas for herbivores such as sheep, cattle and hippos (Fig. 3c). A few individuals of Hibiscus diversifolius Jacq., Solanum nigrum L. and Vigna luteola (Jacq.) Benth. were observed at the edges of the parent papyrus zone. Densities of C. papyrus seedlings in January 2010 (Phase 3) varied between zero in the parent papyrus and adjacent dry zones; 186 ± 51 m−2 in the Sedge zone (consisted of all Cyperus sp. because young plants without inflorescence could not be identified to species levels); 56 ± 12 m−2 in the Sphaeranthus zones; and 5 ± 3 m−2 in the Asteraceae zone. In the Final Phases eight months later, the papyrus seedlings (Fig. 3e and f) had grown, clonally spread and reached maturity forming a fringe papyrus swamp parallel to the parent stand (Fig. 3g). Thus the drawdown succession described above culminated in a newly formed papyrus stand from seedling and local expansion by clonal propagation. A follow-up of these newly formed papyrus was done in 2011 and 2012. In August 2011, we observed that some of the mature papyrus stands from seedlings were drowned; some were stranded inshore as floating mats; and others had been moved by currents and were incorporated into the parent papyrus stand; while still others had occupied new sites such as at Kamere which was devoid of papyrus during drawdown period but was overgrown when we visited it in January, 2012. We also observed that the hygrophilous ephemerals (e.g. Conyza spp., Nicotiana glauca Graham, Fig. 3c–e) could not withstand high water levels inshore and had died and drowned (Fig. 3g). frequency (0.008–0.026), namely observed 1–3 times under heterozygote conditions for 5 loci (Cypap1, Cypap3, Cypap4, Cypap10 and Cypap13S) in 2 mature, 1 juvenile and 6 seedling individuals. HWE conditions were met globally for 6 loci but two loci (Cypap13S and Cypap13F) consistently showed a strong deficiency in heterozygotes within each site, resulting in very high inbreeding values (small f = 0.4 and 0.866 respectively, Table 2). The 8 loci consistently showed no genetic differentiation with an overall Theta close to zero (Table 2). 3.4. Genetic diversity in three groups of life stages in the population Each life stage harbored 35–38 alleles out of the 40 observed for the whole population. Global values for allele diversity (Ae = 2.8) and heterozygosity (Ho = 0.476, He = 0.576) showed little or no variation among life stages, but the seedlings in the drawdown zone contained the highest number of alleles (Table 3). FIS values showed that significant higher levels of inbreeding (p < 0.05) occurred in the seedling zone (Table 3), indicating a local recruitment through seeds from related individuals or seeds from the same umbel. A two-level AMOVA (allelic distance method) revealed that all variance was within (81%) individuals and among individuals (19%) but not among the three life stages, resulting in a full temporal connectivity effect among stages and consequently very high levels (infinite) of gene flow between parent (established) papyrus and the seedlings in the drawdown zone. A PCoA (Fig. 4) revealed a single group of all life stages with seedlings superposing the parent papyrus stand of mature and juvenile individuals (first axis, 27%; second axis, 22% of the variation explained), an indication of seed rain coming from proximate C. papyrus stands. A population assignment test resulted in 36% outcomes to the same life stage and 64% to the other life stage, thereby showing no different outcome for the seedlings. 3.3. Microsatellite loci features for the population The eight microsatellite loci showed allelic diversity among seedlings and individuals of a 1500 m transect (adult and juveniles) within the Fisherman’s Camp population and allowed us to identify nearly as many genets (177) as ramets (181). Within-population tests however showed no linkage for any of the locus pairs and therefore further analyses were done for 8 loci. A total of 40 alleles was observed with 2–8 alleles per locus and Ae ranging from 1.1 to 4.1 (Table 2). Six private alleles were at low Fig. 4. PCoA of multilocus genotypes showing seedlings (sl) that fully overlap with mature (mat) and clonal juvenile (juv) of a nearby established papyrus population. 99 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 Table 3 Cyperus papyrus genetic diversity measures for each sampling site and overall for Fisherman’s Camp population. Population and sites Mature culms Juvenile culms Seedlings *Total (Fisherman’s Camp) SE A PA Na Ae Ho He 66 58 57 64 56 57 36 35 38 2 (2) 1 (1) 3 (6) 4.4 4.4 4.6 2.8 2.8 2.8 0.485 0.502 0.441 0.571 0.590 0.566 0.157 0.157 0.230 181 177 40 6 (9) 4.5 0.3 2.8 0.2 0.476 0.057 0.576 0.044 0.178a −0.123 N Genets FIS 8 loci FIS § 6 loci −0.021 −0.015 0.089* 0.016a 0.027a Number of sampled culms (N), Number of alleles (A), Number of private alleles (PA) and number of individuals with a private allele (between brackets), Mean number of alleles (Na), Effective number of alleles (Ae), observed heterozygosity (Ho), expected heterozygosity (He), within site inbreeding coefficient (FIS ) for all 8 loci and FIS § when omitting two unlinked loci that show strong heterozygote deficiency in each site. * Significant at p < 0.05 level. a Inbreeding values for the total Lake Naivasha population are given by small f and jackknifed over loci. Most of the mature and juvenile individuals at very short distance (10 m) had a different multilocus genotype and in only 4 cases (Table 3), at one outer edge of the transect belt, we found evidence of clonal propagation over 10 m distance, which indicates occasional edge effects through lakeward expansion by perennation. The transect with C. papyrus individuals of either mature clumps or their putative juveniles along the lake margin, both revealed a similar fine-scaled spatial genetic structure at very short distances at 100 m intervals (p = 0.026 and p = 0.036 respectively for mature and juvenile individuals, Fig. 5). 4. Discussion 4.1. Seed production, establishment of papyrus seedlings and role of ephemerals Seed production is one of the life history traits of a species that can be used to predict the composition of wetland vegetation (van der Valk, 1981). Abundant seed production is important in maintaining seed bank and intergenerational genetic diversity which is known to cushion plants against the consequences of habitat fragmentation, genetic drift and differentiation (Honnay et al., 2008). In the present study, we found high densities of seeds in flowering papyrus umbels which provide a constant seed supply that could germinate when drawdown conditions occur. The in situ germination of seedlings per square meter and the observed zone variation are congruent to findings by Gaudet (1977) and Boar (2006). The seedling density/m2 can explain to some extent the fate of the huge seed production found in the seeded umbel. Gaudet (1977) found high densities of viable papyrus seeds (up to 1742) in the drawdown zone soil at horizon depth of 0–10 cm in the Eastern site of the Lake Naivasha, while in a sowing experiment, Boar (2006) found a maximum germination rate of 45% (5846seeds per m2 ) in a varied water level treatment. In the field, once germinated, seedlings are subject to the added hazard of being grazed or browsed by domestic and wild animals. Likewise, a study by Welling et al. (1988) in a prairie wetland showed variation in maximum mean density of seedlings after drawdown compared to maximum mean seed-bank density for Typha spp. (e.g. 250 seedlings m−2 vs. 610 seeds m−2 ), Scirpus spp., Phragmites spp., Carex spp. and Scolochloa spp. This was generally attributed to inconsistencies in distributions and densities between seeds and seedlings for a given species along height gradient in addition to variation in some environmental factors. From both laboratory experiments and in situ seedlings germination (Gaudet, 1977; Boar, 2006, present study), it can be assumed that in the case of papyrus, most seeds are buried in the lake soil mud where germination conditions are unsuitable. In the lake, widespread dispersal by water currents and wind may be responsible for the large variation in distribution and density. Both availability of seeds and microsite are vital in seedling recruitment which is further affected by factors such as dispersal, disturbance frequency or seed predation (Eriksson and Ehrlén, 1992). On the basis of available evidence (Gaudet, 1977; Boar, 2006) and Fig. 5. Spatial structure of mature and clonal juvenile culms in 1500 m long transect considering equal distance classes (r = relatedness; U and L are upper and lower confidence values). 100 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 considering the huge seed production observed in the current study, microsite limitation would seem to be a predominant factor in seedling recruitment of C. papyrus. Such information is crucial in management plans in order to ensure survival of papyrus during drawdown-flooding cycle period. The present study and earlier results by Gaudet (1977) and Boar (2006), show that C. papyrus in Lake Naivasha has a persistent seed bank, although seed longevity is still unknown. The drawdown-flooding cycle is an important factor that shapes population dynamics of C. papyrus. This is because it provides a window of opportunity for seedling recruitment as well as the formation of new populations (during drawdown-flooding interphase) and dispersal to new sites or recruitment into established populations (such as floating mats) during flooding. Terer et al. (2012b) found no seedling recruitment in established papyrus communities in experimental plots monitored for biomass investigations in the same site, thus confirming that seedling recruitment happens under drawdown conditions (van der Valk, 1981; Gaudet, 1977; Harper, 1992). Earlier drawdown episodes in Lake Naivasha corresponded to El Niño-La Niña Southern Oscillation cycles, e.g. in the late 1920s, mid-1930s, early 1940s, early 1970s and 1980s, and these resulted in seedling recruitment and areal expansion of the papyrus swamp (Gaudet, 1977; Harper, 1992; Viles and Goudie, 2003; Boar, 2006). By combining two strategies, seedling recruitment during drawdown and vegetative propagation year-round, C. papyrus appears adapted to the drawdown-flooding conditions in the tropics. It should be noted that the requirement of drawdown-like conditions for seedling recruitment may not apply in C. papyrus populations in a broader sense, since this study was confined to only one lake. Some of the mature papyrus stands originating from recent seedlings were drowned; some were stranded inshore as floating mats; and others had been moved by wind and water currents and were incorporated into the parent papyrus stand. Thus, even after recruitment and early growth, a papyrus swamp may or may not result from this process. This is in line with Morrison (2013) who found that on Lake Naivasha despite its emergence as the dominant species after re-flooding, papyrus seedlings may fail to persist as a fringing swamp. He ascribed this to increased anthropogenic impacts on the riparian zone, and that whilst papyrus seeds are capable of germinating rapidly when sediments are rewetted, new swamps will only persist when lake level recovers slowly enough for seedling growth to keep pace with water depth. Our findings of primary succession and zonation in drawdown zones agree with a detailed documentation including a full list of 108 species for Lake Naivasha provided by Gaudet (1977), a Gleasonian model of succession proposed by van der Valk (1981), and a recent study by Morrison (2013) of plant succession in wetlands and the drawdown region of Lake Naivasha. The drawdown drying phase and early stage of succession captured in Fig. 3a and b was similar in nearly all description aspects provided by Gaudet (1977) in the Eastern site of the lake except the noticeable absence of Nymphaea spp. in the lake-edge mud. In the current study but not mentioned by Gaudet (1977), in both intermediate (Phase 2) and late stages (Phase 3) of succession, the Asteraceae Zone was infested with a new hygrophilous invasive species Nicotiana glauca. Eichhornia crassipes is also absent in Gaudet’s (1977) plant record, whose proliferation seemed to be controlled by drawdown events. Nicotiana glauca is considered native to South America (Brandes and Fritzsch, 2000), but also now common in roadside, e.g. in NairobiAthi River area and Maai Mahiu-Narok road (T. Terer: Personal observation). Regarding the study of drawdown flora on Lake Naivasha carried out by Morrison, he also found three zones during the 2008–2010 drawdown, but there was a relative paucity of species today compared to 35 years ago: total species richness in Morrison’s study was lower by about a half (51%), while over two-thirds (69%) of the species identified by Gaudet (1977) were missing altogether from Morrison’s 2008–2010 study. The presence of hygrophilous ephemerals in the drawdown growing zone together with C. papyrus seedlings and later succumbing to high water levels rise after flooding leading to mass deaths, is a phenomenon that could be important for the survival of C. papyrus and the evolution of the lake shore ecosystems. These hygrophilous ephemerals could be playing a crucial role in the survival of newly recruited seedlings in drawdown zones in that: (1) they provide abundant food for herbivores such as ungulates, thus diverting attention away from feeding on papyrus seedlings. Gaudet (1977) pointed out that C. papyrus was then subjected to grazing pressure from domestic animals, hippo, and buffalo, but mostly when the plants were young and tender. In modern times, Morrison (2013) found a great change in that disturbance of the flora was particularly high during the drought of 2009 when pastoralists and their herds, numbering in their tens of thousands converged on the shoreline from surrounding regions; (2) they provide a mesh-like structure that holds papyrus stems (be it from seedlings or clonal sprouting) firmly anchored, minimizing the possibility of damage due to wind and water current, thus facilitating subsequent dispersal of newly formed papyrus island mats. Ephemeral species such as Conyza and Gnaphalium are known to leave behind a thick root mass when they die which does not decompose easily due to anaerobic conditions after inundation (Gaudet, 1977). In our study we observed standing but dry remnants of ephemerals; and (3) they provide an initial detritus that accumulates, then rots and release a nutrient flush that mimicks the detritus-based nutrient cycling process found in established papyrus stands (Gaudet, 1976) where ramet lifespan is estimated to be 9–12 months (Van Dam et al., 2007). 4.2. Comparison of genetic diversity and gene flow of seedlings, and parent papyrus populations The seedlings in the drawdown zone exhibited comparable high genetic diversity as measured by heterozygosity similar to established C. papyrus populations which included both the mature and their perennate juvenile life stages. As expected for seedlings, no similar multilocus genotypes were encountered (Number of genets equals the number of ramets). The established mature population had fewer genets than ramets, thus explaining the local fine-scaled and year-round clonal growth that continues to occur once a population is established from seed. The overall genetic diversity of the Fisherman’s Camp papyrus population (He = 0.48, Ho = 0.58) is close to the overall Lake Naivasha papyrus population (He = 0.47, Ho = 0.55) (Triest et al., 2013). These gene diversities are comparable to those of other emergents such as Spartina alterniflora L. (He = 0.49–0.77, Ho = 0.61–0.73) or the more related Carex kobomugi Ohwi (He = 0.58–0.60, Ho = 0.60–0.66) but much higher than those reported for Typha angustifolia L. (He = 0.15–0.65, Ho = 0.19–0.67), all using SSR markers (Tsyusko et al., 2005; Novy et al., 2010; Ohsako, 2010). Allelic variation between seedlings and the neighboring established population life stages was noticeable, with more allelic diversity (A, Ae, Na) and private alleles (PA) recorded in seedlings than established stands, an indication of how seedlings maintain or even may increase the gene pool diversity of populations. The private alleles (PA) suggested an external gene flow (outside the site studied at Fisherman’s Camp), though their frequencies were too low to significantly influence the overwhelming local gene flow levels between the seedlings and established stands. T. Terer et al. / Aquatic Botany 116 (2014) 93–102 Consequently, both the seedlings and established stand can be considered here as a single gene pool. As noted above, the formation of new fringing papyrus through seedling recruitment and dispersal of floating mats with shoots by mainly hydrochory and anemochory (and probably to less extent by zoochory, e.g. hippos feeding on umbels with seeds and later defecating in drawdown zones) during drawdown-flooding phenomena shapes the C. papyrus population structure within Lake Naivasha in terms of areal extent and recuperation of genotypes. This dispersal mechanism even could be valid at the lake level because high levels of gene flow were also found between populations from various parts of the lake in which more remote papyrus individuals also appeared to belong to a single genetic cluster after Bayesian analysis (Triest et al., 2013). 4.3. Parent population genetic structure at fine-scale The detection of private alleles in the established stand also indicates a carried forward signature of previous (or past) episodic seedling recruitment, which allows new allele migrants to be incorporated in the recipient gene pool every time it occurs. The 1500 m transect revealed spatial genetic structure at a very short distance of ca. 100 m which indicates the remaining influence caused by an initial seed rain event, since it is clear from the infinite gene flow and PCoA overlap of multilocus genotypes that established papyrus stands release a seed rain nearby the culms (stems) that disperse them. This structuring may reflect the extent of influence of a seed rain, thus the distance may indicate the horizontal transition point among cohorts when established as seedlings in temporary drawdown sites. Additionally, the colonization of papyrus in the Kamere site which was unoccupied during drawdown shows the importance of shoot dispersal and floating mats after reflooding. This demonstrated the action of wind and water currents in the dispersal of floating papyrus clumps originating from cohorts of seedlings. More importantly this also shows how the consequences of either human or natural fragmentation can be mitigated. Indeed, hydrochory and anemochory are important channels of gene flow in riparian and wetland plant communities (Chen et al., 2009; Pollux et al., 2009; Nilsson et al., 2010). Overall, mixed reproduction strategies (sexual recruitment and clonal persistence), combined with ample gene flow across relatively small spatial scales, appear to be responsible for maintaining high levels of genetic diversity and a spatial structure within Fisherman’s Camp C. papyrus population. Remnant C. papyrus at lake margins thus can be hotspots of genetic diversity and consequently important for the recovery from fragmentation in Lake Naivasha caused by human activities, which is known to affect gene exchange among subpopulations (Chen et al., 2012), even within a closed lake such as Lake Naivasha (Triest et al., 2013). Acknowledgments This study was conducted as part of the first author’s PhD program for the Vrije Universiteit Brussel (VUB) funded by Flemish Inter-University Council (VLIR)-University Development Cooperation (UOS). This manuscript was completed under VLIR-UOS postdoctoral scholarship program during winter period (2012) at VUB given to T.T. We acknowledge logistical support from National Museums of Kenya, Nairobi. DNA analysis was financed by VUB (OZR-BOF) and performed with assistance from T. Sierens. We thank Dr. G. Mugambi and N. Muema of the National Museums of Kenya, Nairobi for stereoscope imaging and the papyrus drawings (Fig. 1). Jan Vermaat and two anonymous reviewers provided helpful reviews of an earlier version of the manuscript. 101 References Boar, R.R., 2006. Responses of a fringing Cyperus papyrus L. swamp to changes in water level. Aquat. Bot. 84, 85–92. Boar, R.R., Harper, D.M., Adams, C.S., 1999. Biomass allocation in Cyperus papyrus in a tropical wetland, Lake Naivasha, Kenya. Biotropica 31, 411–421. Bond, N.R., Lake, P.S., Arthington, A.H., 2008. The impacts of drought on freshwater ecosystems: an Australian perspective. Hydrobiologia 600, 3–16. Brandes, D., Fritzsch, K., 2000. Alien plants of Fuerteventura, Canary Islands. Plantas extranjeras de Fuerteventura, Islas Canarias, http://opus.tu-bs.de/ opus/volltexte/2000/79 (accessed 14.10.12). Brock, M.A., Rogers, K.H., 1998. The regeneration potential of seed bank of an ephemeral floodplain in South Africa. Aquat. Bot. 61, 123–135. Chen, Y., Li, X., Yin, L., Cheng, Y., Li, W., 2009. Genetic diversity and migration patterns of the aquatic macrophyte Potamogeton malaianus in a potamo-lacustrine system. Freshw. Biol. 54, 1178–1188. Chen, Y-y., Kong, D-r., Huang, C-h., Xu, Y-x., Li, Z-z., 2012. Microsatellite analysis reveals the genetic structure and gene flow of the aquatic quillwort Isoetes sinensis, a critically endangered species in China. Aquat. Bot. 96, 52–57. de Winton, M.D., Clayton, J.S., Champion, P.D., 2000. Seedling emergence from seed banks of 15 New Zealand lakes with contrasting vegetation histories. Aquat. Bot. 66, 181–194. Eriksson, O., Ehrlén, J., 1992. Seed and microsite limitation of recruitment in plant populations. Oecol. 91, 360–364. Gaudet, J.J., 1976. Nutrient relationships in the detritus of a tropical swamp. Arch Hydrobiol. 78, 213–239. Gaudet, J.J., 1977. Natural drawdown on Lake Naivasha. Kenya and the formation of papyrus swamps. Aquat. Bot. 3, 1–47. Gaudet, J.J., Melack, J.M., 2006. Major ion chemistry in a tropical lake basin. Freshw. Biol. 11, 309–333. Gichuki, J., Dahdouh-Guebas, F., Mugo, J., Rabuor, C.O., Triest, L., Dehairs, F., 2001. Species inventory and the local uses of plants and fishes of the Lower Sondu Miriu wetland of Lake Victoria, Kenya. Hydrobiologia 458, 99–106. Gichuki, N.N., Oyieke, H., Terer, T., 2006. Status and root causes of biodiversity loss in the Eastern Rift Valley Lakes, Kenya. In: Odada, E.O., Olago, D.O., Ochola, W., Ntiba, M., Wandiga, S., Gichuki, N.N., Oyieke, H. (Eds.), Proceedings of 11th World Lakes conference, vol. 2. Nairobi, pp. 511–517. Glenn, T.C., Schable, N.A., 2005. Isolating microsatellite DNA loci. Meth. Enzymol. 395, 202–222. Goudet, J., 1995. FSTAT (Version 1.2). A computer program to calculate F–Statistics. J. Hered. 86, 485–486. Gunn, D.L., 1973. Consequences of cycles in East African climate. Nature 242, 457. Harper, D.M., 1992. The ecological relationships of aquatic plants at Lake Naivasha, Kenya. Hydrobiologia 232, 65–71. Harper, D.M., Morrison, E., Macharia, M.M., Upton, C., 2011. Lake Naivasha, Kenya: ecology, society and future. Freshw. Rev. 4, 89–114. Hickley, P., Muchiri, M., Boar, R.R., Britton, R., Adams, C.S., Gichuru, N., Harper, D.M., 2004. Habitat degradation and subsequent fishery collapse in Lakes Naivasha and Baringo, Kenya. Ecohydrol. Hydrobiol. 4, 503–517. Honnay, O., Bossuyt, B., 2005. Prolonged clonal growth: escape route or route to extinction? Oikos 108, 427–432. Honnay, O., Bossuyt, B., Jacquemyn, H., Shimono, A., Uchiyama, K., 2008. Can a seed bank maintain the genetic variation in the above ground plant population? Oikos 117, 1–5. Howard-Williams, C., 1975. Vegetation changes in a shallow African lake: response of the vegetation to a recent dry period. Hydrobiologia 47, 381–398. IUCN, 2010. IUCN Red List of Threatened Species. Version 2010.1, https://www. iucnredlist.org (accessed 5.06.10). Jones, M.B., Humphries, S., 2002. Impacts of the C4 sedge Cyperus papyrus L. on carbon and water fluxes in an African wetland. Hydrobiologia 488, 107–113. Jones, M.B., 1983. Papyrus a new fuel for the third world. New Sci. 99, 418–421. Mnaya, B., Wolanski, E., 2002. Water circulation and fish larvae recruitment in papyrus wetlands, Rubondo Island, Lake Victoria. Wetl. Ecol. Manage. 10, 133–143. MERPDC (Molecular Ecology Resources Primer Development Consortium), 2013. Permanent Genetic Resources added to Molecular Ecology Resources Database 1 December 2012–31 January 2013. Mol. Ecol. Resour. 13, 546–549. Michalakis, Y., Excoffier, L., 1996. A generic estimation of population subdivision using distances between alleles with special reference for microsatellite loci. Genetics 142, 1061–1064. Morrison, J.H.E., Harper, D.M., 2009. Ecohydrological principles to underpin the restoration of Cyperus papyrus at Lake Naivasha, Kenya. Ecohydrol. Hydrobiol. 9, 83–97. Morrison, J.H.E., (PhD. thesis) 2013. Wetland Restoration at Lake Naivasha, Kenya: Social and Ecological Considerations. University of Leicester, UK. Nilsson, C., Brown, R.L., Jansson, R., Merritt, D.M., 2010. The role of hydrochory in structuring riparian and wetland vegetation. Biol. Rev. 85, 837–858. Novy, A., Smouse, E.P., Hartman, M.J., Struwe, L., Honig, J., Miller, C., Alvarez, M., Bonos, S., 2010. Genetic variation of Spartina alterniflora in New York metropolitan area and its relevance for marsh restoration. Wetlands 30, 603–608. Ohsako, T., 2010. Clonal and spatial genetic structure within populations of a coastal plant. Carex kobomugi (Cyperaceae). Am. J. Bot. 97, 458–470. Owino, A.O., Ryan, P.G., 2006. Habitat associations of papyrus specialist birds at three papyrus swamps in western Kenya. Afr. J. Ecol. 44, 438–443. 102 T. Terer et al. / Aquatic Botany 116 (2014) 93–102 Peakall, R., Smouse, P.E., 2006. GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Mol. Ecol., 288–295, Notes 6. Pollux, B.J.A., Luijten, A., van Groenendael, J.M., Ouborg, N.J., 2009. Gene flow and genetic structure of the aquatic macrophyte Sparganium emersum in a linear unidirectional river. Freshw. Biol. 54, 64–76. Santamaria, L., 2002. Why are most aquatic plants widely distributed? Dispersal, clonal growth and small-scale heterogeneity in a stressful environment. Acta Oecologica 23, 137–154. Saunders, M.J., Jones, M.B., Kansiime, F., 2007. Carbon and water cycles in tropical papyrus wetlands. Wetl. Ecol. Manage. 15, 489–498. Shibayama, Y., Kadono, Y., 2007. The effect of water- level fluctuations on seedling recruitment in an aquatic macrophyte Nymphoides indica (L) Kuntze (Menyantheceae). Aquat. Bot. 87, 320–324. Silvertown, J., 2008. The evolutionary maintenance of sexual reproduction: Evidence from the ecological distribution of asexual reproduction in clonal plants. Int. J. Plant Sci. 169, 157–168. Simpson, A.D., Inglis, A.C., 2001. Cyperaceae of economic, ethnobotanical and horticultural importance: A checklist. Kew Bull. 56, 257–360, http://www. jstor.org/pss/4110962 (accessed 10.03.10). Smith, L.M., Kadlec, J.A., 1983. Seed banks and their role during drawdown of a North American marsh. J. Appl. Ecol. 20, 673–684. Terer, T., (PhD. thesis) 2011. Conservation Genetics, Utilization and Effects of Cyperus Papyrus Harvesting: Making Ecosystem Management Work in Kenyan Wetlands. Vrije Universiteit Brussel. Uitgeverij VUBPRESS Brussels University Press, Brussels, ISBN 978 90 5487 885 8. Terer, T., Gichuki, N.N., Ndiritu, G.G., 2006. Role of wetlands in maintaining stability of tropical lakes: a case study of River Nyando delta in Lake Victoria basin, Kenya. In: Odada, E.O., Olago, D.O., Ochola, W., Ntiba, M., Wandiga, S., Gichuki, N.N., Oyieke, H. (Eds.), Proceedings of 11th World Lakes Conference, vol. 2. Nairobi, pp. 560–567. Terer, T., Muasya, A.M., Dahdouh-Guebas, F., Ndiritu, G., Triest, L., 2012a. Integrating local ecological knowledge and management strategies of an isolated semi-arid papyrus swamp (Loboi, Kenya) into a wider conservation framework. J. Environ. Manage. 93, 71–84. Terer, T., Triest, L., Muasya, A.M., 2012b. Effects of harvesting Cyperus papyrus in undisturbed wetland, Lake Naivasha. Hydrobiologia 680, 135–148. Triest, L., Sierens, T., Terer, T., 2013. Diversity and fine-scale genetic structure of Cyperus papyrus populations in Lake Naivasha (Kenya) using microsatellite markers. Hydrobiologia, 10.1007/s10750-013-1584-8. Tsyusko, V.O., Smith, H.M., Sharitz, R.R., Glenn, T., 2005. Genetic and clonal diversity of two cattail species, Typha latifolia and T. angustifolia (Typhaceae), from Ukraine. Am. J. Bot. 92, 1161–1169. Uesugi, R., Tsumura, K., Washitani, I., 2007. Restoration of genetic diversity from soil seed banks in a threatened aquatic plant, Nymphoides peltata. Conserv. Genet. 8, 111–121. Van Dam, A.A., Dardona, A., Kelderman, P., Kansiime, F., 2007. A simulation model for nitrogen retention in a papyrus wetland near Lake Victoria, Uganda (East Africa). Wetl. Ecol. Manage. 15, 469–480. van der Valk, A.G., 1981. Succession in wetlands: a Gleasonian approach. Ecology 62, 688–696. van der Valk, A.G., 2005. Water-level fluctuations in Northern American Prairie wetlands. Hydrobiologia 539, 171–188. van der Valk, A.G., Davis, C.B., 1978. The role of seed banks in the vegetation dynamics of prairie glacial marshes. Ecology 59, 322–335. Van Geest, G.J., Coops, H., Roijackers, R.M.M., Buijse, A.D., Scheffer, M., 2005a. Succession of aquatic vegetation driven by reduced water-level fluctuations in floodplain lakes. J. Appl. Ecol. 42, 251–260. Van Geest, G.J., Wolters, H., Roozen, F., Coops, H., Roijackers, R.M.M., Buijse, A.D., Scheffer, M., 2005b. Water-level fluctuations affect macrophyte richness in floodplain lakes. Hydrobiologia 539, 239–248. Verschuren, D., Laird, K.R., Cumming, B.F., 2000. Rainfall and drought in equatorial east Africa during the past 1100 years. Nature 403, 410–414. Viles, H.A., Goudie, A.S., 2003. Interannual, decadal and multidecadal scale climatic variability and geomorphology. Earth-Sci. Rev. 61, 105–131. Weir, B.S., Cockerham, C.C., 1984. Estimating F–statistics for the analysis of population structure. Evolution 38, 1358–1370. Welling, C.H., Perderson, R.L., van der Valk, A.G., 1988. Recruitment from the seed bank and the development of emergent vegetation during a drawdown in a prairie wetland. J. Ecol. 76, 483–496. Wilcox, D.A., Meeker, J.E., 1991. Disturbance effects on aquatic vegetation in regulated and unregulated lakes in northern Minnesota (USA). Can. J. Bot. 69, 1542–1551.