Population genetic structure and connectivity of the seagrass Thalassia hemprichii in the Western Indian Ocean is influenced by predominant ocean currents.

Marlene Jahnke,Mathew Ogalo Silas,Said Mgeleka,Maria E Asplund,Jamal Mahafina,Arielle Hoamby,Lina Mtwana Nordlund,Josefine Larsson,Martin Gullström

doi:10.1002/ece3.5420

Abstract

This study is the first large‐scale genetic population study of a widespread climax species of seagrass, Thalassia hemprichii, in the Western Indian Ocean (WIO). The aim was to understand genetic population structure and connectivity of T. hemprichii in relation to hydrodynamic features. We genotyped 205 individual seagrass shoots from 11 sites across the WIO, spanning over a distance of ~2,700 km, with twelve microsatellite markers. Seagrass shoots were sampled in Kenya, Tanzania (mainland and Zanzibar), Mozambique, and Madagascar: 4–26°S and 33–48°E. We assessed clonality and visualized genetic diversity and genetic population differentiation. We used Bayesian clustering approaches (TESS) to trace spatial ancestry of populations and used directional migration rates (DivMigrate) to identify sources of gene flow. We identified four genetically differentiated groups: (a) samples from the Zanzibar channel; (b) Mozambique; (c) Madagascar; and (d) the east coast of Zanzibar and Kenya. Significant pairwise population genetic differentiation was found among many sites. Isolation by distance was detected for the estimated magnitude of divergence (D EST), but the three predominant ocean current systems (i.e., East African Coastal Current, North East Madagascar Current, and the South Equatorial Current) also determine genetic connectivity and genetic structure. Directional migration rates indicate that Madagascar acts as an important source population. Overall, clonality was moderate to high with large differences among sampling sites, indicating relatively low, but spatially variable sexual reproduction rates. The strongest genetic break was identified for three sites in the Zanzibar channel. Although isolation by distance is present, this study suggests that the three regionally predominant ocean current systems (i.e., East African Coastal Current, North East Madagascar Current, and the South Equatorial Current) rather than distance determine genetic connectivity and structure of T. hemprichii in the WIO. If the goal is to maintain genetic connectivity of T. hemprichii within the WIO, conservation planning and implementation of marine protection should be considered at the regional scale—across national borders.

Highlights

Understanding population genetic structure, long‐distance dispersal and connectivity patterns of organisms facilitates conservation actions in spatially widespread coastal seascapes (Jones et al, 2009)
While in the Western Indian Ocean (WIO) we are only aware of one local assessment of the seagrass Thalassodendron ciliatum from 2001 in southern Mozambique using RAPD markers (Bandeira & Nilsson, 2001) and one recent study about the seagrass Zostera capensis sampled along the South African coast, and in one bay in Mozambique and one bay in Kenya indicating the presence of two population clusters broadly corresponding to populations on the west and east coasts of Africa (Phair, Toonen, Knapp, & Heyden, 2019)
Our findings indicate that the genetic structure of T. hemprichii in the WIO is influenced by large oceanic currents (SEC, North East Madagascar Current (NEMC), and East African Coastal Current (EACC)), as well as by local hydrodynamic patterns (Zanzibar channel)

Summary

Introduction

Understanding population genetic structure, long‐distance dispersal and connectivity patterns of organisms facilitates conservation actions in spatially widespread coastal seascapes (Jones et al, 2009). Population genetics may act as a powerful tool for resource management planners to understand the genetic connectivity between populations, which in turn may have implications for decisions regarding number, sizes, and locations of protected areas (Palumbi, 2003; Waycott et al, 2009). Seagrass meadows have experienced a substantial decline (Orth et al, 2006; Waycott et al, 2009) and information regarding genetic structure and variability of meadows could provide an important link for coastal management actions. While in the Western Indian Ocean (WIO) we are only aware of one local assessment of the seagrass Thalassodendron ciliatum from 2001 in southern Mozambique using RAPD markers (Bandeira & Nilsson, 2001) and one recent study about the seagrass Zostera capensis sampled along the South African coast, and in one bay in Mozambique and one bay in Kenya indicating the presence of two population clusters broadly corresponding to populations on the west and east coasts of Africa (Phair, Toonen, Knapp, & Heyden, 2019)

Objectives

Findings

Discussion

Conclusion