5.2.1 Water levels and currents
In this section, we compared our baseline simulation with a simulation without mangrove in order to assess the effect of vegetation on tidal dissipation. The water levels comparison from these 2 configurations at the lagoon mouth (Aquadoop) and upstream inside the mangrove (Mangrove-PT2) is shown in Fig. 4. Without vegetation, the tidal amplitude is excessive in the model and this results in a 20–30% stronger RMSD. Due to the absence of attenuation by vegetation, the maximum of these RMSDs is found in lagoon areas devoid of mangroves. Previous studies also reported the dissipation of tides as it propagates through vegetation (McIvor et al. 2012, Montgomery et al. 2018, Moki et al. 2020; Zhang et al. 2020).
Tidal dissipation by mangrove forest in the lagoon is also well highlighted by the magnitude of the currents, which are substantially overestimated without vegetation (Fig. 5), with a RMSD increased by 60%. In more details, accounting for vegetation has a much larger impact on flood currents than ebb currents, so that the asymmetry in current velocity is much reduced when neglecting vegetation. The spatial extension of this behavior will have to be investigated in future studies, as well as its impact on sediment transport.
Montgomery et al (2018) and Horstman et al 2015 both relayed that the mangrove drag effect is linked to the flow nature (creek and sheet flow). Sheet flow corresponds to the transportation over the vegetated platform through the mangroves, which becomes increasingly important with reduced channelization and at increasing water levels while creek flow dominates in channelized mangroves at low water levels (Montgomery et al. 2018). For sheet flow, Montgomery et al (2018) showed that vegetation properties were important by impeding water exchange across the forest, reducing water levels and slowing down the flood wave propagation in agreement with our results, while for creek flow, the authors highlighted a minimal mangrove contribution to the flow restriction. In the Somone Lagoon, because of the spatial mangrove distribution, tidal circulation allows both sheet and creek flow. Tidal dissipation by the mangrove appears to be related to the water level inside the mangrove, with a stronger dissipation by the mangrove during spring tides (e.g from 28/10 to 05/11 cf. Figure 4, Mangrove-PT2) and a lower tidal dissipation by mangrove during neap tides (e.g from 24/10 to 27/10 cf. Figure 4, Mangrove-PT2) when the tide penetrates very little into the mangrove. A similar behavior was previously obtained by Kitheka (1997) inside mangrove area in Gazy Bay (Kenya).
Contrary to the result obtained by Wu et al 2001 in the Merbok Estuary (Malaysia), showing rather a decrease in both flood and ebb currents in a configuration without mangroves compared to the case of taking into account the mangrove where they observed an acceleration of the floods and ebbs currents in channels, our study shows weaker currents when the mangrove is represented (Fig. 5). However, both studies are supported by local measurements, which makes difficult any generalization.
5.2.2 Frequency-dependence of tidal dissipation
To spatially assess the mangrove effects on tidal dissipation at the lagoon scale, the main tidal harmonic constituents are compared in the presence and absence of vegetation (Fig. 6). Both with and without vegetation, the semi-diurnal tidal wave M2 first decreases from 0.45 m to about 0.35 m (~ 30%) when propagating through the inlet, a process well documented in other similar shallow inlets (e.g. Bertin et al. 2009; Dodet et al. 2013). When entering the mangrove, M2 further decreases down to 0.3 m or less when the vegetation is accounted for, while it remains constant when vegetation is not represented. Diurnal tidal waves suffer much less dissipation when propagating through the inlet and on decrease by a few %. When reaching mangrove areas, diurnal waves also suffer less dissipation than semi-diurnal waves and their amplitude only decrease by 10 to 15%. On the opposite, the M4 overtide mostly develops inside the lagoon and continuously increases from the inlet region to the upstream part of the lagoon. Unlike diurnal and semi-diurnal waves, the amplitude of M4 is increases by 25–40% when the vegetation is accounted for.