HAL Portal UPF (Université de la Polynésie française)
Not a member yet
    3802 research outputs found

    Hybrid Deep Learning Model for Mean Hourly Irradiance Probabilistic Forecasting

    No full text
    International audienceFor grid stability, operation, and planning, solar irradiance forecasting is crucial. In this paper, we provide a method for predicting the Global Horizontal Irradiance (GHI) mean values one hour in advance. Sky images are utilized for training the various forecasting models along with measured meteorological data in order to account for the short-term variability of solar irradiance, which is mostly caused by the presence of clouds in the sky. Additionally, deep learning models like the multilayer perceptron (MLP), convolutional neural networks (CNN), long short-term memory (LSTM), or their hybridized forms are widely used for deterministic solar irradiance forecasting. The implementation of probabilistic solar irradiance forecasting, which is gaining prominence in grid management since it offers information on the likelihood of different outcomes, is another task we carry out using quantile regression. The novelty of this paper lies in the combination of a hybrid deep learning model (CNN-LSTM) with quantile regression for the computation of prediction intervals at different confidence levels. The training of the different machine learning algorithms is performed over a year's worth of sky images and meteorological data from the years 2019 to 2020. The data were measured at the University of French Polynesia (17.5770 • S, 149.6092 • W), on the island of Tahiti, which has a tropical climate. Overall, the hybrid model (CNN-LSTM) is the best performing and most accurate in terms of deterministic and probabilistic metrics. In addition, it was found that the CNN, LSTM, and ANN show good results against persistence

    La protection de l’emploi local en Polynésie française : la genèse

    No full text
    International audienc

    A Systematic Review of UAVs for Island Coastal Environment and Risk Monitoring: Towards a Resilience Assessment

    No full text
    International audienceIsland territories and their coastal regions are subject to a wide variety of stresses, both natural and anthropogenic. With increasing pressures on these vulnerable environments, the need to improve our knowledge of these ecosystems increases as well. Unmanned Aerial Vehicles (UAVs) have recently shown their worth as a tool for data acquisition in coastal zones. This literature review explores the field of UAVs in the context of coastal monitoring on island territories by highlighting the types of platforms, sensors, software, and validation methods available for this relatively new data acquisition method. Reviewing the existing literature will assist data collectors, researchers, and risk managers in more efficiently monitoring their coastal zones on vulnerable island territories. The scientific literature reviewed was strictly analyzed in peer-reviewed articles ranging from 2016 to 2022. This review then focuses on the operationalization of the concept of resilience as a risk management technique. The aim is to identify a procedure from raw data acquisition to quantifying indicators for the evaluation of the resilience of a territory and finally linking the analyzed data to a spatial decision support system. This system could aid the decision-making process and uses the islands of French Polynesia and its Resilience Observatory as a case study

    La construction d'un supermarché à Tahiti autorisée... pour la seconde fois

    No full text
    International audienceAprès l’annulation de sa première autorisation par la cour administrative d’appel de Paris, l’Autorité polynésienne de la concurrence (APC) vient d’autoriser une seconde fois la construction d’un nouveau supermarché sous enseigne Champion dans la commune de Paea à Tahiti

    Ecology of Endozoicomonadaceae in three coral genera across the Pacific Ocean

    No full text
    International audienceHealth and resilience of the coral holobiont depend on diverse bacterial communities often dominated by key marine symbionts of the Endozoicomonadaceae family. The factors controlling their distribution and their functional diversity remain, however, poorly known. Here, we study the ecology of Endozoicomonadaceae at an ocean basin-scale by sampling specimens from three coral genera ( Pocillopora , Porites , Millepora ) on 99 reefs from 32 islands across the Pacific Ocean. The analysis of 2447 metabarcoding and 270 metagenomic samples reveals that each coral genus harbored a distinct new species of Endozoicomonadaceae . These species are composed of nine lineages that have distinct biogeographic patterns. The most common one, found in Pocillopora , appears to be a globally distributed symbiont with distinct metabolic capabilities, including the synthesis of amino acids and vitamins not produced by the host. The other lineages are structured partly by the host genetic lineage in Pocillopora and mainly by the geographic location in Porites . Millepora is more rarely associated to Endozoicomonadaceae . Our results show that different coral genera exhibit distinct strategies of host- Endozoicomonadaceae associations that are defined at the bacteria lineage level

    Mobilisation et polarisation électorales en Nouvelle-Calédonie en 2022

    No full text
    International audienceNew Caledonia’s voters eligible to vote in the three consultative referendums of 2018, 2020, and 2021, were particularly solicited in the run-up to the presidential and legislative elections of 2022. Beyond the continuities, this article highlights the singularities of the party system and electoral behaviour on the occasion of the 2022 national elections in New Caledonia. The article begins by examining electoral mobilisation and demobilisation from a longitudinal perspective. At the same time, it examines them from a comparative perspective in relation to those observed, on the one hand, in France and its overseas territories, and, on the other hand, in the main components of the population of New Caledonia. Between stability and changes, this analysis of the 2022 national elections emphasises the uniqueness of the party system and the evolution of the electoral behaviours in New Caledonia.Les électeurs de la Nouvelle-Calédonie habilités à voter aux trois consultations référendaires de 2018, 2020, et 2021, ont été particulièrement sollicités en amont des élections présidentielles et législatives de 2022. Au-delà des continuités, cet article met en lumière les singularités du système partisan et des comportements électoraux à l’occasion des scrutins nationaux de 2022 en Nouvelle-Calédonie. L’article étudie les mobilisations et démobilisations électorales dans une perspective longitudinale. Il les examine parallèlement dans une perspective comparative au regard de celles observées, d’une part, en France ainsi qu’en Outre-mer, et, d’autre part, dans les principales composantes de l’électorat de la Nouvelle-Calédonie. Entre stabilité et changements, cette analyse des élections nationales de 2022 souligne la singularité du système partisan et l’évolution des comportements électoraux qui opèrent en Nouvelle-Calédonie

    Characterization of 35 new microsatellite markers for the blacktip reef shark (Carcharhinus melanopterus) and cross-species amplification in eight other shark species

    No full text
    International audienceBackgroundShark species are overfished at a global scale, as they are poached for the finning industry or are caught as bycatch. Efficient conservation measures require fine-scale spatial and temporal studies to characterize shark habitat use, infer migratory habits, analyze relatedness, and detect population genetic differentiation. Gathering these types of data is costly and time-consuming, especially when it requires collection of shark tissue samples.Methods and resultsGenetic tools, such as microsatellite markers, are the most economical sampling method for collecting genetic data, as they enable the estimation of genetic diversity, population structure and parentage relationships and are thus an efficient way to inform conservation strategies. Here, a set of 45 microsatellite loci was tested on three blacktip reef shark (Carcharhinus melanopterus) populations from three Polynesian islands: Moorea, Morane and Tenararo. The set was composed of 10 previously published microsatellite markers and 35 microsatellite markers that were developed specifically for C. melanopterus as part of the present study. The 35 novel and 10 existing loci were cross-amplified on eight additional shark species (Carcharhinus amblyrhynchos, C. longimanus, C. sorrah, Galeocerdo cuvier, Negaprion acutidens, Prionacea glauca, Rhincodon typus and Sphyrna lewini). These species had an average of 69% of successful amplification, considered if at least 50% of the individual samples being successfully amplified per species and per locus.ConclusionsThis novel microsatellite marker set will help address numerous knowledge gaps that remain, concerning genetic stock identification, shark behavior and reproduction via parentage analysis

    0

    full texts

    3,802

    metadata records
    Updated in last 30 days.
    HAL Portal UPF (Université de la Polynésie française)
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇