Alessandra Havinga
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alessandra Havinga has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Louis Nusbaumer
University of Geneva, Chambésy, Switzerland
2 shared publications - Patrick Ranirison
Association Famelona, BP 5147, Antananarivo 101, Madagascar
2 shared publications - Gregory Giuliani
Institute for Environmental Sciences, University of Geneva, 1205 Geneva, Switzerland
1 shared publication - Iharivolana
Mention Biologie et Ecologie Végétales, Université d'Antananarivo, BP 906, Antananarivo 101, Madagascar
1 shared publication - Johan Hermans
The Herbarium, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AB, UK
1 shared publication - Pascal Martin
Jardin Botaniques de Genève, Chemin de l’Impératrice 1, 1292 Chambésy, Switzerland
1 shared publication
Latest Publications
- Open Access
- 2 September 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (17) , 2952
Beyond Standalone Geo-Embeddings: Weighted Multi-Model Ensemble Prediction for Tropical Land-Cover Mapping
Accurate land-cover mapping in tropical regions remains challenging because of high environmental heterogeneity, complex vegetation structure, and persistent cloud cover. Recent geospatial foundation models provide pre-computed Geo-embeddings that offer a new representation of Earth Observation data (EO), yet their potential for detailed tropical land-cover mapping and their performance relative to, and in combination with, conventional Sentinel-1 and Sentinel-2 satellite image time series remains largely unexplored. This study evaluated AlphaEarth Foundation Geo-embeddings for detailed vegetation mapping in northern Madagascar and investigated whether combining them with satellite image time series processed using the Satellite Image Time Series (SITS) R package could improve classification performance. Geo-embeddings were classified using a Random Forest model (GEO), while four supervised classifiers were trained on Sentinel-1 and Sentinel-2 time series. GEO achieved the highest standalone performance (OA = 0.74), outperforming all classifiers trained on the satellite image time series. Probability-level ensemble models were then used to assess whether both data representations could be beneficially combined. The best-performing ensemble, combining GEO (75%) with Temporal Convolutional Neural Network (TempCNN; 25%), increased overall accuracy from 0.74 to 0.82. These results suggest that Geo-embeddings provide an effective standalone representation for detailed tropical land-cover mapping and indicate that combining them with conventional satellite image time series may exploit complementary information. As geospatial foundation models continue to evolve, understanding how they can be integrated with established Earth observation workflows may support future operational land-cover mapping.
- 2 September 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (17) , 2952
Open AccessBeyond Standalone Geo-Embeddings: Weighted Multi-Model Ensemble Prediction for Tropical Land-Cover Mapping
Accurate land-cover mapping in tropical regions remains challenging because of high environmental heterogeneity, complex vegetation structure, and persistent cloud cover. Recent geospatial foundation models provide pre-computed Geo-embeddings that offer a new representation of Earth Observation data (EO), yet their potential for detailed tropical land-cover mapping and their performance relative to, and in combination with, conventional Sentinel-1 and Sentinel-2 satellite image time series remains largely unexplored. This study evaluated AlphaEarth Foundation Geo-embeddings for detailed vegetation mapping in northern Madagascar and investigated whether combining them with satellite image time series processed using the Satellite Image Time Series (SITS) R package could improve classification performance. Geo-embeddings were classified using a Random Forest model (GEO), while four supervised classifiers were trained on Sentinel-1 and Sentinel-2 time series. GEO achieved the highest standalone performance (OA = 0.74), outperforming all classifiers trained on the satellite image time series. Probability-level ensemble models were then used to assess whether both data representations could be beneficially combined. The best-performing ensemble, combining GEO (75%) with Temporal Convolutional Neural Network (TempCNN; 25%), increased overall accuracy from 0.74 to 0.82. These results suggest that Geo-embeddings provide an effective standalone representation for detailed tropical land-cover mapping and indicate that combining them with conventional satellite image time series may exploit complementary information. As geospatial foundation models continue to evolve, understanding how they can be integrated with established Earth observation workflows may support future operational land-cover mapping.
- Open Access
- 1 June 2026
- Published by Society of Conservatoire at Jardin Botaniques de la Ville de Geneve in Candollea
- Vol. 81 (1) , 1-6
Platylepis avaratraensis (Orchidaceae), a new species from northern Madagascar and its taxonomic position in Goodyerinae
Les recherches taxonomiques et moléculaires sur les genres de la sous-tribu Goodyerinae de Madagascar et des îles adjacentes de l'océan Indien occidental sont discutées. Les caractéristiques de Goodyera R.Br. et Platylepis A.Rich. sont comparées, et leur histoire est présentée. Les espèces actuellement acceptées par les auteurs pour la région sont détaillées. Une nouvelle espèce d'orchidée, Platylepis avaratraensis Hermans & Nusb., est décrite dans la région de Sava, au nord de Madagascar. Elle est illustrée et comparée à d'autres espèces du genre. Elle est la plus proche de P. humicola (Schltr.) Schltr., mais se distingue par ses feuilles plus larges, ses bractées florales plus courtes, la couleur de ses fleurs, des sépales plus étroits et un labelle avec des callosités distinctes. La distribution et l'écologie de la nouvelle espèce sont discutées et son statut de conservation préliminaire est évalué comme «En danger» suivant les Catégories et Critères de la Liste rouge de l'UICN.
- 1 June 2026
- Published by Society of Conservatoire at Jardin Botaniques de la Ville de Geneve in Candollea
- Vol. 81 (1) , 1-6
Open AccessPlatylepis avaratraensis (Orchidaceae), a new species from northern Madagascar and its taxonomic position in Goodyerinae
Les recherches taxonomiques et moléculaires sur les genres de la sous-tribu Goodyerinae de Madagascar et des îles adjacentes de l'océan Indien occidental sont discutées. Les caractéristiques de Goodyera R.Br. et Platylepis A.Rich. sont comparées, et leur histoire est présentée. Les espèces actuellement acceptées par les auteurs pour la région sont détaillées. Une nouvelle espèce d'orchidée, Platylepis avaratraensis Hermans & Nusb., est décrite dans la région de Sava, au nord de Madagascar. Elle est illustrée et comparée à d'autres espèces du genre. Elle est la plus proche de P. humicola (Schltr.) Schltr., mais se distingue par ses feuilles plus larges, ses bractées florales plus courtes, la couleur de ses fleurs, des sépales plus étroits et un labelle avec des callosités distinctes. La distribution et l'écologie de la nouvelle espèce sont discutées et son statut de conservation préliminaire est évalué comme «En danger» suivant les Catégories et Critères de la Liste rouge de l'UICN.