Wood spectral variability in Entandrophragma cylindricum across northern Congo: limited association with genetic structure despite high genetic diversity

Le résumé fourni par la source

Abstract Wood plays a crucial role in the economies of Central African countries, yet knowledge of the ecological and genetic determinants of intraspecific wood quality variation remains limited. In this study, we combined microsatellite-based population genetic analyses with short-wave near-infrared (NIR) hyperspectral imaging (HSI) of transverse wood slices to investigate the genetic diversity and wood spectral variability of three Entandrophragma cylindricum populations across distinct forest management units in northern Republic of the Congo. Genotyping of adult trees using 10 nuclear microsatellite loci revealed high levels of genetic diversity and low inbreeding across all populations. No significant genetic differentiation was detected among sites, although a weak but significant spatial genetic structure was observed, consistent with an isolation-by-distance (IBD) pattern, suggesting ongoing gene flow over the landscape. Hyperspectral imaging of transverse wood samples highlighted substantial inter-individual and site-related variability in NIR spectral signatures, with the highest spectral heterogeneity observed in the site characterized by the long-term logging history. However, this variability was not correlated with genetic distance, as confirmed by Mantel tests, indicating that site-specific context and logging intensity are the primary drivers of wood spectral variation. The results suggest that E. cylindricum populations retain considerable genetic potential for conservation and improvement. Moreover, the observed site-related variation in wood spectral traits highlights the importance of considering local growing context when interpreting wood spectral variability in tropical timber species. Further research combining near-infrared spectroscopy with detailed wood anatomical analyses and direct measurements of environmental variables would be necessary to disentangle the mechanisms underlying site-associated spectral variation.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.