Seedling emergence of Isoberlinia doka seeds depends more on storage duration than on collection condition under tropical ambient storage

Abstract

Introduction Seeds of Isoberlinia doka (Fabaceae), a keystone tree of West African woodlands, have been reported to be recalcitrant, with viability tightly coupled to moisture content and therefore presumed vulnerable to desiccation and thermal stress. Yet the practical limits of their handling after dispersal remain poorly defined, particularly under the high ambient temperatures prevailing in the non-refrigerated nurseries where the species is propagated. This study evaluated how collection condition (canopy versus ground) and storage duration affect seedling emergence capacity and emergence dynamics in Isoberlinia doka . Methods Seeds were stored without artificial cooling at ambient temperature (mean 34 °C) for 0, 2, 4 or 6 weeks, reflecting the resource constrained conditions of smallholder nurseries in West Africa. The experiment followed a completely randomized design with 8 treatments (2 collection conditions × 4 storage durations), each replicated 3 times (720 seeds). Emergence was monitored daily for 30 days and analyzed using beta regression (with the Smithson-Verkuilen transformation) for the final emergence proportion and a Cox proportional-hazards model, complemented by Kaplan-Meier estimation and log-rank tests, for emergence dynamics. Results Storage duration strongly affected emergence (p< 0.001), with a marked decline beyond two weeks. Freshly collected seeds reached the highest final emergence proportion (0.844). Two week-stored seeds reached about 0.55, and seeds stored for four or six weeks showed no emergence. The Kaplan-Meier emergence probability stabilized at 84.4% by day 14 for fresh seeds and at 55% by day 28 for two week-stored seeds. Collection condition affected emergence in a storage-dependent way, with a significant storage × collection interaction (p< 0.001). Canopy-collected seeds emerged more than ground-collected seeds when fresh (p = 0.004). After two weeks, the ranking reversed and ground-collected seeds emerged more (p = 0.008). Discussion and conclusion These results indicate that storage duration under ambient tropical conditions, rather than soil contact during early post-dispersal, governs the rapid loss of emergence capacity. This loss is consistent with the desiccation and temperature-sensitive behavior previously reported for this presumed recalcitrant species, although no physiological marker was measured here. We establish a practical two-week emergence threshold for seed sourcing, nursery management and forest restoration.