This flagship species is fairly widespread but has a fragmented distribution with different subpopulations showing varying degrees of vigour: in the granite Haggeher mountains (particularly in the centre and western parts) Dracaena cinnabari is abundant at higher altitudes with populations showing good age structure and active regeneration (particularly in sheltered gullies and in the dense woodland amongst the granite pinnacles); on the limestone plateaus adjacent to the Haggeher (e.g., Firmihin) the woodland has a good age structure but shows no sign of regeneration; on certain outlying sites on the eastern limestone plateau (e.g., Hamadero) there are no signs of regeneration except in places inaccessible to livestock - these woodlands appear in good health but consist almost entirely of mature trees; finally over large areas the tree is absent except for small relict populations (Iksha) or as isolated individuals (Qatariyah).
It is likely that D. cinnabari was, in the past, widely distributed over large parts of Soqotra. What then could account for its decline? Information from local informants and the historic record suggests that there has been no change in traditional land management practices until very recently, and, as regards Dracaena specifically: (i) the harvesting of Dracaena resin has lessened over recent years as the market for it has declined, and (ii) the use of the growing leaflets for cordage has almost died out with the introduction of nylon rope. However, it is true that during severe droughts, leaflets are removed and cracked open as fodder for goats. Nevertheless, it must be noted that this feed causes scour in goats, and so can only be fed in very small quanitities. The same goes for the fruit which are gathered in the summer and fed to goats but especially to cattle. However, once more, large quantities of the berries make both ill, so they must be fed in very small quanities (typically 5–8 berries for a goat, for instance). A starving goat will indeed nibble at the softer, younger leaflets of a young plant if it can find absolutely nothing else to eat, this is not a common occurrence. It would therefore be an over-simplification to blame the decline on over-grazing. It seems likely that grazing is only a factor preventing regeneration when populations are already under stress. Regeneration is also particularly problematic in relict populations which lack the dense under-storey of shrubs, which provide a protective nursery for young plants in healthy Dracaena woodland. Herders who feed their cattle on Dracaena fruit report observing seeds sprouting in feed areas; however, these young seedlings have not been observed surviving the subsequent dry season
Islanders have observed that in areas where Dracaena is failing to regenerate younger trees are not branching out to form the typical umbrella-shaped canopy, but are producing taller trunks instead. The exception to this is the high Haggeher mountains, where what are seen as ‘normal’ young trees are still common. They offer various possible explanations for this. The trees do best in the areas affected by the mists, low cloud and, in particular, the constant drizzle of the monsoon. Indeed, they regard the trees as constituting a natural indicator of this micro-climate. Oral history reports that the extent of this area has decreased over past generations, and this is one reason for the trees fialing to flourish. Oral history also reports that in earlier times the crowns were frequently completely covered in flowering spikes, whereas now it is more common for only part, or one side, of the crown to have flowering spikes, and these are less productive. Another contributing factor is believed to be the increased feeding of livestock with the flowers and fruit. This is due to increased livestock numbers, the high cost of supplementary cereal feeds, and the pressure of the years of drought in the 1970s and 1980s. This sometimes resulted in the complete stripping of the fruiting stalks from trees. Herders say that if too few fructescences are left on a tree year after year, the tree begins to die back.
The best Dracaena woodland occurs in Rokeb di Firmihin and the northern end of the Igalis plateaus which run from behind the Haggeher mountains down to the southern escarpment. After the Haggeher mountains themselves, these are the areas most affected by the monsoon clouds and mists. Islanders also say that the best trees occur in areas where the terrain consist of solid rock pavement with extensive cracks, down which water and soil flows after rains, providing moisture and nourishment for the roots of the trees. It is not difficult to produce seedlings from seed: cattle herders even in the drier areas have observed that when they feed the ripe fruit to their cows many seeds sprout later in the rains and cloud of the winter season. However, as soon as the cloud lifts and sun comes out, they are burned off, or, if the seed has sprouted in areas of herbage, they are grazed down along with the grasses and herbs.
It seems likely that a major cause of the decline in extent and quality of Dracaena woodland is the very gradual drying of the Archipelago. No direct palaeoclimatic data is available from Soqotra, however, there is strong circumstantial evidence that the islands have been drying over the last few hundred years. Local expertise describes Dracaena as occurring only in those high plateaux regions affected by the mists and cloud of the monsoon season (a period when the rest of the archipelago suffers strong and dessicating winds). Oral history and traditional reports that not only has the extent of this cloud/mist coverage diminished over the years, but also its duration and continuity: what used to be a predictable 4–5 months of cloud and drizzle in the area (resulting in the total evacuation of both people and livestock from the region down to the lower hills) has become instead patchy and discontinuous.
Other potential threats to the long-term survival of Dracaena cinnabari are the over-exploitation of Dragon’s Blood - a resin obtained from the bark of the tree. At present there is little demand for the resin but an increase in demand could lead to rapid over-collecting. The almost complete eradication of the closely related Dracaena draco on the Canary Islands has been linked to over-exploitation of the trees for Dragon’s Blood in the Middle Ages (Lucas and Synge 1978). Another threat is from the use of the trunks to make traditional beehives. There was recently a case when over 20 trees were felled to make beehives for export to the mainland. This incident was generally condemned on the island but illustrates how a breakdown in traditional practices poses a very real threat to species on the islands.