On the shoulders of giants: Factors impacting baobab survival in the Kruger National Park

By Dr Dave Thompson, Biodiversity Scientist, SAEON Ndlovu Node and Rob Taylor, Ecologist, GroundTruth Tweet

To stand on the shoulders of giants - a concept originating from the 12^th century, expresses the meaning of ‘discovering the truth by building on previous discovery’.

Not only does this notion define the ever-evolving nature of science, but it also succinctly highlights the need for ongoing observation and retrospective comparison. Or, put slightly differently, SAEON’s focus areas of long-term observation science and data archiving.

Figure 1. The savanna elephant (L) and the baobab (R) are the giants of the African continent. The destructive behaviour of the savanna elephant has attracted negative attention regarding the persistence of baobabs, but is this justified?

The African baobab (Adansonia digitata) - the giant of the African continent (Figure 1, right) - has highly specific requirements for recruitment, a slow life history that spans millennia, low dispersal capabilities, and experiences heavy utilisation. These characteristics reduce its ability to accommodate and respond to changing environmental conditions.

Consequently, an apparent lack of young trees and the high numbers of dead large individuals seen during aerial surveys carried out in the early 1990s in the Kruger National Park (KNP) prompted concern over the status, and likely future, of this protected-area population.

Figure 2. A composite image of a baobab individual in northern Kruger National Park - the left side of the image from a photograph taken in July 1995 by Dr Ian Whyte; the right taken in September 2013. Changes in the extent of elephant damage are clear.

Figure 3. Comparison of the size class distributions of sampled baobabs across areas differing in maximum fire return period calculated for 1941-2006. Fire return periods of less than 30 years resulted in lower proportions of smaller sized trees and deviation from the inverse J-shaped distribution that typifies healthy populations.

Near 20-year data series

An MSc study recently completed by Rob Taylor under the supervision of Prof. Ed Witkowski (University of the Witwatersrand - WITS) and Dr Dave Thompson (SAEON), and with financial support from the National Research Foundation, SAEON, WITS and the Botanical Education Trust, resampled 758 baobabs previously sampled in 1995/6 by Dr Ian Whyte (SANParks, retired) and 2001 by Michele Hofmeyr (SANParks). The outcome: a valuable near 20-year data series of population structure and change in the status of individual baobabs (Figure 2) across the dominant vegetation and climatic zones of the species’ KNP range.

The destructive impacts of another African giant - the elephant (Figure 1, left), have borne the brunt of blame for reduced baobab survival in conservation areas. However, and using the existing SANParks data as the starting point, this project revealed that survival is impacted by three main culprits: i) maximum fire return interval (the longest interval that a seedling or sapling baobab has to establish in the absence of fire); ii) mean annual temperature; and iii) degree of elephant damage.

A comparison of the population structure of baobabs growing in areas with differing maximum fire return intervals suggests that smaller baobabs (with stem diameter <2m) are particularly sensitive to fire, being comparatively absent where fire is more frequent and more common where fire is rare (Figure 3).

Those individuals most likely to escape the effects of being burned, grow where low fuel-loads cannot support fire, or where landscape features act as natural fire breaks.

Baobab populations growing across different mean annual temperatures (MAT) also had different size class structures, with the population growing in that area defined by the 23 ^°C MAT isotherm showing a healthy inverse J-shape curve (many young individuals, progressively fewer older ones) relative to the bell-shaped curves of populations growing in cooler regions (Figure 4).

Figure 4. Comparison of the size class distributions of baobab populations found in the areas defined by the 21, 22, and 23 °C mean annual temperature isotherms in the Kruger National Park. Survival through to maturity is most likely for individuals occurring in the warmer, more northern parts of the Park, where the healthiest population structure also exists.

Elephant damage is escalating

Favoured by elephants, the baobab is utilised destructively. This study suggests that the percentage and severity of damage to baobabs has increased in line with recent increasing elephant densities in KNP.

Elephants use their tusks to pierce baobab bark and then strip the bark both upwards and downwards. In this manner an elephant is able to ‘peel’ large areas of the stem (Figure 5). The preferred forage height of <3m suggests a convenient height for elephants to manoeuvre their tusks.

Figure 5. Baobabs are a species favoured by elephant as a source of food. Typically, the tusks are used ‘peel’ off vertical strips of bark and the moisture-rich underlying stem tissues. If repeated too frequently, the stem cannot heal and the extent and severity of damage becomes accumulative.

Elephants removed more bark in 2013 than they had in 1995/6 (Figure 6).

All baobabs >2m in stem diameter had some level of elephant damage and all baobabs >3.5m in stem diameter had more than just slight damage. This suggests that damage is accumulative and that frequent elephant returns to the same tree do not allow for baobabs to heal between visits.

Figure 6. Comparison of the percentage of the stem damaged per metre of vertical stem height in 1995/6 and 2013. Bars indicate standard error. Repeat photographs taken in 1995 (left, by Dr Ian Whyte) and during the recent survey (right) clearly show increased stem damage.

Figure 7. Comparison of the size class distributions of baobabs showing different levels of elephant utilisation in 2013. Small trees (<1m stem diameter) either escape detection and utilisation or, more typically, are heavily and often fatally impacted.

Further, the size class distributions revealed that small trees either displayed no or slight damage, or were severely/very severely impacted (Figure 7). The suggestion here is that these smaller individuals either escape detection and utilisation by elephants, or more typically, are found and then suffer severe and often (61%) fatal damage.

No seedlings were observed during any of the sampling campaigns since 1995, suggesting long-lived baobabs are reliant on the survival of mature trees in order for the population to persist. However, year-on-year mortality increased three-fold between 1995/6-2001 and 2001-2013, with most deaths being small individuals (<1m stem diameter), or individuals displaying more than moderate stem damage.

Predicted climate warming may favour the recruitment of new individuals into the population in the future, but the survival of these recruits is reduced by impacts from two factors that can both be managed - fire and elephants.

This project was made possible by the exemplary data records and data-sharing philosophies of SANParks and SAEON. The 18-year dataset represents a mere ~1% of a baobab’s possible lifespan - so continued monitoring of these trees will add significantly to our understanding of this species. This should be done at intervals relevant to changes in elephant density, fire management and climate.