It All Adds Up: Modeling the Baobab’s Survival

Keywords: baobab, mathematical modeling, spatial distribution, growth patterns, climate, fractal-branching, longevity

Abstract

Mathematics offers a way to understand how the baobab (Adansonia) grows, spreads, and adapts to its surroundings, uncovering patterns that explain and analyze its survival today and its prospects for the future. Its growth, drought tolerance, and remarkable carbon storage capacity reflect an adaptable strategy shaped by environmental pressures. Mathematical models can explain far more than might be expected at first glance; they are responsible for predicting and determining complex natural behaviors such as how baobabs are distributed across landscapes and how they may respond to future climates. In studying spatial distribution, tools like Voronoi tessellations, Ripley’s K-function, and nearest-neighbour analysis reveal whether trees cluster, spread evenly, or occur randomly, showing how temperature, elevation, and seed dispersal shape their arrangement. Through climate modelling, physical equations such as the Navier–Stokes system simulate atmospheric changes and project how shifts in temperature and rainfall could alter the baobab’s range over time.

Introduction

Across the dry plains of Africa and the red soils of Madagascar, few sights are as striking as the baobab (Adansonia), as seen in Figure 1.

Fig. 1. Avenue of Baobabs in Morondava, Madagascar (DeWeese, 2025).

The baobab has evolved a range of adaptations, both structural and physiological, that enables it to survive in environments where water and nutrients are scarce. The baobab’s strategies can be understood through chemistry, physics, and mathematics alike.

From a chemical perspective, the baobab’s longevity arises from antioxidant-rich tissues that protect against heat- and light-induced stress. Its parenchyma layers serve as natural water reservoirs, while polysaccharides function like hydrogels to retain moisture and preserve cellular structure. Symbiotic fungi further support survival by enhancing nutrient recycling and uptake in poor soils. From a physical standpoint, the baobab’s fibrous, lightweight wood provides strength and flexibility, allowing the trunk to remain stable even when hollow. Its geometric structure distributes mass efficiently for balance, while the thick bark and high heat capacity regulate temperature and prevent heat damage. Its adaptive growth patterns, drought tolerance, and efficient carbon storage support survival during early development and beyond. Together, these characteristics explain how the baobab endures for centuries, thriving in environments where few other trees can persist.

Yet beneath these biological and structural features lies another layer of understanding, one grounded in mathematics. Patterns of growth, spatial organization, and climate interaction reveal that the baobab’s survival follows quantifiable relationships that link biology, physics, and environment through mathematical principles.

Spatial Distribution

Spatial distribution refers to how individuals or objects are arranged within a given space (Borregaard et al., 2008). In the case of baobabs, it describes how the trees are positioned in their environment. Mathematical tools are used to analyze this distribution to determine whether individuals are clumped together, evenly spaced, or randomly scattered. Two main factors influence these patterns: first-order effects, which relate to environmental conditions such as soil type, water availability, or sunlight that affect where individuals can exist, and second-order effects, which describe interactions between individuals themselves, such as competition or facilitation. [hyperlink to competition 2021] Studying spatial distribution is important because the patterns observed reveal how plants survive, reproduce, and interact with their surroundings. For baobabs, this analysis helps explain how the species adapts to harsh environments (Ben-Said, 2021).

First-Order Effects

First-order effects are largely determined by environmental factors such as temperature, altitude, and soil quality. For the baobab, information on spatial distribution is limited; however, one study conducted in Zimbabwe has examined these first-order effects. The type of modelling they used was maximum entropy, which is an ecological niche modelling that has higher predictive power, works with presence-only data, and selects ecologically relevant variables. Zimbabwe has roughly 4.21 million baobabs, and in this study, they focused on analyzing the distribution at different altitudes and at different temperatures. Figure 2a shows that the highest density of baobab trees is in the north and south of the country. When comparing these findings to Figure 2b, a map showing the change in elevation, it is found that most baobabs grow in lower elevations between 350-700 m (Moyo et al., 2019).

Fig. 2. (a) The spatial distribution of baobabs in Zimbabwe. (b) The change in elevation over Zimbabwe (Moyo et al., 2019).

Additionally, in this study, the baobab’s distribution was also influenced by the mean temperature of the driest quarter and the mean temperature of the coldest quarter. Baobabs can tolerate high temperatures up to 40 - 42°C; however, the ideal environment for growth is between 15 - 24°C. The mean temperature of the driest quarter measures how hot or cold it gets when there is little to no rain. During this period, baobabs rely on moisture stored in their trunks; therefore, if the temperature is too hot, the water evaporates more quickly, drying out seedlings and hindering the survival of the tree. The mean temperature of the coldest quarter measures how cold it gets in the coldest part of the year, which for Zimbabwe is around April to July. If temperatures are too cold, seed germination and metabolic activity slow down, and frost can kill younger trees. This explains why in high altitudes where it is very cold, baobab growth is limited, and in lower lands and valleys where it is warmer, there is greater growth (Moyo et al., 2019).

Overall, the temperature and altitude of the area are key predictors of the spatial distribution of baobabs. Clusters of baobabs mostly appear in temperature-friendly zones, which are zones of low altitude that vary in temperature from 15 - 24°C. In higher altitudes, there are a few baobab trees; however, the extreme temperatures lead to sparser tree distribution (Moyo et al., 2019).

Second-Order Effects

Second-order effects are based on the interactions between surrounding organisms, such as competition or facilitation. In another study done, this time in southeastern Zimbabwe, they compared two protected areas to analyze the spatial distribution of baobab trees. The first area was the Save Valley Conservancy (SVC), which has a high elephant density and many herbivores, and the second area was the Chipinge Safari Area (CSA), which has few herbivores and no elephants. These two areas are complete opposites in terms of the level of competition that is present, and the reason these areas were chosen is to analyze the effect that competition from other species has on the distribution of baobabs (Khosa et al., 2023).

To accomplish this, baobabs within 1 km² plots were sampled, and data was recorded on the location, height, diameter, and canopy of each tree. The sampled trees were then classified into two categories: “large trees,” with a circumference ≥ 5 m, and “small trees,” with a circumference < 5 m. The analysis employed mathematical tools including Voronoi tessellations, Ripley’s K-function, pair-correlation function, and nearest-neighbour analysis. These methods were used to examine competition and interactions between small and large baobabs (Khosa et al., 2023).

The results found from this study, when comparing the two areas, were that the baobabs in CSA were found to be larger in terms of height, canopy and basal area and in the SVC area, it was found that the baobabs have a higher tree density seen in Figure 3. On top of this, it was found that small baobabs grow in clusters only 0-50 m apart, likely due to suitable micro-sites and not facilitation. There is a potential influence of animals such as elephants and other herbivores helping spread seeds, affecting where new baobabs appear. This helps explain why in the SVC area, which is high in elephant and herbivore density, there is higher baobab density of younger trees. Finally, large baobabs were found to be randomly distributed (Khosa et al., 2023).

Fig. 3. Canopy cover versus the distance between surrounding baobabs in the (a) SVC area and (b) the CSA area (Khosa et al., 2023).

Altogether, this study found that in terms of the second-order effects, the spatial distribution of baobabs is influenced more by environmental heterogeneity and seed dispersal than by competition. There are some signs of competition, but it is weak in both areas studied, and there are also very few plant-to-plant interactions, as younger baobabs do not rely on older baobabs for survival (Khosa et al., 2023).

Mathematical Tools Used in These Studies

Instead of always relying on the analysis of maps to determine the spatial distribution of the baobab tree, mathematicians use mathematical functions to analyze these patterns. Voronoi tessellation is one of these methods, and it is a method that divides space to see how trees interact with neighbours. In terms of the baobab, the way this method works is, mathematicians analyze the area they want to study and plot the location of each tree observed as a dot on their diagram. Each tree acts like a seed point in space, and then a polygon is drawn around the point, representing the tree's territory and influence zone. This creates a Voronoi diagram that looks similar to the those seen in Figure 4.

Fig. 4. Voronoi tessellation diagrams demonstrating Voronoi cells of (a) Two points, (b) Three points, (c) Five points, (d) Many points (Zhang & Wong, 2018).

The size and shape of each polygon, which is also called the Voronoi cell, help analyze the competition with neighbouring trees of this specific area. Larger cells indicate less nearby competition, as the trees are more widely spaced. In contrast, smaller cells signify more immediate neighbours, leading to greater competition among the trees (Zhang & Wong, 2018). The mathematical definition of Voronoi tessellation is given by “For a set of points {p₁, p₂, ... , p_n} in a plane, a Voronoi diagram divides the plane into n regions. The i^th region, V_i, is the set of all points x such that the distance from x to p_i is less than or equal to the distance from x to any other point p_j where j does not equal i ” (Horn & Weber, 2004). In two-dimensional (2D) Voronoi cells, the boundaries between regions are determined using Euclidean distance, that is, the straight-line distance between points, calculated using Equation 1 (Zhang & Wong, 2018).

Along with Voronoi tessellations, Ripley’s K function and the pair-correlation function are utilized to measure and compare clustering versus randomness. The Ripley’s K-function measures overall clustering and dispersion over all distances up to r, whereas the pair-correlation function measures a detailed, distance-specific interaction exactly at r. Essentially, the pair-correlation function is a more precise version of Ripley’s K-function. It measures point density changes at a specific distance r from a reference point and is better at detecting short-range interactions than the K-function. The purpose of these functions is to help measure the spatial distribution of the baobab trees across a region and determine if the trees are clustered, random, or regularly spaced. The way it works is that for a set of points (trees) recorded, the function is able to calculate the expected number of points within a distance of a certain point. This value is calculated using Equation 2, where A represents the total study area, n is the number of points, d_ij is the distance between points i and j, and I( d_ij≤ r) equals 1 when the distance is less than or equal to r, and 0 otherwise. This number is then compared to an expected number that is found under the conditions of complete spatial randomness (CSR). The results from these functions show that if the value calculated is greater than the CSR, the points are clustered, if it is equal to the CSR, the points are randomly distributed, and if it is less than the CSR, the points are evenly spaced (Dixon, 2002).

Finally, nearest-neighbour analysis is used to help determine the spatial distribution of baobab trees, and it is a very similar method to Ripley’s K function. This method sees if tree size affects spacing and is measured by Equation 3 where λ is the density of points and E(d) is the expected nearest-neighbour distance if points were random. This number is also compared to the CSR and works the same way as Ripley’s K function to determine if the trees are clustered, dispersed, or random. The difference between the two functions is that the nearest-neighbour analysis determines how the trees are distributed based on the average distance to the nearest neighbour, whereas Ripley's K function determines this based on how the patterns change across multiple scales by using all neighbours within a given radius (Parihari et al., 2023).

Climate Modeling

In the last two papers written on the baobab, the physics-oriented and chemistry-oriented essays, numerous adaptations that the baobab has developed to survive and thrive in its environment were discussed. From its hollow trunk to water storage to heat capacity, all these design solutions are crucial for its survival. However, as we currently advance in the Anthropocene, the climate is evolving at an ever-increasing pace, and all species will be affected. The following section will utilize climate change modeling to explore the potential impacts on the baobab and the adaptations it implements.

Models

First, there is a broad range of climate models that one can use to predict climate; all of them, though, rely on mathematical formulas based in chemistry, physics, and biology. Due to their wide range of applications, climate models are utilized in various fields (Gresham College, 2018). Here, a distinction must be made between weather and climate; weather is a short-term prediction (tomorrow or in 48 hours), whilst climate predictions are based on longer scales (years) (Gresham College, 2018). A nice visualization of specific models that exist is shown in Figure 5.

Fig. 5. Graph showing a variety of models used to predict conditions on a short to long scale and a small to a large area (Gresham College, 2018, https://www.youtube.com/watch?v=w4O4jK-lZrI).

While all the models shown in Fig. 4 have different scales and applications, they all rely on complex equations known as the Navier-Stokes equations (Scaife et al., 2007). These are a set of non-linear partial differential equations, which can allow, for example, the modelling of the atmosphere as a continuous, compressible fluid (Scaife et al., 2007).

Equation 4 shows the mass continuity for the atmosphere in 3 dimensions, while Equation 5 shows the energy continuity of the atmosphere in 3 dimensions (MIT, 2008). However, there is famously no analytical solution to the Navier-Stokes equations, meaning that we cannot find a closed-form solution and must rely on approximations and numerical methods (MIT, 2008).

Now that we have these powerful mathematical models, how can they help predict the effects of climate change and the impacts on the baobab?

Baobab Future Climates

Now that we have an understanding of the mathematics behind climate modeling, simulations can be run, often with multiple models combined to predict what the climate will look like in the future. Moreover, multiple potential scenarios can be run to get a better understanding of possible future paths. For example, in an interesting study predicting future inhabitable zones for the baobab in Sudan, two possible scenarios were studied (Gadallah et al., 2025). The future climate data were derived from the Project Phase 6, a global event where climate scientists run the same climate experiments so that their models can be compared (Eyring et al., 2016). The two scenarios, SSP245 and SSP585, represent intermediate to high emission trajectories, respectively. Also note that 3 global climate models (GCMs) were employed to enhance the robustness of future predictions (these GCMs are a type of climate model described earlier), and these were evaluated over two timestamps, 2021-2040 and 2040-2081 (Gadallah et al., 2025). The findings of the researchers are shown in Table 1 below.

Table. 1. Distribution of baobab habitat in Sudan under current and future projections with rates of change. Gain (areas newly becoming suitable in future scenarios); Loss (areas currently suitable that become unsuitable); Stable (areas remaining suitable across different time periods); Unsuitable (areas remaining unsuitable across current and future periods) (Gadallah et al., 2025).

Interestingly, in the intermediate carbon emission scenario, SSP245, at the end of the century, the baobab suitable land actually increases compared to the current suitable area. Moreover, in the high carbon emission scenario, the baobab gains suitable area for the first timestamp, 2021-2040, but loses around 7 percent of its suitable area by the end of the century, compared with its current-day area (Gadallah et al., 2025).

However, while the area might grow or shrink, the makeup of the climate is not static (Agbohessou et al., 2025). Indeed, a research group studying baobab climate distribution using the model SSP585 until the year 2070 found that climates shifted and expanded as seen in Table 2.

Table. 2. Showing expansion and shrinking of various climates from projections of the year 2070, adapted from (Agbohessou et al., 2025). Niche dynamic indices, Schoener’s D index. Rho: Spearman correlation.

Before looking at the results, different niche dynamics statistics are used that need explaining. First, Schoener’s D represents the extent to which two pairs of species, in this case, baobabs, form different climates, may interact in the same space (Smith, 2021). It is given by Equation 6.

Here, P_x,I, and P_y,I, represent the frequency of each species. Low values of D mean low overlap (Smith, 2021). In Table 2, Rho denotes Spearman’s correlation, which measures how strongly two variables are related by comparing the order, or rank, of their values. It ranges from -1 to +1 (Geeks For Geeks, 2025). A positive value means the variables increase together, a negative value means one increases while the other decreases, and a value of 0 means there is no relationship between them. The formula for Spearman’s correlation is given by Equation 7.

Where n represents the total number of paired observations in the dataset, and dᵢ represents the difference between the rank of each observation in variable 1 and its rank in variable 2. The difference in rank between two variables could also be explained as follows: neighborhood A is ranked 4^th in income but 6^th in rent; therefore, d_i = 4 – 6 = -2 (Geeks For Geeks, 2025).

Coming back to the results from Table 2 (Agbohessou et al., 2025), the low Schoener index indicates that the environmental niches of the baobab are quite distinct and do not overlap much. Moreover, the near-zero Spearman correlation indicates weak spatial associations in environmental suitability between each zone. Or in other words, if one zone is highly suitable for the baobab, it does not mean that the other one it is being compared to will be too.

Table 2 also shows that while some baobab in certain climatic zones remain stable (for example, arid to semi-arid with 97% stability), other baobab groups will have a dramatically shifting climate (for example, arid to humid with 30% stability) (Agbohessou et al., 2025).

While the models do show how climates may contract and vary with time, they do not show the specific impacts that they could have on individuals of the species. Although linking these two is no easy task, some inferences from the data explored above could illuminate the effects on the baobab.

Effects on the Baobab

Phenology is defined as a “periodic biological phenomena that are correlated with climatic conditions” (Merriam-Webster). In the case of the baobab, this could relate to environmental triggers that determine the timing of leaf production. What some researchers found was that day length and water stress affected leaf phenology (Di Lucchio et al., 2018). However, another study found that only rainfall affected leaf phenology rather than temperature and day length (Venter & Witkowski, 2019). These contrasting findings suggest that baobab leaf phenology may not be driven by a single dominant factor but rather by a combination of environmental cues that vary depending on regional conditions, in which slight changes to climate can wildly alter leaf flushing and growing. As we have seen in the results discussed above, the climates of the baobab will dramatically shift, and this may alter and disrupt the normal leaf cycle. Moreover, a decrease in rainfall has already been observed in southern Africa (Alahacoon et al., 2021): leaf phenology is already being affected. While this is a very specific example of how the shifting climates could affect the baobab, the principle could be applied much more broadly. For example, it could affect flowering, germination, seedling establishment, thermal stress, and many other aspects crucial for baobab survival and proliferation (Gadallah et al., 2025).

However, there is cause for optimism. As was shown in Table 1, under an intermediate carbon emission scenario, suitable land for the baobab actually increases at the end of the century compared to today. Furthermore, all the adaptations the baobab already employs that have been explored in the previous physics and chemistry papers show its resilience to an already harsh climate. Its porous tissue, along with the larger vacuoles in parenchymal cells, allows it to store more water. The fibrous tissue also creates a thermal insulation from the hot African air. Because of these adaptations and many others, the baobab appears to be the survivalist expert.

While it is challenging to determine the exact effects of climate change on the baobab, some insight can be gained from mathematical models. Ultimately, time will tell if the baobab continues to thrive. However, if any tree could adapt and thrive in the shifting climate, it would certainly be the “tree of life”.

Growing A Baobab

There are many factors that influence the growth of the baobab seedling; given that it inhabits a dry environment, the seedling must be resilient enough to withstand the climate and unintended changes. Water scarcity, sun exposure and temperature are all things that could impede the tree’s survival, as well as regulate the flowering and production of fruit of adult baobabs. Carbon makes up most of the content of the tree, contributing to the carbon content of the ecosystem it is a part of.

Carbon Uptake

The importance of carbon in the earth’s energy cycles cannot be stressed enough by the countless studies focused on defining the structure of an ecosystem (Marelign & Mekonen, 2022). Forest biomass is used as a metric in such cases, quantifying forest aboveground biomass (AGB) to determine carbon stocks and balances. In the Dirmaga Watershed in Ethiopia, home to many species including Adansonia digitata, a baobab species, carbon stock assessments were conducted with a high degree of uncertainty in mind using Species-specific allometric Equation 8. AGB is calculated according to the height (H) and the diameter at breast height (DBH), at around 1.3 meters or 4.3 feet tall (Marelign & Mekonen, 2022).

The principles of Pearson et al. (2005) allow the aboveground carbon and the AGB CO₂ equivalent (AGB CO2eq) to be found in Equations 9 and 10 respectively (Marelign & Mekonen, 2022, Pearson et al., 2005).

The total AGC and belowground carbon (BGC), the latter being 10% of the above-ground tree biomass, of the baobab tree is found to be 2657.4 t/ha, one of the largest portions among species studied in the area. Its AGB content reached 4920.0 t/ha, the second highest after the Kirkira tree (Marelign & Mekonen, 2022). This correlates with the high quantity of parenchyma cells found in the baobab stem, as they tend to have a high tendency to store carbon.

A tentative spectral index to map vegetated regions from non-vegetated regions would be NDVI, which is a term that indicates photosynthetically active radiation (Marelign & Mekonen, 2022). The spatial distribution of AGC in Dirmaga Watershed can be found in Figure 6 where 25.6% has no vegetation, and 5.5% of the map includes the highest estimated value of AGC of 331.35 to 532.86 t/ha. The latter seems to overlap with where the Dirmaga river runs in Figure 7. Further study is needed to properly represent the various species and their contributions to the total AGC of the area studied (Marelign & Mekonen, 2022).

Fig. 6. The above ground carbon (t/ha) distribution of Dirmaga Watershed (Marelign & Mekonen, 2022).

Fig. 7. The location map of the study area (Marelign & Mekonen, 2022).

Water Stress-induced Growth Rate

Drought affects seedling growth in ways which are becoming increasingly clearer (Bouda et al., 2015). In a study conducted by Bouda et al, they found that compared to seedlings under high water content (HWC), those under medium water content (MWC) and low water content (LWC) had a diameter and a stem height that were significantly smaller. Their rate of diameter growth also decreases under water stress, especially in seedlings from Mkundi in the Eastern province, as seen in Figure 8. Even when that region is excluded, seedlings from Kolangal and Komodiguili still show strong reductions in growth, indicating that they are more sensitive to dry conditions than the others. Interestingly, between 12 and 18 months, dry matter increased again for MWC and LWC trees, after the cold season had passed. Part of the initial decrease can be attributed to the shedding of leaves for 6 months (Bouda et al., 2015).

Fig. 8. Effects of provenance and water regime on relative growth rate of root collar diameter of Adansonia digitata seedlings. Error bars denote upper and lower 95% confidence intervals (n = 3) (Bouda et al., 2015).

Despite the intuition, the drought stress is not a major factor in baobab survivability as all the survival rates of all provenances, with the exception of Mkundi, was above 80% (Bouda et al., 2015). However, the height and diameter growth of seedlings slows down during the cold and dry periods in the Sahelian environment. During the first 18 months, biomass was gathered in the roots when water stressed, but shifted to allocating more biomass to the shoots, possibly as a way to facilitate carbon capture (Bouda et al., 2015).

Understanding the biomass distribution and carbon uptake of the baobab tree, especially the variables that affect it, could help scientists map out the general layout of a forest’s carbon distribution. Currently, one of the bigger challenges in mapping is looking at individual trees. Instead, satellite imagery and artificial intelligence are being developed as tools for large-scale initiatives (Tucker et al., 2023). Having a clearer picture of how water affects the growth rates in terms of diameter and stem height could give additional context to include in the calculations.

Seedlings Growing in Shade

Similarly, knowing that baobabs are often found in arid climates, the effect of sunlight on their growth rate could be helpful for mapping, as it relates directly to the biomass. Day length has also been related to flowering, which is crucial for reproduction. Growing baobab seedlings in the nurseries requires a proper look into the effect of shade on the development of the seedling (Egbadzor et al., 2023). The rates of growth are indicative of what primarily affects the baobab’s health. By compiling how studies optimize growing efficiency, it is possible to understand which conditions reflect traditional agriculture in these areas. Farmers often opt to use topsoil, which is poor in nutrient content, mixed with animal manure due to the soil’s low cost; therefore, differentiating soil from a soilless medium is imperative when replicating the expected conditions of growth. Bacteria and pathogens can be problematic in the latter case; therefore, a soilless medium would be a safer alternative. Given that baobabs can remain in their juvenile state for up to two decades, grafting is being used as a countermeasure while improved growth conditions continue to be refined (Egbadzor et al., 2023).

Under the full sun, seedlings grown in a soilless medium had around 34 leaves by the ninth week, whereas those grown in soil only had around 14. In Figure 9, the seedlings are shown with their different heights and leaf quantities. (Egbadzor et al., 2023). Likewise, in the soilless medium, partially shaded seedlings also had 13.8 leaves. The higher number of leaves in the presence of sun and daily watering indicate that the baobab, well adapted to the arid and sunny conditions of its habitat, is able to fully profit from the sunlight and the nutrients in the soilless medium. Unlike seedlings of other species, the tree does not need to be partially shaded, instead it handles direct sun better than even its counterparts. In terms of stem diameter, again the largest of 10.4 cm belonged to those raised in the soilless medium under the full sun (Egbadzor et al., 2023).

Fig. 9. Photographs of baobab seedlings under different treatments in the ninth week. (a) Represents plants raised in soilless medium in shade, (b) raised in soil in shade. (c) Represents plants raised in a soilless medium in full sun, (d) raised in soil in full sun (Egbadzor et al., 2023).

Chlorophyll content flipped halfway throughout the study (Egbadzor et al., 2023). This was detected using a SPAD meter, which gives a unitless value corresponding to the relative chlorophyll content in the leaf. By week 3, the highest value of 29.6 SPAD value was attributed to seedlings grown in partial shade in soil, but by week 9, the soilless medium in the sun had reached a SPAD value of 33.2 as seen in Table 3. Since nitrogen and potassium are abundant in the medium, being active ingredients in chlorophyll formation, they would have contributed to the high content. However, due to the porous nature of the medium, the seedlings were subjected to water stress until week 6, when frequent watering became available (Egbadzor et al., 2023).

Table. 3. Results of analysis of variance for the mean chlorophyll content of baobab seedlings raised in different media under different shading regimes (Egbadzor et al., 2023).

Flowering and Fruit

Unlike leaves, flowering in baobabs tends to follow a more deterministic mechanism, following day-length or temperature influence. In a study conducted by Venter & Wikowski, the peak flowering happened around the same time in November each year, despite the rains coming earlier in Year 2 as seen in Figure 10 (Venter & Witkowski, 2019). The length of flowering adapted to the early rains, showing that the baobab is flexible enough to take advantage of the rains (Venter & Witkowski, 2019).

Fig. 10. (a) Proportion of adult trees (n = 79) in flower and exhibiting peak flowering, as well as rainfall each month in Year 1 and (b) Year 2. (c) Proportion of trees flowering for different lengths of times in Year 1 and Year 2 (Venter & Witkowski, 2019).

Consequently, trees that produced 50 to 400 flowers had a fruit-set average of 30% (Venter & Witkowski, 2019). This is significant as for trees with over 1500 flowers the value was less than 1%, and trees with less than 50 flowers did not have any fruit whatsoever. With regression analysis, a weak negative trend can be observed in Figure 11 ( R² = 0.0973, P = 0.0065). The proportion of trees as a function of fruit-set percentages has an inverse J-shaped distribution, where the majority of cases will be found on either extreme. 23% of baobabs had no fruit and 31% had less than 1% fruit-set. This may be due to meiotic aberrations, leading to sterility as often seen in polyploids. Furthermore, it could be a form of sexual dimorphism, where the function is altered instead of a morphological feature. With low geographic diversity but high diversity within populations, baobabs are highly polymorphic and can have unusual segregation patterns due to polysomic inheritance, allowing for evolutionary change to occur (Venter & Witkowski, 2019).

Fig. 11. Proportion of fruit set (a) per number of flowers-class, (b) as well as the proportion of trees per fruit-set-class (n = 79) (Modified from Venter & Witkowski, 2023).

Conclusions

The baobab’s survival can be understood through the language of mathematics, which reveals the patterns behind its growth, organization, and adaptability. Across every scale, from the early development of seedlings to the spread of trees across vast landscapes, measurable relationships explain how this species endures. Mathematical models reveal that what appears to be random natural behavior actually follows patterns of logic and efficiency, uncovering the design solutions that allow the baobab to adapt and survive.

Through spatial analysis, it becomes clear that baobabs adapt their growth to crowding and limited resources. They “partition space,” arranging themselves to reduce competition and share what the landscape offers. Methods such as Voronoi tessellations and Ripley’s K-function show that baobabs naturally space themselves at efficient distances, using geometry to minimize competition and make the most of available resources.

Climate models built on the Navier–Stokes equations help predict how changing temperature and rainfall will affect the baobab’s range. When global climate models were run under two scenarios, SSP245 (moderate emissions) and SSP585 (high emissions), they showed that under moderate conditions, suitable habitats for baobabs may expand, while under high emissions, only a slight decline is expected by the end of the century. These results highlight a design that is flexible and resilient. The baobab’s porous wood, fibrous tissue, and water-storing vacuoles give it the ability to withstand and even benefit from environmental change, reflecting a structure naturally built for adaptation.

The baobab’s overall growth strategy reflects this same resilience. Whether adjusting biomass allocation under drought, maintaining high survival across provenances, responding flexibly to light and nutrient conditions, or storing significant carbon within its massive parenchyma-rich stems, its growth is dynamic and efficient. Across seedlings and mature trees alike, the baobab continuously optimizes its structure and physiology to match changing conditions.

From its growth patterns to its spatial arrangement and climatic adaptability, the baobab demonstrates a single unifying principle, design through optimization. At larger scales, baobabs demonstrate this adaptability not only in how they position themselves across the landscape but also in how they adjust to changing climates, store carbon efficiently, and maintain growth under varying environmental pressures. Each level reflects the same mathematical order, minimizing stress, maximizing survival, and maintaining balance in a changing world.