1. Individual Planat Level

Do you think angiosperms or conifers deal better with dry conditions and why?

Fiona Fang, Trinity Hall

Water is indispensable for plant physiology. Its dipole nature and hydrogen bonding support cohesion and adhesion in the xylem, enabling transpiration-driven water transport against the gravitational force. Water facilitates photosynthesis, especially in the photolysis process which provides electrons for ATP and NADPH formation. Water potential gradients also regulate osmotic pressure, essential for nutrient uptake through root hairs. It also maintains turgor pressure, preserving cell structure and enables growth (Schulze et al., 2005). However, water scarcity under dry condition will disrupts these processes, leading to metabolic dysfunction, structural collapse, and even death. With climate change altering rainfall patterns and intensifying drought frequency (Dai, 2013), understanding plant adaptations to dry conditions is increasingly critical for predicting ecological resilience and informing conservation strategies. Conifers and angiosperms have developed distinct strategies to cope with water stress, which will be compared and evaluated in this essay from the following three perspectives: 1) water transport and cavitation resistance 2) stomatal control and the strategies balancing carbon assimilation and water conservation 3) leaf adaptations and morphological differences.

Water transport in both conifer and angiosperm plants occurs through the xylem via cohesion-tension mechanism (Dixon and Joly, 1895). It is driven by surface tension during leaf transpiration and the integrity of the water column is maintained by cohesion between the water molecules and adhesion between the water column and xylem conduit walls (Tyree 1997; Tyree and Zimmermann 2002). Tracheids, spindle shaped, very elongate xylem elements, are the only type of water-conducting cells in most gymnosperms. Vessel members are the principal water-conducting cells in angiosperms (though most species also have tracheids). They are more specialized than tracheids, and are characterized by areas that lack both primary and secondary cell walls, known as perforations, through which water flows relatively unimpeded from vessel to vessel (Britannica, 2024). The lack of cell wall and greater diameter of vessels compared to tracheids lead to a greater water conduct efficiency in angiosperms (Sperry, Hacke, Pittermann, 2006). Under dry conditions, however, normal water conduction will be disrupted. Due to low levels of water input from the roots, the tension in the xylem increases, eventually exceeding a critical point and leading to the breakdown of the water column, a process referred to as cavitation. The gas bubble (i.e., embolism) formed during cavitation in one cell can spread to other adjacent cells (air-seed), effectively blocking water movement. The ability for the plant to mitigate cavitation is dependent on the pit membrane, a tissue lies in the center of each pit, i.e. cavities in the lignified cell walls of xylem conduits. Pit membrane allows water to pass between xylem conduits but limits the spread of embolism. Angiosperm vessels have homogenous pit membranes with smaller pores, while tracheids in conifers are perforated by complex pits membranes with central, impermeable torus surrounded by permeable margo (Hacke, Sperry, and Pittermann, 2005).

For angiosperm, the cavitation resistance depends on the size of the largest pores in the pit membrane. Smaller pit pores produce greater capillary forces in a water-filled vessel, thereby air bubbles in a neighbouring empty vessel cannot enter with ease. However, angiosperm pit membranes lack a mechanism to completely seal off a pit (Delzon et al., 2010). Indeed, by comparing angiosperms with and without vessels, Sperry et al. (2007) argued that the evolution of vessels in angiosperms is compromised safety from cavitation as a cost. However, this is challenged by Trueba et al. (2016) who didn’t observe significant differences in embolism vulnerability between vesselless and vessel-bearing angiosperms in the tropical forest. On the other hand, Conifers possess an intrinsically safer hydraulic system due to their torus-margo pit membranes: In the center of each bordered pit is a thick, circular torus that is largely impermeable, and around it is a thinner, porous margo region that allows water flow (Hacke et al., 2015). When adjacent tracheids are both water-filled and under tension, the torus stays centered, and water moves through the coarse margo pores with little resistance. If an embolism forms in one tracheid, the pressure difference pulls the pit membrane toward the embolized side. The torus is pressed against the pit aperture, sealing the opening between the two cells. This is called pit aspiration, and it effectively isolates the air-filled tracheid, preventing the spread of air to its neighbor (Hacke et al., 2015). This structure effectively mitigates cavitation risk and explains why conifers often exhibit high cavitation resistance (P50 < -8 MPa in some species, Choat et al., 2012, Figure 1). Thereby, we can conclude that conifers generally better adapt to the dry conditions than angiosperms, especially under chronic drought. However, there are also other factors beyond the vessel-tracheid differences can impact the cavitation vulnerability including climate variables such as precipitation, temperature, and altitude (Choat et al. 2012; Maherali et al. 2004; Trueba et al., 2016). For example, as Figure 1 shows, while gymnosperms possess greater embolism resistance compared to angiosperms, the resistance of both species decreases with the increase in mean annual precipitation.

Figure 1 Embolism resistance and mean annual precipitation data for 384 angiosperm and 96 gymnosperm species (Source: Choat et al., 2012)

It is also important to consider the stomata control when talking about adaptation to dry conditions, because the balance between transpiration and photosynthesis is an essential compromise in the existence of plants: To make sugars, plants must absorb CO2 from the atmosphere through stomata; however, when stomata open, water is lost to the atmosphere at a prolific rate relative to the small amount of CO2 absorbed. Across plant species an average of 400 water molecules are lost for each CO2 molecule gained (Ruggiero et al., 2017). This highlights the importance of comparing the stomata control of angiosperms and conifers as this mechanism governs the balance between carbon assimilation (growth) and water conservation (survival). Conifers tend to maintain a conservative water-use strategy, characterized by low baseline stomatal conductance and limited stomatal responsiveness to environmental changes. This means that even under extreme drought, they avoid rapid stomatal fluctuations, allowing them to sustain slow but steady transpiration rates. In contrast, angiosperms exhibit more dynamic stomatal control, often responding to water availability more flexibly by closing their stomata rapidly to prevent water loss, but at the cost of reducing carbon assimilation (Klein & Ramon, 2019). This flexibility could give angiosperms an advantage in short-term droughts. However, if drought conditions persist, excessive stomatal closure in angiosperms could limit carbon uptake, leading to reduced growth or hydraulic failure. A special case to consider in the adaptation of stomatal control in dry condition is the CAM photosynthesis (Crassulacean Acid Metabolism). CAM plants open stomata only at night to take in CO₂, and keep them closed in the hot daytime. This adaptation allows gas exchange with minimal water loss, since nighttime temperatures are cooler and humidity is higher. While majority great majority of plants using CAM are angiosperms (flowering plants), it is also found in ferns, Gnetopsida and in quillworts (relatives of club mosses).

Finally, the distinct leaf morphology can reflect the differences of water conservation strategies between conifers and angiosperms. Conifers typically have needle-like leaves, which offer a low surface-area-to-volume ratio and limits water loss through transpiration. These leaves are often covered by thick, waxy cuticles, which create a waterproof barrier and further reduce evaporation from the leaf surface. Meanwhile, conifer stomata are often recessed in pits below the leaf surface. These sunken stomata trap moist air and reduce the gradient for water vapor loss (Šantrůček, 2022). By maintaining higher humidity around the stomata, conifers minimize transpiration. Angiosperms, by contrast, exhibit greater diversity in leaf morphology, employing three main strategies. Firstly, many angiosperms avoid drought stress by shedding some or all of their leaves during the driest seasons in response to water shortage, then enter a dormant state until rain returns (Gaff and Oliver, 2013). This strategy is common in seasonal dry tropical forests (e.g. teak trees lose leaves in the dry season) and desert shrubs like Ocotillo or Bursera. Secondly, some broadleaf evergreen angiosperms keep their leaves year-round but make them drought-resistant, known as Sclerophyllous leaves. These leaves tend to be small, thick, leathery, and often coated with wax or resin, which can effectively reduce water loss and reflect sunlight. Examples include many Mediterranean-climate evergreens, such as olive trees, holly, and Quercus ilex (holm oak) (Bussotti and Pollastrini, 2020). Thirdly, succulent angiosperm plants, including cacti, aloes, and many euphorbias, adapt to dry conditions by turning their leaves or stems into fleshy water-storage organs. Cacti are an extreme case – most cacti have no conventional leaves at all. Instead, their stems do the photosynthesis and are covered in a thick waxy skin. The spines of a cactus are actually modified leaves, evolved to provide shade and break up airflow across the cactus surface, thereby reducing evaporation. By eliminating broad leaves, cacti drastically cut water loss.

To summarize, I would argue that conifers generally withstand dry conditions better than angiosperms due to their cavitation-resistant tracheids, conservative stomatal regulation, and needle-like leaves that minimize water loss. However, certain angiosperms, including drought-deciduous trees, sclerophyllous species, and CAM plants, have evolved specialized strategies that allow them to thrive in specific arid environments, sometimes surpassing conifers in drought resilience. As climate change intensifies drought stress worldwide, understanding these differences in adaptation strategies is essential for predicting plant responses, conserving biodiversity, and guiding reforestation and agricultural practices.

Bibliography

Bussotti, F., & Pollastrini, M. (2020). Opportunities and threats of Mediterranean evergreen sclerophyllous woody species subjected to extreme drought events. Applied Sciences, 10(23), 8458.

Choat, B., Jansen, S., Brodribb, T. J., Cochard, H., Delzon, S., Bhaskar, R., … & Zanne, A. E. (2012). Global convergence in the vulnerability of forests to drought. Nature, 491(7426), 752-755.

Dai, A. (2013). Increasing drought under global warming in observations and models. Nature climate change, 3(1), 52-58.

Delzon, S., Douthe, C., Sala, A., & Cochard, H. (2010). Mechanism of water‐stress induced cavitation in conifers: bordered pit structure and function support the hypothesis of seal capillary‐seeding. Plant, Cell & Environment, 33(12), 2101-2111.

Dixon, H. H., & Joly, J. (1895). XII. On the ascent of sap. Philosophical Transactions of the Royal Society of London.(B.), (186), 563-576.

Hacke, U. G., Sperry, J. S., & Pittermann, J. (2005). Efficiency versus safety tradeoffs for water conduction in angiosperm vessels versus gymnosperm tracheids. In Vascular transport in plants (pp. 333-353). Academic Press.

Hacke, U. G., Lachenbruch, B., Pittermann, J., Mayr, S., Domec, J. C., & Schulte, P. J. (2015). The hydraulic architecture of conifers. Functional and ecological xylem anatomy, 39-75.

Gaff, D. F., & Oliver, M. (2013). The evolution of desiccation tolerance in angiosperm plants: a rare yet common phenomenon. Functional Plant Biology, 40(4), 315-328.

Klein, T., & Ramon, U. (2019). Stomatal sensitivity to CO2 diverges between angiosperm and gymnosperm tree species. Functional Ecology, 33(8), 1411-1424.

Maherali, H., Pockman, W. T., & Jackson, R. B. (2004). Adaptive variation in the vulnerability of woody plants to xylem cavitation. Ecology, 85(8), 2184-2199.

McElrone, A. J., Choat, B., Gambetta, G. A. & Brodersen, C. R. (2013) Water Uptake and Transport in Vascular Plants. Nature Education Knowledge 4(5):6

Ruggiero, A., Punzo, P., Landi, S., Costa, A., Van Oosten, M. J., & Grillo, S. (2017). Improving plant water use efficiency through molecular genetics. Horticulturae, 3(2), 31.

Šantrůček, J. (2022). The why and how of sunken stomata: does the behaviour of encrypted stomata and the leaf cuticle matter?. Annals of Botany, 130(3), 285-300.

Schulze, E. D., Beck, E., & Müller-Hohenstein, K. (2005). Plant ecology. Springer Science & Business Media.

Sperry, J. S., Hacke, U. G., & Pittermann, J. (2006). Size and function in conifer tracheids and angiosperm vessels. American journal of botany, 93(10), 1490-1500.

Sperry, J. S., Hacke, U. G., Feild, T. S., Sano, Y., & Sikkema, E. H. (2007). Hydraulic consequences of vessel evolution in angiosperms. International Journal of Plant Sciences, 168(8), 1127-1139.

The Editors of Encyclopaedia Britannica. “vessel”. Encyclopedia Britannica, 19 Sep. 2019, https://www.britannica.com/science/vessel. Accessed 28 February 2025.

Tyree, M. T. (1997). The cohesion-tension theory of sap ascent: current controversies. Journal of Experimental Botany, 48(10), 1753-1765.

Tyree, M. T., Zimmermann, M. H., Tyree, M. T., & Zimmermann, M. H. (2002). The cohesion-tension theory of sap ascent. Xylem structure and the ascent of sap, 49-88.

2. Ecosystem Level: Rainforest-savanna bistability

Rainforest-savanna bistability theory is proposed to explain the observation from satellite and field studies that across much of the humid-subhumid tropics (mean annual rainfall ≈ 1 000–2 000 mm), there is a bimodal distribution of tree cover: landscapes cluster around either ~20% (savanna) or >80 % (closed forest) while intermediate cover is rare. This pattern holds globally: Staver et al 2011 confirmed the pattern for sub-Saharan Africa using 250 m MODIS data, and Hirota et al (2011) observed similar patterns in South America and Australia.

These two principal equilibrium states of tropical vegetation are self-stabilized through coupled hydrological and fire feedbacks. In the rainforest state, dense tree cover enhances evapotranspiration, which recycles atmospheric moisture and reinforces local rainfall (Spracklen et al., 2012). In the central Amazon, models estimate that 25–50 % of precipitation originates from basin-internal recycling—roughly 1,000–2,000 mm yr⁻¹ of the ~2,000 mm annual total (Flores et al., 2024). Additionally, the closed canopy limits light penetration and wind at the ground level, suppressing the establishment of flammable grasses and reducing fire frequency (Hoffmann et al., 2012). For example, satellite-based analyses in the southern Amazon revealed that above a 40 % tree-cover threshold, fire occurrence drops sharply, and this threshold itself rises by ~10 % for every additional 500 mm yr⁻¹ of water deficit (Wuyts et al., 2017; Machida et al., 2021). In contrast, in savanna, sparse tree cover and an open canopy lower evapotranspiration and increase surface albedo, thereby reducing convective rainfall (Snyder et al., 2004). Meanwhile, the grassy understory promotes frequent fires, which prevent the recruitment of tree saplings into mature, fire-resistant individuals (Hoffmann et al., 2009). NASA MODIS fire counts show that fires burn 608 Mha every year, 86% in tropical savanna. Experimental burns across Brazilian savannas report that fire-return intervals of 1–4 years maintain tree cover below ~20 %—long enough for grass fuels to accumulate but short enough to prevent saplings escaping the “fire trap” (Oliveras et al., 2013).

Shift between states: This feedback-driven bistability means that the same mean annual precipitation—particularly within intermediate rainfall zones (~1000–2000 mm yr⁻¹)—can support either dense forest or open savanna, depending on the system’s initial condition and disturbance trajectory (Hirota et al., 2011; Staver et al., 2011). The accompanying diagram illustrates this concept using a system with tree cover fraction T=0.8 and precipitation P=2500 mm yr⁻¹ at point A. A modest (20%) reduction in precipitation alone (1) does not shift the system out of the forest basin of attraction; tree cover remains stable. However, if this reduction is coupled with a 30% tree loss due to deforestation (2), the system can cross the unstable intermediate equilibrium (3), triggering a transition toward the savanna attractor (4). Importantly, due to hysteresis, the reversal of these disturbances—such as restoring precipitation or halting deforestation—may not be sufficient to recover the original forest state. According to Hirota et al (2011) the precipitation threshold for savanna→forest conversion (∼2,200 mm yr⁻¹) exceeds that for forest→savanna (∼1,800 mm yr⁻¹). This yield a bistable band of ∼400 mm where both states can coexist.

Climate-vegetation models predict 14% to 47% of the Amazon forest to be exposed to compound stress in the coming decades (warming temperatures, annual rainfall, rainfall seasonality intensity, dry season length, and accumulated deforestation) (Rosa-Rodrigues et al., 2024). Since Amazon stores 150-200 Pg C, which equals 15-20 years of human carbon emissions, this is very important for the global carbon budget. CMIP6-driven dynamic vegetation models predict that under continued drying, by 2050, up to 6 % of the Amazon will shift from stable forest to bistability, primarily in the southern and central parts. 3 % will cross fully into a stable-savanna state, notably in the southern Bolivian Amazon. Conversely, 8 % (western Amazon) will become more stable forest, gaining resilience to disturbance. (Flores et al., 2024)

Debates: (1) Zwann et al (2024) also highlight factors beyond precipitation – models which only account for mean annual precipitation midclassifies 14% of the landscape type, but when adding consideration of factors such as precipitation seasonality, mean annual temperature, topographic roughness, elevation and soil sand fraction, the model was able to predict the landscape pattern with 99% accuracy. (2) Wuyts et al. (2017) used MODIS-VCF tree cover data and logistic regression to show that the bimodal feature of Amazon region is not just maintained by natural feedbacks alone but significantly enhanced by human land use, since human‐unaffected landscapes show weak bimodality, while areas near agriculture exhibit pronounced dual modes. This is likely explained by the edge effect related to clearing, logging, and agriculture. for example, Bauer et al (2024) found that aboveground biomass in forest edge areas is 27% lower than in forest core areas. (3) Zwann et al. (2024)’s study revealed that while the tree cover fraction (and thus vegetation distribution) is bimodal when looking from local scale (30m), when aggregating them into 0.1° “landscape” scale, 28% of the area is within the coexistence classification, thus challenging the classical bistability theory which the minor possibility for co-existence.