Much competition between plants occurs belowground1,2. Studying the differentiation of root systems has been thought of as a mechanism that reduces competition and facilitates species coexistence, and can be used to understand the spatial distribution of plants, plant coexistence, and soil and water budgets3,4,5. Many studies show that niche differentiation can occur if plant species occupy different environmental niche axes, including soil moisture and root depth6,7,8.
Competition for water is considered to occur by reduction in availability, favoring plants that can withstand the lowest water potential9. However, competition for water is less studied than other factors such as nutrients or light9. Water availability is a particularly important factor in the survival and establishment of very young plants10. The impact of soil-moisture depletion on plants is expressed in lower emergence rate, diminished plant size, and decreased biomass11. Moreover, plants growing in water-limited soils are more prone to hydraulic failure and drought-induced mortality. Drought leads to higher xylem tensions, as the tension of water within the xylem exceeds atmospheric pressure, leading to xylem embolism or cavitation. This embolism (gas bubble) blocks water flow through xylem conduits and reduce the delivery of water from the soil to other parts of the plant12,13. Water limitation can also affect a plant’s nutrient uptake, alter plant-microbe interactions, and affect transpiration rates. This will result in a decline in available resources for the different parts of the plant, including the leaves and seeds14. The presence of trees in a system can alter soil water content, affecting all associated plants.
In one of the most prominent theories of niche separation between trees and grasses, Walter (1939) proposed the two-layer hypothesis as an equilibrium explanation for the coexistence of savanna trees and grasses. Trees and grasses have very distinct root morphologies and different water-use strategies. Root-morphology differences can affect the efficiency of trees and grasses to extract water, independently of rooting depth. Walter (1939) proposed that grasses predominate over trees in savannas because grasses have shallow roots, allowing them to use water efficiently and take advantage of the available rainfall. Grasses are considered superior competitors for water in the upper soil profile15 due to their shallow rooting system with high root length and surface area. Contrastingly, trees have access to deeper water because of their extensive rooting systems4,16. However, in grasslands, trees and grasses may use overlapping areas in the soil profile, especially while trees are growing their roots through the soil profile6,17. Daly et al. (2000)18 studied tree-grass interactions in water-limited ecosystems and established that in the absence of fire, trees eventually become the dominant life form because they are more deeply rooted than grasses. Conversely, grass dominance is expected if trees and grasses occupy the same rooting region because of their greater water-use efficiency4. One aspect that Walter’s (1939) two-layer hypothesis did not focus on is how the roots of two or more tree species interact to allow these trees to coexist8. A few studies comparing the depth of water uptake by co-occurring woody species reported some tree species use only deep soil water, while others take water from both shallow and deep layers4,8,19.
Eastern redcedar (Juniperus virginiana; hereafter ERC) is the most widespread conifer in the eastern United States20. Due to drought and fire exclusion in the past century, ERC is encroaching into new habitats in eastern and Central US, including grasslands in the Great Plains (grassland located in the interior of North America)21,22, and midwestern prairies as far west as Nebraska23,24. ERC are known to be drought-tolerant25 and have the ability to photosynthesize in low soil-moisture conditions26,27. In addition, ERC has a relatively high leaf-level water-use efficiency under water limitation27, giving this tree a competitive advantage over other species in water-limited environments19,22.
An important factor that contributes to the improved growing conditions of ERC is reduced plant competition with neighboring plants28, such as grasses (Ward, 2020). A major grass competitor is Bromus inermis (hereafter smooth brome). Smooth brome is an invasive perennial grass that dominates many grasslands and old fields in every state of the contiguous United States30. An additional competitor for ERC are Quercus spp. ERC is encroaching in areas where oak (Quercus) trees, such as Quercus stellata (hereafter post oak), are dominant22,25. For example, ERC encroached into the Cross Timbers (Western edge of the eastern deciduous forest of the U.S.), transforming post oak-dominated forests to ERC and post oak co-dominant woodlands22.
In order to accurately model and forecast species dynamics and the impact of climate change on ecosystem processes, identifying the source of water uptake of species and their role in the water balance of an ecosystem is important31. In this study we sought to determine the rooting depth and water uptake across soil layers for ERC both growing alone and in competition with smooth brome and/or post oak. We conducted a greenhouse experiment and used frequency domain reflectometry (FDR) to measure water availability at different depths. We also accounted for ERC root length and water status by examining midday leaf water potential and relative water content. We further tested Walter’s (1939) two-layer hypothesis by focusing on how the roots of two or more species interact to allow trees to coexist.
We made three predictions based on Walter’s (1939) two-layer hypothesis:
1) When ERC and smooth brome co-occur, smooth brome will take up water from the topsoil layers, because of their shallow roots, leaving less water for ERC. As a result, ERC, having deeper roots than smooth brome, will invest in belowground development and will further extend their roots to get to the deeper water in the soil profile.
2) When ERC and post oak are competing, we predict that there will be niche separation between ERC and post oak roots22. ERC will avoid low water potentials by developing a deep root system that gives them access to deeper soil moisture.
3) When all three species (ERC, post oak, smooth brome) co-occur, soil moisture will be consistently low throughout the soil profile. Moreover, ERC will be more stressed (lower water potential and relative water content) when all three species co-occur compared to the other treatments (ERC-alone, ERC + post oak, and ERC + smooth brome). As a result, ERC sapling performance and encroachment will be negatively affected when the three species co-occur together due to lower water availability for ERC.
4) We expect that the water requirements for the ERC, post oak, and smooth brome will increase with time (higher in the second growing season) as the plants grow. As a result, there will be lower soil moisture content in second growing season for all ERC treatments.