Root dynamics influence tree–grass coexistence in an Australian savanna

Both resource and disturbance controls have been invoked to explain tree persistence among grasses in savannas. Here we determine the extent to which competition for available resources restricts the rooting depth of both grasses and trees, and how this may influence nutrient cycling under an infrequently burned savanna near Darwin, Australia. We sampled fine roots <2 mm in diameter from 24 soil pits under perennial as well as annual grasses and three levels of canopy cover. The relative proportion of C₃ (trees) and C₄ (grasses) derived carbon in a sample was determined using mass balance calculations. Our results show that regardless of the type of grass both tree and grass roots are concentrated in the top 20 cm of the soil. While trees have greater root production and contribute more fine root biomass grass roots contribute a disproportional amount of nitrogen and carbon to the soil relative to total root biomass. We postulate that grasses maintain soil nutrient pools and provide biomass for regular fires that prevent forest trees from establishing while savanna trees, are important for increasing soil N content, cycling and mineralization rates. We put forward our ideas as a hypothesis of resource‐regulated tree–grass coexistence in tropical savannas.     

The distribution of tree and grass roots in savannas in relation to soil nitrogen and water

Here we describe the fine root distribution of trees and grasses relative to soil nitrogen and water profiles. The primary objective is to improve our understanding of edaphic processes influencing the relative abundance of trees and grasses in savanna systems. We do this at both a mesic (737 mm MAP) site on sandy-loam soils and at an arid (547 mm MAP) site on clay rich soils in the Kruger National Park in South Africa. The proportion of tree and grass fine roots at each soil depth were estimated using the δ¹³C values of fine roots and the δ¹³C end members of the fine roots of the dominant trees and grasses at our study sites. Changes in soil nitrogen concentrations with depth were indexed using total soil nitrogen concentrations and soil δ¹⁵N values. Soil water content was measured at different depths using capacitance probes. We show that most tree and grass roots are located in the upper layers of the soil and that both tree and grass roots are present at the bottom of the profile. We demonstrate that root density is positively related to the distribution of soil nitrogen and negatively related to soil moisture. We attribute the negative correlation with soil moisture to evaporation from the soil surface and uptake by roots. Our data is a snapshot of a dynamic process, here the picture it provides is potentially misleading. To understand whether roots in this system are primarily foraging for water or for nitrogen future studies need to include a dynamic component.     

Small differences in root distributions allow resource niche partitioning

1. Deep roots have long been thought to allow trees to coexist with shallow‐rooted grasses. However, data demonstrating how root distributions affect water uptake and niche partitioning are uncommon.
2. We describe tree and grass root distributions using a depth‐specific tracer experiment six times over two years in a subtropical savanna, Kruger National Park, South Africa. These point‐in‐time measurements were then used in a soil water flow model to simulate continuous water uptake by depth and plant growth form (trees and grasses) across two growing seasons. This allowed estimates of the total amount of water a root distribution could absorb as well as the amount of water a root distribution could absorb in excess of the other rooting distribution (i.e., unique hydrological niche).
3. Most active tree and grass roots were in shallow soils: The mean depth of water uptake was 22 cm for trees and 17 cm for grasses. Slightly deeper rooting distributions provided trees with 5% more soil water than the grasses in a drier season, but 13% less water in a wetter season. Small differences also provided each rooting distribution (tree or grass) with unique hydrological niches of 4 to 13 mm water.
4. The effect of rooting distributions has long been inferred. By quantifying the depth and timing of water uptake, we demonstrated how even small differences in rooting distributions can provide plants with resource niches that can contribute to species coexistence. Differences in total water uptake and unique hydrological niche sizes were small in this system, but they indicated that tradeoffs in rooting strategies can be expected to contribute to tree and grass coexistence because 1) competitive advantages change over time and 2) plant growth forms always have access to a soil resource pool that is not available to the other plant growth form.

     

Asynchronicity in root and shoot phenology in grasses and woody plants

Phenology is central to understanding vegetation response to climate change, as well as vegetation effects on plant resources, but most temporal production data is based on shoots, especially those of trees. In contrast, most production in temperate and colder regions is belowground, and is frequently dominated by grasses. We report root and shoot phenology in 7‐year old monocultures of 10 dominant species (five woody species, five grasses) in southern Canada. Woody shoot production was greatest about 8 weeks before the peak of root production, whereas grass shoot maxima preceded root maxima by 2–4 weeks. Over the growing season, woody root, and grass root and shoot production increased significantly with soil temperature. In contrast, the timing of woody shoot production was not related to soil temperature (r=0.01). The duration of root production was significantly greater than that of shoot production (grasses: 22%, woody species: 54%). Woody species produced cooler and moister soils than grasses, but growth forms did not affect seasonal patterns of soil conditions. Although woody shoots are the current benchmark for phenology studies, the other three components examined here (woody plant roots, grass shoots and roots) differed greatly in peak production time, as well as production duration. These results highlight that shoot and root phenology is not coincident, and further, that major plant growth forms differ in their timing of above‐ and belowground production. Thus, considering total plant phenology instead of only tree shoot phenology should provide a better understanding of ecosystem response to climate change.     

Root niche partitioning among grasses, saplings, and trees measured using a tracer technique

Niche partitioning of resources by plants is believed to be a fundamental aspect of plant coexistence and biogeochemical cycles; however, measurements of the timing and location of resource use are often lacking because of the difficulties of belowground research. To measure niche partitioning of soil water by grasses, planted saplings, and trees in a mesic savanna (Kruger National Park, South Africa), we injected deuterium oxide into 102,000 points in 15, 154-m² plots randomly assigned to one of five depths (0–120 cm) and one of three time periods during the 2008/2009 growing season. Grasses, saplings and trees all demonstrated an exponential decline in water uptake early in the season when resources were abundant. Later in the season, when resources were scarce, grasses continued to extract the most water from the shallowest soil depths (5 cm), but saplings and trees shifted water uptake to deeper depths (30–60 cm). Saplings, in particular, rapidly established roots to at least 1 m and used these deep roots to a greater extent than grasses or trees. Helping to resolve contradictory observations of the relative importance of deep and shallow roots, our results showed that grasses, saplings and trees all extract the most water from shallow soils when it is available but that woody plants can rapidly shift water uptake to deeper soils when resources are scarce. Results highlight the importance of temporal changes in water uptake and the problems with inferring spatial and temporal partitioning of soil water uptake from root biomass measurements alone.     

Interspecific effects on plant size inequality: evidence from a temperate savanna community

The influences of intraspecific competition on plant size inequality have been well documented, but interspecific effects on this topic remain little understood. Here we examined the effects of canopy shading and fine roots of the trees (Elaeagnus angustifolia) on size inequality of the grasses (Achnatherum splendens) in a temperate savanna community in northwest China. Three study plots of 400 m² were divided into 4-m² quadrats, within each of which (1) canopy shading was quantified by modeling cumulative direct solar radiation (CDSR) and (2) the root effect was quantified using an empirical relationship between tree fine root density (TFRD) and relative distance to tree bases. Morphological traits were measured to represent grass size. Redundancy analysis (RDA) was conducted to examine the relative influences of grass density, CDSR and TFRD on the coefficient of variation of grass size. Results showed that no significant correlation occurred between grass density and grass size inequality. Both CDSR and TFRD had significant negative correlations with grass size inequality, suggesting that canopy shading and the presence of fine roots of trees can, respectively, increase and reduce grass size inequality. Canopy shading and TFRD played competitive roles in determining grass size inequality, where the root effect was a stronger factor than canopy shading. The tree effects can substantially alter the intensity of water stress. In response, size inequality of the grasses could be influenced through size-specific growth/mortality and slowed size divergence. These mechanisms could operate together in the savanna community.     

Seasonal leaf dynamics across a tree density gradient in a Brazilian savanna

Interactions between trees and grasses that influence leaf area index (LAI) have important consequences for savanna ecosystem processes through their controls on water, carbon, and energy fluxes as well as fire regimes. We measured LAI, of the groundlayer (herbaceous and woody plants <1-m tall) and shrub and tree layer (woody plants >1-m tall), in the Brazilian cerrado over a range of tree densities from open shrub savanna to closed woodland through the annual cycle. During the dry season, soil water potential was strongly and positively correlated with grass LAI, and less strongly with tree and shrub LAI. By the end of the dry season, LAI of grasses, groundlayer dicots and trees declined to 28, 60, and 68% of mean wet-season values, respectively. We compared the data to remotely sensed vegetation indices, finding that field measurements were more strongly correlated to the enhanced vegetation index (EVI, r ²=0.71) than to the normalized difference vegetation index (NDVI, r ²=0.49). Although the latter has been more widely used in quantifying leaf dynamics of tropical savannas, EVI appears better suited for this purpose. Our ground-based measurements demonstrate that groundlayer LAI declines with increasing tree density across sites, with savanna grasses being excluded at a tree LAI of approximately 3.3. LAI averaged 4.2 in nearby gallery (riparian) forest, so savanna grasses were absent, thereby greatly reducing fire risk and permitting survival of fire-sensitive forest tree species. Although edaphic conditions may partly explain the larger tree LAI of forests, relative to savanna, biological differences between savanna and forest tree species play an important role. Overall, forest tree species had 48% greater LAI than congeneric savanna trees under similar growing conditions. Savanna and forest species play distinct roles in the structure and dynamics of savanna–forest boundaries, contributing to the differences in fire regimes, microclimate, and nutrient cycling between savanna and forest ecosystems.     

Fine Root Biomass and Its Relationship to Evapotranspiration in Woody and Grassy Vegetation Covers for Ecological Restoration of Waste Storage and Mining Landscapes

Production and distribution of fine roots (≤2.0 mm diameter) are central to belowground ecological processes. This is especially true where vegetation serves as a pump to prevent saturation of soil and possible drainage of excess water into or from potentially toxic waste material stored underground or in mounds aboveground. In this study undertaken near Sydney in Australia, we determined fine root biomass and evapotranspiration (ET) on a waste disposal site restored with either a 15-year-old grass sward or plantations of mixed woody species that were either 5 years old (plantation-5) with a vigorous groundcover of pasture legumes and grasses, or 3 years old (plantation-3) with sparse groundcover. These sites were compared with nearby remnant woodland; all four were located within 0.5-km radius at the same site. Ranking of fine root biomass was in the order woodland (12.3 Mg ha⁻¹) > plantation-5 (8.3 Mg ha⁻¹) > grass (4.9 Mg ha⁻¹) > plantation-3 (1.2 Mg ha⁻¹) and was not correlated with nutrient contents in soil or plants, but reflected the form and age of the vegetation covers. Trends in root length density (RLD) and root area index (RAI) followed those in root biomass, but the differences in RAI were larger than those in biomass amongst the vegetation covers. Annual ET in the dry year of 2009 was similar in the three woody vegetation covers (652–683 mm) and was at least 15% larger than for the grass (555 mm), which experienced restrained growth in winter and periodic mowing. This resulted in drainage from the grass cover while there was no drainage from any of the woody vegetation covers. In plantation-5, root biomass, RAI and RLD were reduced in the rain shadow side of the tree rows. Similarly, the amount and depth of rooting in the groundcover were reduced close to the trees compared to midway between rows. Differences in the root variables were larger than those in ET, which suggested that more roots were produced than were needed for water uptake and/or presence of considerable amounts of necromass. We conclude that vegetation covers, such as plantation-5 consisting of widely spaced trees and a heavy groundcover containing winter-active pasture legumes, will promote year-round water-use with a reduced risk of deep rooting that could breach buried wastes. This function could be sustained through progressive thinning of trees to account for not more than 25% of the whole canopy cover; this will minimize competition for limited soil-water and thereby constrain deep rooting as vegetation ages and attains climax.     

Fine Root Distribution in Dehesas of Central-Western Spain

A Dehesa is a structurally complex agro-silvo-pastoral system where at least two strata of vegetation, trees and herbaceous plants coexist. We studied the root distribution of trees (Quercus ilex L.) and herbaceous plants, in order to evaluate tree and crops competition and complementarity in Dehesas of Central Western Spain. 72 soil cores of 10 cm diameter (one to two metre deep) were taken out around 13 trees. Seven trees were intercropped with Avena sativa L. and six trees were in a grazed pasture dominated by native grasses. Soil coring was performed at four distances from the tree trunks, from 2.5 (beneath canopy) till 20 m (out of the canopy). Root length density (RLD) of herbaceous plants and trees was measured using the soil core-break method. Additionally, we mapped tree roots in 51 profiles of 7 recently opened road cuts, located between 4 and 26 m of distance from the nearest tree. The depth of the road cuts varied between 2.5 and 5.5 m. Herbaceous plant roots were located mostly in the upper 30 cm, above a clayey, dense soil layer. RLD of herbaceous plants decreased exponentially with depth until 100 cm depth. Holm-oak showed a much lower RLD than herbs (on average, 2.4 versus 23.7 km m⁻³, respectively, in the first 10 cm of the soil depth). Tree RLD was surprisingly almost uniform with depth and distance to trees. We estimated a 5.2 m maximum depth and a 33 m maximum horizontal extension for tree roots. The huge surface of soil explored by tree roots (even 7 times the projection of the canopy) could allow trees to meet their water needs during the dry Mediterranean summers. The limited vertical overlap of the two root profiles suggests that competition for soil resources between trees and the herbaceous understorey in the Dehesa is probably not as strong as usually assumed.     

Soil organic carbon content at a range of north Australian tropical savannas with contrasting site histories

Soils play an important role in the global carbon cycle, and can be major source or sink of CO₂ depending upon land use, vegetation type and soil management practices. Natural and human impact on soil carbon concentration and storage is poorly understood in native north Australian savanna, yet this represents the largest carbon store in the ecosystem. To gain understanding of possible management impacts on this carbon pool, soil organic carbon (SOC) of the top 1m of red earth sands and sandy loams common in the region was sampled at 5 sites with different vegetation cover and site history (fire regime and tree removal). SOC was high when compared to other published values for savannas and was more comparable with dry-deciduous tropical forests. Sites sampled in this study represent high rainfall savannas of northern Australia (> 1700 mm annual rainfall) that feature frequent burning (2 in 3 years or more frequent) and a cycle of annual re-growth of tall C₄ grasses that dominate the savanna understorey. These factors may be responsible for the higher than expected SOC levels of the surface soils, despite high respiration rates. Medium term fire exclusion (15–20 years) at one of the sampled sites (Wildlife Park) dramatically reduced the grassy biomass of the understorey. This site had lower SOC levels when compared to the grass dominated and frequently burnt sites, which may be due to a reduction in detrital input to surface (0–30 cm) soil carbon pools. Exclusion of trees also had a significant impact on both the total amount and distribution of soil organic carbon, with tree removal reducing observed SOC at depth (100 cm). Soil carbon content was higher in the wet season than that in the dry season, but this difference was not statistically significant. Our results indicated that annual cycle of grass growth and wildfire resulted in small carbon accumulation in the upper region of the soil, and removal of woody plants resulted in significant carbon losses to recalcitrant, deep soil horizons greater than 80 cm depth.     

Spatial partitioning of the soil water resource between grass and shrub components in a West African humid savanna

Most savanna water balance models assume water partitioning between grasses and shrubs in a two-layer hypothesis, but this hypothesis has not been tested for humid savanna environments. Spatial partitioning of soil water between grasses and shrubs was investigated in a West African humid savanna by comparing the isotopic composition (oxygen-18 and deuterium) of soil water and plant stem water during rainy and dry conditions. Both grass and shrub species acquire most of their water from the top soil layer during both rainy and dry periods. A shift of water uptake pattern towards deeper horizons was observed only at the end of the dry season after shrub defoliation. The mean depth of water uptake, as determined by the isotopic signature of stem water, was consistent with grass and shrub root profiles and with changes in soil water content profiles as surveyed by a neutron probe. This provides evidence for potentially strong competition between shrubs and grasses for soil water in these humid savannas. Limited nutrient availability may explain these competitive interactions. These results enhance our understanding of shrub-grass interactions, and will contribute to models of ecosystem functioning in humid savannas.     

Length densities,occupancies and weights of apple root systems

The root systems of apple trees from five orchards ranging in age from 1.5-y to 14-y were sampled to depths of between one and two metres using soil cores. Although trees came from orchards which differed in soil-type, tree spacings and management, consistent patterns were found in root systems. In orchards of 4-y and older, roots of adjacent trees met so that soil volumes within the planting grids (i.e, tree spacings of approximately 5 m inter-row×4 m intra-row distances) were completely explored, although not completely occupied by roots. Mean root-length densities declined with depth for these orchards. In the 1.5-y orchard, roots from adjacent trees did not meet and root-length densities declined with radial distance from the stem as well as with depth.Root-length densities in the top 1 m ranged from zero to about 1.0 cm.cm^–3 in all orchards and were highly variable. The proportions of core samples having zero values for root-length density were used to subdivide the root zone into volumes in which all samples contained roots, and volumes in which some samples had no roots.Results suggest that roots in an average tree penetrate to at least one metre depth in all but very young orchards so that soil in this volume is fully explored. Volumes filled by roots and volumes occupied at any particular root-length density appear to reach a maximum at about 4 years. Volumes of soil occupied at any particular root-length density were equal in all orchards older than 4 years. This suggests that root growth was balanced by root death. In contrast woody roots continue to accumulate with time.     

Decomposition and nutrient release of grass and tree fine roots along an environmental gradient in southern Patagonia

Decomposition of fine roots is a fundamental ecosystem process that relates to carbon (C) and nutrient cycling in terrestrial ecosystems. However, this important ecosystem process has been hardly studied in Patagonian ecosystems. The aim of this work was to study root decomposition and nutrient release from fine roots of grasses and trees (Nothofagus antarctica) across a range of Patagonian ecosystems that included steppe, primary forest and silvopastoral forests. After 2.2 years of decomposition in the field all roots retained 70–90% of their original mass, and decomposition rates were 0.09 and 0.15 year^?1 for grass roots in steppe and primary forest, respectively. For N. antarctica roots, no significant differences were found in rates of decay between primary and silvopastoral forests (k = 0.07 year^?1). Possibly low temperatures of these southern sites restricted decomposition by microorganisms. Nutrient release differed between sites and root types. Across all ecosystem categories, nitrogen (N) retention in decomposing biomass followed the order: tree roots > roots of forest grasses > roots of steppe grasses. Phosphorus (P) was retained in grass roots in forest plots but was released during decomposition of tree and steppe grass roots. Calcium (Ca) dynamics also was different between root types, since trees showed retention during the initial phase, whereas grass roots showed a slow and consistent Ca release during decomposition. Potassium (K) was the only nutrient that was rapidly released from both grass and tree roots in both grasslands and woodlands. We found that silvopastoral use of N. antarctica forests does not affect grass or tree root decomposition and/or nutrient release, since no significant differences were found for any nutrient according to ecosystem type. Information about tree and grass root decomposition found in this work could be useful to understand C and nutrient cycling in these southern ecosystems, which are characterized by extreme climatic conditions.     

Biomass and distribution of fine and coarse roots from blue oak (Quercus douglasii) trees in the northern Sierra Nevada foothills of California

This research adds to the limited data on coarse and fine root biomass for blue oak (Quercus douglasii Hook and Arn.), a California deciduous oak species found extensively throughout the interior foothills surrounding the Central Valley. Root systems of six blue oak trees were analyzed using three methods — backhoe excavation, quantitative pits, and soil cores. Coarse root biomass ranged from 7 to 177 kg per tree. Rooting depth for the main root system ranged from 0.5 to 1.5 m, with an average of 70% of excavated root biomass located above 0.5 m. Of the total biomass in excavated central root systems, primary roots (including burls) accounted for 56% and large lateral roots (> 20 mm diameter) accounted for 36%. Data from cores indicated that most biomass outside of the root crown was located in fine roots and that fine root biomass decreased with depth. At surface depths (0–20 cm), small-fine (< 0.5 mm diameter) roots accounted for 71%, large-fine (0.5–2.0 mm) for 25%, and coarse (> 2 mm) for 4% of total root biomass collected with cores. Mean fine root biomass density in the top 50 cm was 0.43 kg m⁻³. Fine root biomass did not change with increasing distance from the trees (up to approximately 5 m). Thus, fine roots were not concentrated under the tree canopies. Our results emphasize the importance of the smallest size class of roots (<0.5 mm), which had both higher N concentration and, in the area outside the central root system, greater biomass than large fine (0.5–2.0 mm) or coarse (> 2.0 mm) roots. This revised version was published online in June 2006 with corrections to the Cover Date.     

Root distribution of poplar at varying densities on pastoral hill country

Spaced poplar (Populus spp.) trees are used widely in New Zealand for soil conservation on erodible pastoral hill country. Their root distribution in this environment, and factors that affect it, are poorly understood. Robust recommendations on effective tree spacing depend on knowledge of root systems. This study determined the effect of tree density, position between trees, and soil depth (0–90 cm) on root number, root diameter distribution, root area ratio (RAR), and cross sectional area per root for young trees on slopes. Data were collected for lateral roots using trenches. Greater than 80% of roots were < 5 mm diameter and root attributes were highest in shallow soil. Trees at 770 stems per hectare (sph) had 3–12 times more roots and 3–9 times greater RAR than those at densities of ≤ 237 sph, representative of most tree-pasture systems. Mean cross sectional area per root was similar across densities. Positions close to trees had twice as many roots (46 vs. 23/m²) and RAR (109 vs. 52 mm²/m²) as positions midway between trees. The study provided quantitative understanding of variation in root distribution with tree density and information useful for supporting and strengthening recommendations on densities for effective erosion control.     

Root vascular anatomy predicts maximum growth rates in savanna trees and grasses

Root-based functional traits are relatively overlooked as drivers of savanna plant community dynamics, an important gap in water-limited ecosystems. Recent work has shed light on patterns of trait coordination in roots, but less is known about the relationship between root functional traits, water acquisition, and plant demographic rates. Here, we investigated how fine-root vascular and morphological traits are related in two dominant PFTs (C₃ trees and C₄ grasses from the savanna biome), whether root traits can predict plant relative growth rate (RGR), and whether root trait multivariate relationships differ in trees and grasses. We used root data from 21 tree and 18 grass species grown under greenhouse conditions, and quantified a suite of vascular and morphological root traits. We used a principal components analysis (PCA) to identify common axes of trait variation, compared trait correlation matrices between the two PFTs, and investigated the relationship between PCA axes and individual traits and RGR. We found that there was no clear single axis integrating vascular and morphological traits, but found that vascular anatomy predicted RGR in both trees and grasses. Trait correlation matrices differed in trees and grasses, suggesting potentially divergent patterns of trait coordination between the two functional types. Our results suggested that, despite differences in trait relationships between trees and grasses, root conductivity may constrain maximum growth rate in both PFTs, highlighting the critical role that water relations play in savanna vegetation dynamics and suggesting that root water transport capacity is an important predictor of plant performance in the savanna biome.     

Differential water resource use by herbaceous and woody plant life-forms in a shortgrass steppe community

We conducted a study to test the predictions of Walter's two-layer model in the shortgrass steppe of northeastern Colorado. The model suggests that grasses and woody plants use water resources from different layers of the soil profile. Four plant removal treatments were applied in the spring of 1996 within a plant community codominated by Atriplex canescens (a C₄ shrub) and Bouteloua gracilis (a C₄ grass). During the subsequent growing season, soil water content was monitored to a depth of 180 cm. In addition, stem and leaf tissue of Atriplex, Bouteloua and the streamside tree Populus sargentii were collected monthly during the growing seasons of 1995 and 1996 for analysis of the δ¹⁸O value of plant stem water (for comparison with potential water sources) and the δ¹³C value of leaves (as an indicator of plant water status). Selective removal of shrubs did not significantly increase water storage at any depth in the measured soil profile. Selective removal of the herbaceous understory (mainly grasses) increased water storage in the top 60 cm of the soil. Some of this water gradually percolated to lower layers, where it was utilized by the shrubs. Based on stem water δ¹⁸O values, grasses were exclusively using spring and summer rain extracted from the uppermost soil layers. In contrast, trees were exclusively using groundwater, and the consistent δ¹³C values of tree leaves over the course of the summer indicated no seasonal changes in gas exchange and therefore minimal water stress in this life-form. Based on anecdotal rooting-depth information and initial measurements of stem water δ¹⁸O, shrubs may have also had access to groundwater. However, their overall δ¹⁸O values indicated that they mainly used water from spring and summer precipitation events, extracted from subsurface soil layers. These findings indicate that the diversity of life-forms found in this shortgrass steppe community may be a function of the spatial partitioning of soil water resources, and their differential use by grasses, shrubs, and trees. Consequently, our findings support the two-layer model in a broad sense, but indicate a relatively flexible strategy of water acquisition by shrubs. Received: 23 December 1997 / Accepted: 16 September 1998     

Oxidative stress in relation to reproduction,contaminants, gender and age in a long-lived seabird

A key question in savanna ecology is how trees and grasses coexist under N limitation. We used N stable isotopes and N content to study N source partitioning across seasons from trees and associated grasses in a semi-arid savanna. We also used ¹⁵N tracer additions to investigate possible redistribution of N by trees to grasses. Foliar stable N isotope ratio (δ¹⁵N) values were consistent with trees and grasses using mycorrhiza-supplied N in all seasons except in the wet season when they switched to microbially fixed N. The dependence of trees and grasses on mineralized soil N seemed highly unlikely based on seasonal variation in mineralization rates in the Kruger Park region. Remarkably, foliar δ¹⁵N values were similar for all three tree species differing in the potential for N fixation through nodulation. The tracer experiment showed that N was redistributed by trees to understory grasses in all seasons. Our results suggest that the redistribution of N from trees to grasses and uptake of N was independent of water redistribution. Although there is overlap of N sources between trees and grasses, dependence on biological sources of N coupled with redistribution of subsoil N by trees may contribute to the coexistence of trees and grasses in semi-arid savannas.     

Below-ground competition between trees and grasses may overwhelm the facilitative effects of hydraulic lift

Under large East African Acacia trees, which were known to show hydraulic lift, we experimentally tested whether tree roots facilitate grass production or compete with grasses for below‐ground resources. Prevention of tree–grass interactions through root trenching led to increased soil water content indicating that trees took up more water from the topsoil than they exuded via hydraulic lift. Biomass was higher in trenched plots compared to controls probably because of reduced competition for water. Stable isotope analyses of plant and source water showed that grasses which competed with trees used a greater proportion of deep water compared with grasses in trenched plots. Grasses therefore used hydraulically lifted water provided by trees, or took up deep soil water directly by growing deeper roots when competition with trees occurred. We conclude that any facilitative effect of hydraulic lift for neighbouring species may easily be overwhelmed by water competition in (semi‐) arid regions.