Legume and Non-legume Trees Increase Soil Carbon Sequestration in Savanna

Savanna ecosystems are increasingly pressured by climate and land-use changes, especially around populous areas such as the Mt. Kilimanjaro region. Savanna vegetation consists of grassland with isolated trees or tree groups and is therefore characterized by high spatial variation and patchiness of canopy cover and aboveground biomass. Both are major regulators for soil ecological properties and soil-atmospheric trace gas exchange (CO₂, N₂O, CH₄), especially in water-limited environments. Our objectives were to determine spatial trends in soil properties and trace gas fluxes during the dry season and to relate above- and belowground processes and attributes. We selected a Savanna plain with vertic soil properties, south east of Mt. Kilimanjaro. Three trees were chosen from each of the two most dominant species: the legume Acacia nilotica and the non-legume Balanites aegyptiaca. For each tree, we selected one transect with nine sampling points, up to a distance of 4 times the crown radius from the stem. At each sampling point, we measured carbon (C) and nitrogen (N) content, δ¹³C of soil (0–10, 10–30 cm depth) and in plant biomass, soil C and N pools, water content, available nutrients, cation exchange capacity (CEC), temperature, pH, as well as root biomass and greenhouse-gas exchange. Tree species had no effect on soil parameters and gas fluxes under the crown. CEC, C, and N pools decreased up to 50% outside the crown-covered area. Tree leaf litter had a far lower C:N ratio than litter of the C₄ grasses. δ¹³C in soil under the crown shifted about 15% in the direction of tree leaf litter δ¹³C compared to soil in open area reflecting the tree litter contribution to soil organic matter. The microbial C:N ratio and CO₂ efflux were about 30% higher in the open area and strongly dependent on mineral N availability. This indicates N limitation and low microbial C use efficiency in the soil of open grassland areas. We conclude that the spatial structure of aboveground biomass in savanna ecosystems leads to a spatial redistribution of nutrients and thus C mineralization and sequestration. Therefore, the capability of savanna ecosystems to act as C sinks is both directly and indirectly dependent on the abundance of trees, regardless of their N-fixing status.     

Contribution of paddock trees to the conservation of terrestrial invertebrate biodiversity within grazed native pastures

Abstract: Paddock trees are a common feature in the agricultural landscapes of Australia. Recent studies have demonstrated the value of scattered paddock trees for soil fertility, native pasture plants and arboreal faunas; however, the degree to which scattered paddock trees contribute to the conservation of terrestrial invertebrate biodiversity within grazed landscapes remains unknown. We ask three questions: (i) Is there a difference between the terrestrial invertebrate assemblages found under paddock trees compared with surrounding grazed native pastures? (ii) Can gradients in soil and litter variables from the base of trees explain patterns in invertebrate assemblages? and (iii) Does the presence of scattered paddock trees have implications for the conservation of terrestrial invertebrate biodiversity within grazed native pastures? We used pitfall trapping and extraction from soil cores to sample the invertebrate assemblages under six New England Peppermint trees (Eucalyptus nova‐anglica Deane and Maiden) and compared them with assemblages sampled from the open paddock. Formicidae and Collembola univariate and multivariate data were analysed along with a range of soil and litter variables. We found (i) significant differences in the assemblages of invertebrates under trees compared with surrounding grazed pastures; (ii) that most soil and litter variables revealed gradients away from tree bases and these variables explained significant variation in invertebrate assemblages; and (iii) more native invertebrates and more species of invertebrates were found under trees compared with the surrounding pastures. We discuss the relationships between paddock trees, the ground and soil environments and the invertebrate communities that inhabit these environments, and conclude with a discussion of the future for paddock trees and the biota supported by them.     

Clumped and isolated trees influence soil nutrient levels in an Australian temperate box woodland

Isolated paddock trees are a common feature of agri-pastoral landscapes in south-eastern Australia. We assessed the impact of trees on soil nutrients by examining (1) changes in soil nutrients under clumped and isolated (living and dead) trees at four microsites corresponding with increasing distance from the trunk (trunk, mid-canopy, drip line, open), and (2) changes with depth under trees growing in clearly-defined clumps. We detected significantly greater concentrations of organic C, and total N and S under trees growing in clumps compared with either isolated living or dead trees. Levels of soluble Ca²⁺, K⁺ and Mg²⁺, pH, electrical conductivity (EC) and available P declined with increasing distance from the trees, but there were no significant trends for organic C, or total N and S. The concentration of most nutrients declined with depth, particularly at microsites close to the trunk, while pH increased with depth. We believe that differences in chemistry were largely driven by greater inputs of organic matter under the trees. This study reinforces the view that trees, whether scattered or in clumps, are important for soil nutrient conservation in agri-pastoral landscapes.     

Spatial and temporal litterfall heterogeneity generated by woody species in the Central Monte desert

In arid and semiarid environments, the presence of woody species generates a series of environmental gradients that increase spatial heterogeneity and modify the pattern of distribution of the other species. We postulate that the temporal and spatial variability in litter input generated by woody species is a relevant factor in the generation of edaphic heterogeneity by redistribution of nutrients and the physical effects of litter. The objective of this study was to determine the temporal and spatial variability in the amount of litter input under the canopy of dominant woody plants (Prosopis flexuosa and Larrea divaricata) and in exposed areas at the Ñacuñán Reserve, in the central zone of the Monte desert. Litterfall was collected during 2 years from 30-cm-diameter litter traps distributed at three microsites: under P. flexuosa canopy, under L. divaricata canopy, and in exposed areas. Microhabitats beneath Prosopis showed the highest litter input per m² (between 320 and 527 g/m²), and, consequently, more than 50% of it fell to the soil beneath the canopy of P. flexuosa. Only 10% fell on exposed areas, which exhibited an annual input rate per m² of a lower order of magnitude than the sites under Prosopis. Litterfall presented a peak in summer as a consequence of convective storms, and a second one in autumn due to phenological shedding. Our results suggest that woody species have a central importance in the dynamics of nutrients in arid lands by both the increase of total productivity and litterfall, and the spatial and temporal regulation of litter input.     

Influence of scattered paddock trees on surface soil properties: A study of the Northern Tablelands of NSW

Summary Surface soil conditions were assessed under three tree species on a property near Armidale on the Northern Tablelands of NSW. In both a stocked and adjacent destocked paddock, five trees each of three eucalypt species: Eucalyptus melliodora, Eucalyptus blakelyi and Eucalyptus nova‐anglica, were selected. Soil samples were collected (depth 0–10 cm) along transects 20 m in length running from beneath the tree canopy progressively outwards into the open paddock. Six additional transects were also sampled outside the influence of the trees. Soil properties at a distance from the trees differed little between the stocked and destocked paddock with only a slight acidification in the stocked paddock. However, soil properties around the scattered trees showed considerable variation between stocked and destocked equivalents and most notably in a systematic pattern with distance from the trees themselves. For example, bulk density increased significantly, whereas soil pH, carbon, nitrogen and extractable phosphorus contents all decreased significantly with distance from the trees. However, stocking and camping had modified some of these soil properties. In the stocked paddock, the systematic change in nitrogen and phosphorus with distance from the trees was less clear and the degree of dispersion of the data was largest at the most heavily camped site. In this paddock, bulk density was also generally higher whereas pH, carbon and nitrogen contents were lower compared with the destocked equivalent. Extractable phosphorus content was also higher around the trees in the stocked paddock especially where camping activity was most intense. It is concluded that, although animal camping can modify their effects, scattered trees have a beneficial effect on soil properties and in this respect they have value in the grazing system from a soil conservation perspective.     

Variation in litter under individual tree crowns: Implications for scattered tree ecosystems

In forest ecosystems litter is usually assessed in terms of the average amount produced by the canopy. In scattered tree ecosystems this approach is problematic because the canopy is discontinuous and the spatial arrangement of litter highly variable. We addressed this problem by quantifying the spatial variation in litter load and litter composition associated with individual trees in a Eucalyptus melliodora – Eucalyptus blakelyi woodland. Litter was sampled under crowns and in grassland adjacent to 10 E. blakelyi and 10 E. melliodora trees ranging in diameter at breast height (dbh) from 14 to 129 cm. A total of 302 L samples were collected from these trees, at distances ranging from 0 to 42 m from main stem. The sampled litter loads ranged from 0.02 to 109.3 t ha^?1 and were significantly higher under tree crowns than in grassland for litter and each component of litter (leaves, bark, fine twigs, coarse twigs). In particular, the mean litter load under tree crowns (12.5 t ha^?1) was an order of magnitude higher than the mean litter load in grassland (1.27 t ha^?1). There was a significant (P = 0.0103) positive relationship between mean litter load under the tree crown and dbh, indicating larger trees produced more litter per unit area of ground than smaller trees. Generalized Linear Modelling produced highly significant (P < 0.0001) models predicting the spatial variation in litter load and litter composition in terms of distance from main stem and dbh. Our models demonstrate gradients in litter load and composition under tree crowns. These gradients were most pronounced for the large trees in our study. The disproportionate input of litter and variety of litter components associated with large trees in our study supports their keystone role in scattered tree ecosystems and highlights the need to maintain these structures in agricultural landscapes.     

Carbon pools and fluxes in Bruguiera parviflora dominated naturally growing mangrove forest of Peninsular Malaysia

Estimation of carbon pools and fluxes were conducted in Bruguiera parviflora dominated naturally growing protected mangrove forest in Kuala Selangor Nature Park of Peninsular Malaysia. Above and below-ground carbon pools in seedlings were estimated from destructive methods. While, carbon pools and fluxes in saplings and trees were estimated from the derived allometric biomass equations. Carbon concentrations in different parts of seedlings, saplings and trees; and litter were measured during the dry, wet and intermediate seasons. Soil cores up to 1 m were analyzed to measure carbon concentrations and bulk densities at different depths. Litter standing crop of the study area was measured at the dry, wet and intermediate seasons and the range of total amount of litter standing crop was from 0.66 to 0.88 Mg/ha. Carbon concentration found to vary with the plant and litter parts; and also with the seasons and the range of mean weighted carbon concentration was 40.19 ± 0.87–56.52 ± 1.01 %. The carbon pools in seedling, sapling, tree and litter were 0.69, 0.51, 82.62 and 0.41 Mg C/ha respectively. However, 13.95 Mg C/ha/year of carbon flux was associated with saplings, trees and litter. The estimated carbon pool in the soil (up to 1 m) of the study area was 488.04 Mg C/ha. The findings of this study are the first estimation of carbon pools and fluxes in B. parviflora dominated sites and suggests the potential of this site as a carbon pool.     

Relationship of plant diversity with litter and soil available nitrogen in an alpine meadow under a 9-year grazing exclusion

Grazing exclusion is widely used globally to restore degraded grasslands. Plant diversity has important impacts on grassland ecosystem functions, including grassland productivity and carbon storage. In this study, we selected a Kobresia meadow on the Qinghai–Tibetan Plateau to investigate how grazing exclusion affects plant diversity. Inorganic nitrogen (NH₄ ⁺ and NO₃ ^?) was also measured because its availability impacts plant growth. We found that plant diversity in the meadow was significantly lower under grazing exclusion (fenced meadow) for 9 years compared with moderate grazing. Accumulated litter was significantly higher under grazing exclusion (386.41 g m^?2) compared with grazing (58.77 g m^?2). Soil inorganic nitrogen at 0–5 cm depth was significantly higher under grazing exclusion (13.60 × 10^?2 g kg^?1) than under grazing (9.40 × 10^?2 g kg^?1). The composition of the four functional groups (grasses, sedges, legumes, and forbs) might alter in response to significant changes in the amount of litter and soil available nitrogen content under grazing exclusion compared with grazing. However, the enhanced soil available nitrogen content showed weak feedbacks on plant diversity. In conclusion, light limitation induced by increased amounts of litter may be the main factor causing decreased plant diversity in grazing-excluded meadows compared with moderately grazed meadows.     

Spatial variability of nitrous oxide emissions from an Australian irrigated dairy pasture

Understanding spatial variability of emissions of nitrous oxide (N₂O) is essential to understanding of N₂O emissions from soils to the atmosphere and in the design of statistically valid measurement programs to determine plot, farm and regional emission rates. Two afternoon, ‘snap-shot’ experiments were conducted; one in the summer and one in the autumn of 2004, to examine the statistics and soil variables affecting the spatial variability of N₂O emissions at paddock scale. Small, static chambers (mini-chambers) were placed at 100 locations over an 8,100 m² area of irrigated dairy pasture in northern Victoria, Australia. Chamber headspace was sampled for N₂O and soil samples taken below each mini-chamber were analysed for soil nitrate (NO₃ ^-), ammonium (NH₄ ⁺) and other chemical and physical properties known to affect N₂O emissions. The experiments took place immediately after the sequence of grazing, urea application and irrigation. Nitrous oxide emissions and soil variables were analysed using classical statistics to investigate the effect of soil variables on N₂O emissions. Geostatistics were used to investigate spatial patterns of N₂O emissions and soil variables over the measurement area. Nitrous oxide emissions were extremely variable; 45–765 ng N₂O–N m^?2 s^?1 and 20–953 ng N₂O–N m^?2 s^?1 for the two experiments with corresponding averages of 165 and 138 ng N₂O–N m^?2 s^?1. Nitrous oxide emissions showed spatial dependence up to 73 and 51 m for the two experiments. Nitrous oxide emissions showed significant correlation with soil nutrients in decreasing order of NO₃ ^-, NH₄ ⁺ and available-P concentrations. There was no significant correlation of N₂O emissions with measured soil physical properties.     

Effects of exotic and native tree leaf litter on soil properties of two contrasting sites in the Iberian Peninsula

Aims

We assessed the effects of native and exotic tree leaf litter on soil properties in two contrasting scenarios. The native Quercus robur and Pinus pinaster tree species coexist with the aliens Eucalyptus globulus and Acacia dealbata in acid soils of NW Spain. The native trees Fraxinus angustifolia and Ulmus minor coexist with the aliens Ailanthus altissima, Robinia pseudoacacia and Ulmus pumila in eutrophic basic riparian soils in Central Spain.

Methods

Four plastic trays per species were filled with homogenized top-soil of the site and covered with leaf litter. Before and after 9?months of incubation, litter mass, soil pH, organic matter, mineral and total N were measured. Available mineral N (NO ₃ ^? -N and NH ₄ ⁺ -N) was assessed every 2?months.

Results

Soil biological activity was higher in the basic than in the acid soil. Litter of the exotic trees tended to decompose less than litter of native species, probably due to the presence of secondary metabolites in the former. Soil pH, mineral and total N responded differently to different litter types, irrespective of their exotic or native origin (acid soil), or was similar across litter treatments (basic riparian soil). The similar response of the basic soil to the addition of different litter types may be due to the low contrast of litter quality between the species. E. globulus litter inhibitied soil microbial activity much more than the rest of the studied litter types, leading to a drastic impoverishment of N in soils.

Conclusion

Litter of exotic N-fixing trees (A. dealbata and R. pseudoacacia) did not increase soil N pools because of the inhibition of microbial activity by secondary compounds. Therefore, secondary metabolites of the litter played a major role explaining exotic litter impact on soil properties.     

Changes to the N cycle following bark beetle outbreaks in two contrasting conifer forest types

Outbreaks of Dendroctonus beetles are causing extensive mortality in conifer forests throughout North America. However, nitrogen (N) cycling impacts among forest types are not well known. We quantified beetle-induced changes in forest structure, soil temperature, and N cycling in Douglas-fir (Pseudotsuga menziesii) forests of Greater Yellowstone (WY, USA), and compared them to published lodgepole pine (Pinus contorta var. latifolia) data. Five undisturbed stands were compared to five beetle-killed stands (4–5 years post-outbreak). We hypothesized greater N cycling responses in Douglas-fir due to higher overall N stocks. Undisturbed Douglas-fir stands had greater litter N pools, soil N, and net N mineralization than lodgepole pine. Several responses to disturbance were similar between forest types, including a pulse of N-enriched litter, doubling of soil N availability, 30–50 % increase in understory cover, and 20 % increase in foliar N concentration of unattacked trees. However, the response of some ecosystem properties notably varied by host forest type. Soil temperature was unaffected in Douglas-fir, but lowered in lodgepole pine. Fresh foliar %N was uncorrelated with net N mineralization in Douglas-fir, but positively correlated in lodgepole pine. Though soil ammonium and nitrate, net N mineralization, and net nitrification all doubled, they remained low in both forest types (<6 μg N g soil^?1 NH₄ ⁺or NO₃ ^?; <25 μg N g soil^?1 year^?1 net N mineralization; <8 μg N g soil^?1 year^?1 net nitrification). Results suggest that beetle disturbance affected litter and soil N cycling similarly in each forest type, despite substantial differences in pre-disturbance biogeochemistry. In contrast, soil temperature and soil N–foliar N linkages differed between host forest types. This result suggests that disturbance type may be a better predictor of litter and soil N responses than forest type due to similar disturbance mechanisms and disturbance legacies across both host–beetle systems.     

Grazing history effects on above- and below-ground litter decomposition and nutrient cycling in two co-occurring grasses

Large herbivores may alter carbon and nutrient cycling in soil by changing above- and below-ground litter decomposition dynamics. Grazing effects may reflect changes in plant allocation patterns, and thus litter quality, or the site conditions for decomposition, but the relative roles of these broad mechanisms have rarely been tested. We examined plant and soil mediated effects of grazing history on litter mass loss and nutrient release in two grazing-tolerant grasses, Lolium multiflorum and Paspalum dilatatum, in a humid pampa grassland, Argentina. Shoot and root litters produced in a common garden by conspecific plants collected from grazed and ungrazed sites were incubated under both grazing conditions. We found that grazing history effects on litter decomposition were stronger for shoot than for root material. Root mass loss was neither affected by litter origin nor incubation site, although roots from the grazed origin immobilised more nutrients. Plants from the grazed site produced shoots with higher cell soluble contents and lower lignin:N ratios. Grazing effects mediated by shoot litter origin depended on the species, and were less apparent than incubation site effects. Lolium shoots from the grazed site decomposed and released nutrients faster, whereas Paspalum shoots from the grazed site retained more nutrient than their respective counterparts from the ungrazed site. Such divergent, species-specific dynamics did not translate into consistent differences in soil mineral N beneath decomposing litters. Indeed, shoot mass loss and nutrient release were generally faster in the grazed grassland, where soil N availability was higher. Our results show that grazing influenced nutrient cycling by modifying litter breakdown within species as well as the soil environment for decomposition. They also indicate that grazing effects on decomposition are likely to involve aerial litter pools rather than the more recalcitrant root compartment.     

The influence of mistletoes on nutrient cycling in a semi-arid savanna, southwest Zimbabwe

Parasitic plants are increasingly becoming the focus of research in many ecosystems. They have been shown to alter litterfall properties and decomposition rates in environments where they occur. Despite this recognition, the role of mistletoes in nutrient cycling in semi-arid savanna remains poorly understood. We investigated the litter input, element returns, and associated below-canopy soil nutrient concentrations of three mistletoe species (Erianthemum ngamicum, Plicosepalus kalachariensis, and Viscum verrucosum) that parasitize Acacia karroo trees in a semi-arid savanna, southwest Zimbabwe. Element concentrations in mistletoe leaf litter were enriched relative to the host. Litterfall from mistletoes significantly increased overall litterfall by up to 173 %, with E. ngamicum and P. kalachariensis having greater litterfall than their host trees. Associated with these changes in litterfall was an increase in element returns and the below-canopy soil nutrient concentrations. The increase in nutrient returns was due to both the effect of enriched mistletoe litter and increased volumes of litterfall beneath host trees. Litterfall, element returns, and the below-canopy soil nutrient concentrations were significantly influenced by mistletoe density, with higher values at high mistletoe density. Overall, E. ngamicum and P. kalachariensis had greater influence on litterfall, element returns, and soil nutrient concentrations than V. verrucosum. These findings are consistent with current understanding of enhanced nutrient cycling in the presence of parasitic plants particularly in nutrient-poor ecosystems. We conclude that the introduction of nutrients and associated increase in resource heterogeneity play an important role in determining ecosystem structure and function in semi-arid savannas.     

Rehabilitation of Semiarid Landscapes in Australia. II. Restoring Vegetation Patches

This paper describes a practical technique, tested experimentally, for rehabilitating degraded semiarid landscapes in Australia. This rehabilitation technique is based on the ecological principle that semiarid landscapes are spatially organized as patchy, source-sink systems; this patchy organization functions to conserve limited water and nutrients within the system. The aim was to rebuild vegetation patchiness, lost through decades of utilization of these landscapes as rangelands. Patches were reconstructed from large tree branches and shrubs obtained locally and placed in elongated piles along contours. These piles of branches were very effective in recreating productive soil patches within the landscape, as described in part I of this study. These new patchy habitats promoted the establishment and growth of perennial grasses. Although the foliage cover of these grasses declined into a drought, which started before the end of the experiment, plant survivorship remained high. This suggests that patches also function as refugia for organisms during droughts. The patches of branches remained robust and functional, even under grazing impacts, although plant growth and survival were significantly higher within an ungrazed paddock than in a grazed paddock.     

Deposition and decomposition of cattle dung and its impact on soil properties and plant growth in a cool-temperate pasture

Livestock dung provides an important direct pathway by which carbon and nutrients enter soils in pasture ecosystems and affects carbon and nitrogen cycling indirectly through changes in soil and plant properties. Here, we quantify dung deposition, decomposition, and the effects of dung on soil and plants in a Zoysia japonica grassland in Japan. We determined (1) the distribution of dung, (2) the mass loss rate of dung and the amount of carbon respired as CO₂, and (3) changes in soil properties and aboveground biomass of Z. japonica. Dung deposition was 4.0–9.7 g C and 0.4–1.0 g N m^?2 year^?1 and distributed patchily (Morishita’s I _δ  > 1). Most (71 %) of the carbon in dung deposited in June was lost within a single grazing period by aerobic decomposition, more than mass loss rate of Z. japonica litter in the first year (about 50 %), suggesting that grazing and defecation can accelerate carbon cycling compared with the typical litterfall–decomposition regime. Nitrogen in dung mass entered the soil as ammonium nitrogen and was nitrified. The spatiotemporal distribution of these processes corresponded to that of stimulated Z. japonica growth. These results suggested that dung deposition significantly affected the inorganic nitrogen status of soil and, therefore, the growth of Z. japonica. However, these effects were very restricted temporally (July–August) and spatially (within 10 cm from dung edge). Thus, such spatiotemporally restricted effects combined with the patchy distribution of dung may contribute to the heterogeneous structure of pasture ecosystems.     

The vertical distribution of N and K uptake in relation to root distribution and root uptake capacity in mature Quercus robur, Fagus sylvatica and Picea abies stands

We have measured the uptake capacity of nitrogen (N) and potassium (K) from different soil depths by injecting ¹⁵N and caesium (Cs; as an analogue to K) at 5 and 50 cm soil depth and analysing the recovery of these markers in foliage and buds. The study was performed in monocultures of 40-year-old pedunculate oak (Quercus robur), European beech (Fagus sylvatica) and Norway spruce (Picea abies (L.) Karst.) located at an experimental site in Palsgård, Denmark. The markers were injected as a solution through plastic tubes around 20 trees of each species at either 5 or 50 cm soil depth in June 2003. After 65 days foliage and buds were harvested and the concentrations of ¹⁵N and Cs analysed. The recovery of ¹⁵N in the foliage and buds tended to be higher from 5 than 50 cm soil depth in oak whereas they where similar in spruce and beech after compensation for differences in immobilization of ¹⁵N in the soil. In oak more Cs was recovered from 5 than from 50 cm soil depth whereas in beech and spruce no difference could be detected. Out of the three investigated tree species, oak was found to have the lowest capacity to take up Cs at 50 cm soil depth compared to 5 cm soil depth also after compensating for differences in discrimination against Cs by the roots. The uptake capacity from 50 cm soil depth compared with 5 cm was higher than expected from the root distribution except for K in oak, which can probably be explained by a considerable overlap of the uptake zones around the roots and mycorrhizal hyphae in the topsoil. The study also shows that fine roots at different soil depths with different physiological properties can influence the nutrient uptake of trees. Estimates of fine root distribution alone may thus not reflect the nutrient uptake capacity of trees with sufficient accuracy. Our study shows that deep-rooted trees such as oak may have lower nutrient uptake capacity at deeper soil layers than more shallow-rooted trees such as spruce, as we found no evidence that deep-rooted trees obtained proportionally more nutrients from deeper soil layers. This has implications for models of nutrient cycling in forest ecosystems that use the distribution of roots as the sole criterion for predicting uptake of nutrients from different soil depths.     

Spatial patterns of P fractions and chemical properties in soils of two native shrub communities in Senegal

Two shrub species (Piliostigma reticulatum (D.C.) Hochst (Caesalpinioideae) and Guiera senegalensis J.F. Gmel (Combretaceae) are commonly found in farmers’ fields at varying densities in semi-arid Senegal and throughout the Sahel where soils have chronically low phosphorus (P) availability. It seems plausible that shrub litter and the rhizospheres could influence P fractions and other chemical soil properties that affect crop productivity. Thus, a study was done at two sites, on the distribution of inorganic and organic soil P pools, organic C levels, and pH in soil beneath and outside the canopies of P. reticulatum and G. senegalensis (0-30 cm depth). Both sites had low total P ranging from 64 mg P kg^?1 to 135 mg P kg^-1, and low extractable PO₄ (resin Pi) (1–6 mg P kg^?1) with P fractions dominated by NaOH-P. Organic P (Po) made up about 50% of total P, and most of the organic P (>60%) was found in the NaOH-P fractions. The labile P, particularly bicarb-Po was higher in soil beneath shrub canopies (8.4 mg P kg ^?1), than outside the canopy (6.2 mg P kg ^?1). Similarly, C, N and P to a lesser extent, were more concentrated beneath shrub canopies. P. reticulatum soil was dominated by the NaOH-Po fraction, whereas G. senegalensis had higher bicarb-Po at one of the study sites. An index of biologically available organic P (Bicarb-Po) / (Bicarb-Po?+?Bicar-Pi?+?Resin Pi) was ?>?60% and indicates that biological processes represent an important part of P cycling in these shrub ecosystems. The differential ability of shrubs in modifying soil chemical properties under their canopies has major implications for biogeochemical cycling of nutrients and C in sandy soils of semi arid Sahelian ecosystems.     

Spatial pattern of soil nitrogen availability and its relationship to stand structure in a coniferous-broadleaved mixed forest with a dense dwarf bamboo understory in northern Japan

Natural disturbances create spatial patterns of the ecosystem processes and functions in natural forests. However, how dynamics and the spatial structure of forests relate to soil nitrogen dynamics is not well understood. We examined the spatial relationship between the distributions of canopy and understory species, and soil nitrogen dynamics in a natural coniferous-broadleaved mixed forest with a dense understory of Sasa dwarf bamboo in northern Japan. The O horizon was thick where coniferous litter predominated, and it was thin where broadleaved litter predominated. The soil water content was low in areas with a thick O horizon and a high abundance of coniferous trees. The soil nitrate content was low where the soil water content was low, and the soil nitrate content increased linearly with increasing net nitrification potential. These results suggest that the soil nitrate content under the coniferous canopy was lower because of the low nitrification potential of soil microbes in soils with low water contents. The soil nitrate content and nitrification potential were higher in the canopy gap than under the canopy. Our results suggest that forest structure, specifically the thickness of the forest floor, significantly affects the spatial pattern of the soil water content, thereby creating a spatial pattern of soil nitrogen availability at a relatively small scale with flat topography. The higher nitrification potential under the canopy gap could pose a long-term risk of nitrate leaching because of the suppression of the natural regeneration of canopy species by dense Sasa dwarf bamboo in this forest ecosystem.     

Spatial and temporal changes in ecosystem carbon pools following juniper encroachment and removal

Proliferation of woody plants is a predominant global land cover change of the past century, particularly in dryland ecosystems. Woody encroachment and its potential impacts (e.g., livestock forage, wildlife habitat, hydrological cycling) have led to widespread brush management. Although woody plants may have substantial impacts on soils, uncertainty remains regarding woody encroachment and brush management influences on carbon (C) pools. Surface C pools (shallow soils and litter) may be particularly dynamic in response to encroachment and brush management. However, we have limited understanding of spatiotemporal patterns of surface C responses or how surface pools respond relative to aboveground C, litter, roots, and deep soil organic C. Spatial variability and lack of basic ecological data in woody-encroached dryland ecosystems present challenges to filling this data gap. We assessed the impact of western juniper (Juniperus occidentalis) encroachment and removal on C pools in a semi-arid sagebrush ecosystem. We used spatially-intensive sampling to create sub-canopy estimates of surface soil C (0–10 cm depth) and litter C pools that consider variation in tree size/age and sub-canopy location for live juniper and around stumps that were cut 7 years prior to sampling. We coupled the present size distribution of junipers with extensive existing allometric information about juniper in this region to estimate how landscape-level C pools would change through time under future management and land cover scenarios. Juniper encroachment and removal leads to substantial changes in C pools. Best-fit models for surface soil and litter C included positive responses to shrub basal diameter and negative responses to increasing relative distance from the bole to dripline. Juniper removal led to a net loss of surface C as a function of large decreases in litter C and small increases in surface soil C. At the landscape scale, deep soil C was the largest C pool (77 Mg C ha^?1), with an apparent lack of sensitivity to management. Overall, encroachment led to substantial increases in C storage over time as juniper size increased (excluding deep soil C, ecosystem C pools increased from 13.5 to 30.2 Mg C ha^?1 with transition from sagebrush-dominated to present encroachment levels). The largest pool of accumulation was juniper aboveground C, with important other pools including juniper roots, litter, and surface soil C. Woody encroachment and subsequent brush management can have substantive impacts on ecosystem C pools, although our data suggest the spatiotemporal patterns of surface C pools need to be properly accounted for to capture C pool responses. Our approach of coupling spatially-intensive surface C information with shrub distribution and allometric data is an effective method for characterizing ecosystem C pools, offering an opportunity for filling in knowledge gaps regarding woody encroachment and brush management impacts on local-to-regional ecosystem C pools.     

Soil properties associated with the tree decline ‘Mundulla Yellows’

Czerniakowski, B., Crnov, R., Smith, I. W. and Luck, J.E. 2006. Mundulla Yellows (MY) is a progressive dieback syndrome of Eucalyptus and other native species that was first reported in the 1970s. Despite being observed in Australia for over 30 years, the cause of MY has not been determined. To investigate the role of soil properties in MY, foliage and soil from underneath 40 Eucalyptus camaldulensis Dehnh., E. leucoxylon F. Muell. or E. cladocalyx F. Muell. trees from ten sites in South Australia and Victoria, Australia, were analysed. Soil from sites with symptomatic trees had significantly higher pH, EC and lower available iron when compared to soil from sites with asymptomatic trees. High levels of carbonates (CO₃²⁻/HCO₃⁻) dominated the aqueous soil extract from sites with symptomatic trees. Foliage analysis of 20 symptomatic trees indicated lower levels of total Fe and Mn and higher levels of Na and Cl, compared to 20 asymptomatic trees. This is the first report that associates soil and nutrients with Mundulla Yellows tree decline.