Fence Experiment Research Articles

Rapid Ice‐Wedge Collapse and Permafrost Carbon Loss Triggered by Increased Snow Depth and Surface Runoff

AbstractThicker snow cover in permafrost areas causes deeper active layers and thaw subsidence, which alter local hydrology and may amplify the loss of soil carbon. However, the potential for changes in snow cover and surface runoff to mobilize permafrost carbon remains poorly quantified. In this study, we show that a snow fence experiment on High‐Arctic Svalbard inadvertently led to surface subsidence through warming, and extensive downstream erosion due to increased surface runoff. Within a decade of artificially raised snow depths, several ice wedges collapsed, forming a 50 m long and 1.5 m deep thermo‐erosion gully in the landscape. We estimate that 1.1–3.3 tons C may have eroded, and that the gully is a hotspot for processing of mobilized aquatic carbon. Our results show that interactions among snow, runoff and permafrost thaw form an important driver of soil carbon loss, highlighting the need for improved model representation.

Stochastic Processes Drive Plant Community Assembly in Alpine Grassland during the Restoration Period

Enclosure (prohibition of grazing) is an important process to restore alpine grassland on the Qinghai-Tibetan Plateau. However, few studies have quantified the extent to which the long-term enclosure may contribute to the changes in plant phylogenetic diversity and community assembly in alpine grassland under environmental change. In this study, based on an 11-year fencing experiment along an altitudinal gradient ranging from 4400 m to 5200 m in central Tibet, we conducted an observation of species composition and coverage within and outside the fences in the fifth, eighth and eleventh year, and monitored the related climate and soil factors at 7 sites. Our aim is to quantify the relative effects of environmental change and grassland management on the alpine plant community assemblage. The results were: (1) the overall phylogenetic structure (NRI) of the alpine plant communities, whether inside or outside the enclosure, was divergent at altitudes where the environment was relatively unextreme (4800–5100 m), but aggregative at altitudes with low precipitation (4400–4650 m) or with low temperature (5200 m). (2) The phylogenetic structure of the nearest taxon of species (NTI) was more aggregative along the whole gradient. (3) Precipitation was the dominant factor driving the changes in species richness, phylogenetic diversity and community α-phylogenetic structure indices (NRI and NTI), followed by enclosure duration and soil C:N ratio. (4) The phylogenetic structure of the communities was similar at higher altitudes under grazing or enclosure treatments, and was opposite at lower sites. Stochastic processes have driven the changes in the communities between inside and outside the fences at all altitudes. In addition, homogeneous dispersal occurred in communities at higher sites. In summary, the 11-year enclosure had little effect on community structure of alpine meadows where the grazing pressure is relative lower, whereas it could help restore the community of steppe meadow at lower altitudes where the grazing pressure is extensively higher. This study may provide a vital theoretical support for the formulation of differential management for alpine grassland on the Tibetan Plateau.

Assessment of Soil Erosion Potential From the Disturbed Surface of Skid Trails in Small Shovel Harvesting System

Forest roads, haul roads, and especially skid trails have been associated with sedimentation and soil erosion risk. Despite the widespread small shovel harvesting system on steep terrains in South Korea, the subsequent risks of deep (rut depth >5 cm) and compact disturbances, and erosion rates in skid trails are largely unknown. Therefore, the main objectives of this study were to compare the soil erosion rate in each skid trail and predict the total soil erosion rate in a small shovel harvesting area. The soil erosion rate was measured at the plot scale (5 × 3 m) in different skid trail parts (bladed skid trail by small-shovel loader passage, BT; and compacted skid trail CT by carrier passage with construction by a small-shovel loader) using a silt fence experiment. In addition, we investigated the applicability of the Water Erosion Prediction Project (WEPP) model to each disturbance. Among all disturbances, the highest erosion rate (average value of 9.13 ± 0.96 kg m−2 4 months−1) was because of CT. The model predictions were over- and under-estimated and showed particularly poor performance where uncovered soil was exposed (less than 1%) to high machine traffic frequency and excavation. Further, the annual soil erosion rates ranged from 11.59 to 28.94 ton ha−1 year−1. The results suggested that the WEPP model could partially validate the soil erosion results, and further research is still required to improve the accuracy of the model.

Plant foliar nutrient response to active layer and water table depth in warming permafrost soils

AbstractThawing permafrost in northern latitudes has led to deepening active soil layers and fluctuating water tables. This could increase plant access to permafrost‐derived nitrogen (N), phosphorus (P) and other nutrients such as calcium (Ca) and magnesium (Mg), and subsequently increase plant productivity and ecosystem carbon storage and nutrient cycling. We hypothesized that deepening permafrost thaw and water table fluctuations would alter species‐specific foliar N:P ratios. Since there is often more P, Ca and Mg available in the deeper mineral soil layers and more N available in the shallow organic layers, we expected that deeply rooted species would decrease foliar N:P ratios due to root proximity to thawing mineral soil, and plants with shallower rooting systems mostly in the organic layer would increase foliar N:P ratios.We assessed foliar and canopy nutrient responses of seven vascular plant species in moist acidic tussock tundra vegetation in the northern foothills of the Alaska Range to variable soil thaw depths and water table levels induced by either a natural thermokarst gradient or a winter warming snow fence experiment.In both the natural thermokarst gradient and the warming experiment, wet or deeply thawed areas generally led to an increase in foliar nutrient concentrations and greater canopy mass and canopy nutrients. For the majority of species, foliar N:P ratios remained proportional or decreased in deciduous species in wet sites, with the exception of one shallowly rooted species that increased foliar N:P ratio in deeply thawed sites. Overall, plant acquisition of P was more related to water table level than to thaw depth, and water table modulated the canopy biomass response of the species at the warming experiment.Synthesis. Foliar N:P ratios suggest that plant species in this tussock tundra ecosystem are either remaining or becoming more N‐limited as thaw depth deepens and water table level rises, indicating that P is not likely to become the primary limiting nutrient with the progression of permafrost thaw. However, the amount of deeply thawed, wet areas that develop on the landscape as permafrost thaws will be important contributors in the total movement of nutrients above‐ground.

Feral ungulate and macropod responses to resource scarcity and predation risk at savanna waterholes

When exotic species are introduced to new environments, they often have a competitive advantage over native species. In northern Australia, pigs, cattle, and water buffalo have established widespread, feral populations. As ungulates have high water requirements, they typically congregate near waterpoints. We used a fencing experiment to test whether native macropods preferentially visited savanna waterholes where large ungulates were excluded. We also investigated whether water scarcity affected the visitation behaviour and temporal activity patterns of herbivores at waterholes and whether increasing prey aggregation at waterholes increased dingo presence. We found that macropods did not use fenced waterholes preferentially over unfenced ones. Cattle presence at waterholes increased as water became scarce, while macropod and pig presence peaked in the middle of the dry season. Macropod activity declined rapidly at the end of the dry season when cattle activity was greatest, suggesting that macropods may avoid waterholes in areas utilised by cattle when competition for resources is high. Macropods and all ungulates visited waterholes more during a drought year compared to an average rainfall year. Despite increasing prey activity, dingo presence at waterholes did not increase when water became scarce. However, dingo presence increased significantly on moonless nights. Our results suggest that competition between macropods and ungulates may intensify during periods of water scarcity. Climate change and pastoral intensification are likely to increase competition for resources between ungulates and macropods in Australian savannas, potentially threatening macropod populations across the landscape in the future.Significance statementIn northern Australia, feral populations of pigs, cattle, and water buffalo compete with native wildlife for access to water sources. As interspecific competition favours species with a size advantage, we tested whether kangaroos and wallabies (macropods) preferentially use waterholes where large ungulates (cattle and buffalo) were excluded. We found that macropods avoided waterholes when cattle presence was high but did not preferentially use waterholes where livestock were excluded. When water scarcity peaked during a drought, macropods and all three feral ungulate species visited waterholes more. However, increased prey presence at waterholes during the drought did not correspond with increased predator (dingo) presence. Our study advances the understanding of behavioural interactions between invasive and native species at important shared resources, and how this may affect wildlife conservation in an increasingly unpredictable environment.

Development of a snowdrift model with the lattice Boltzmann method

We developed a snowdrift model to evaluate the snowdrift height around snow fences, which are often installed along roads in snowy, windy locations. The model consisted of the conventional computational fluid dynamics solver that used the lattice Boltzmann method and a module for calculating the snow particles’ motion and accumulation. The calculation domain was a half channel with a flat free-slip boundary on the top and a non-slip boundary on the bottom, and an inflow with artificially generated turbulence from one side to the outlet side was imposed. In addition to the reference experiment with no fence, experiments were set up with a two-dimensional and a three-dimensional fence normal to the dominant wind direction in the channel center. The estimated wind flow over the two-dimensional fence was characterized by a swirling eddy in the cross section, whereas the wind flow in the three-dimensional fence experiment was horizontally diffluent with a dipole vortex pair on the leeward side of the fence. Almost all the snowdrift formed on the windward side of the two-dimensional and three-dimensional fences, although the snowdrift also formed along the split streaks on the leeward side of the three-dimensional fence. Our results suggested that the fence should be as long as possible to avoid snowdrifts on roads.

Exploring drivers of litter decomposition in a greening Arctic: results from a transplant experiment across a treeline.

Decomposition of plant litter is a key control over carbon (C) storage in the soil. The biochemistry of the litter being produced, the environment in which the decomposition is taking place, and the community composition and metabolism of the decomposer organisms exert a combined influence over decomposition rates. As deciduous shrubs and trees are expanding into tundra ecosystems as a result of regional climate warming, this change in vegetation represents a change in litter input to tundra soils and a change in the environment in which litter decomposes. To test the importance of litter biochemistry and environment in determining litter mass loss, we reciprocally transplanted litter between heath (Empetrum nigrum), shrub (Betula nana), and forest (Betula pubescens) at a sub‐Arctic treeline in Sweden. As expansion of shrubs and trees promotes deeper snow, we also used a snow fence experiment in a tundra heath environment to understand the importance of snow depth, relative to other factors, in the decomposition of litter. Our results show that B. pubescens and B. nana leaf litter decomposed at faster rates than E. nigrum litter across all environments, while all litter species decomposed at faster rates in the forest and shrub environments than in the tundra heath. The effect of increased snow on decomposition was minimal, leading us to conclude that microbial activity over summer in the productive forest and shrub vegetation is driving increased mass loss compared to the heath. Using B. pubescens and E. nigrum litter, we demonstrate that degradation of carbohydrate‐C is a significant driver of mass loss in the forest. This pathway was less prominent in the heath, which is consistent with observations that tundra soils typically have high concentrations of “labile” C. This experiment suggests that further expansion of shrubs and trees may stimulate the loss of undecomposed carbohydrate C in the tundra.

Winter Ecosystem Respiration and Sources of CO2 From the High Arctic Tundra of Svalbard: Response to a Deeper Snow Experiment

AbstractCurrently, there is a lack of understanding on how the magnitude and sources of carbon (C) emissions from High Arctic tundra are impacted by changing snow cover duration and depth during winter. Here we investigated this issue in a graminoid tundra snow fence experiment on shale‐derived gelisols in Svalbard from the end of the growing season and throughout the winter. To characterize emissions, we measured ecosystem respiration (Reco) along with its radiocarbon (14C) content. We assessed the composition of soil organic matter (SOM) by measuring its bulk‐C and nitrogen (N), 14C content, and n‐alkane composition. Our findings reveal that greater snow depth increased soil temperatures and winter Reco (25 mg C m−2 d−1 under deeper snow compared to 13 mg C m−2 d−1 in ambient conditions). At the end of the growing season, Reco was dominated by plant respiration and microbial decomposition of C fixed within the past 60 years (Δ14C = 62 ± 8‰). During winter, emissions were significantly older (Δ14C = −64 ± 14‰), and likely sourced from microorganisms decomposing aged SOM formed during the Holocene mixed with biotic or abiotic mineralization of the carbonaceous, fossil parent material. Our findings imply that snow cover duration and depth is a key control on soil temperatures and thus the magnitude of Reco in winter. We also show that in shallow Arctic soils, mineralization of carbonaceous parent materials can contribute significant proportions of fossil C to Reco. Therefore, permafrost‐C inventories informing C emission projections must carefully distinguish between more vulnerable SOM from recently fixed biomass and more recalcitrant ancient sedimentary C sources.

Land’s complex role in climate change

To mitigate climate change at local, regional, and global scales, we must begin to think beyond greenhouse gases.

The fence experiment — a first evaluation of shelter models

We present a preliminary evaluation of shelter models of different degrees of complexity using full-scale lidar measurements of the shelter on a vertical plane behind and orthogonal to a fence. Model results accounting for the distribution of the relative wind direction within the observed direction interval are in better agreement with the observations than those that correspond to the simulation at the center of the direction interval, particularly in the far-wake region, for six vertical levels up to two fence heights. Generally, the CFD results are in better agreement with the observations than those from two engineering-like obstacle models but the latter two follow well the behavior of the observations in the far-wake region.

The fence experiment – full-scale lidar-based shelter observations

Abstract. We present shelter measurements of a fence from a field experiment in Denmark. The measurements were performed with three lidars scanning on a vertical plane downwind of the fence. Inflow conditions are based on sonic anemometer observations of a nearby mast. For fence-undisturbed conditions, the lidars' measurements agree well with those from the sonic anemometers and, at the mast position, the average inflow conditions are well described by the logarithmic profile. Seven cases are defined based on the relative wind direction to the fence, the fence porosity, and the inflow conditions. The larger the relative direction, the lower the effect of the shelter. For the case with the largest relative directions, no sheltering effect is observed in the far wake (distances ⪆ 6 fence heights downwind of the fence). When comparing a near-neutral to a stable case, a stronger shelter effect is noticed. The shelter is highest below ≈ 1.46 fence heights and can sometimes be observed at all downwind positions (up to 11 fence heights downwind). Below the fence height, the porous fence has a lower impact on the flow close to the fence compared to the solid fence. Velocity profiles in the far wake converge onto each other using the self-preserving forms from two-dimensional wake analysis.

Long-term increase in snow depth leads to compositional changes in arctic ectomycorrhizal fungal communities.

Many arctic ecological processes are regulated by soil temperature that is tightly interconnected with snow cover distribution and persistence. Recently, various climate-induced changes have been observed in arctic tundra ecosystems, e.g. shrub expansion, resulting in reduction in albedo and greater C fixation in aboveground vegetation as well as increased rates of soil C mobilization by microbes. Importantly, the net effects of these shifts are unknown, in part because our understanding of belowground processes is limited. Here, we focus on the effects of increased snow depth, and as a consequence, increased winter soil temperature on ectomycorrhizal (ECM) fungal communities in dry and moist tundra. We analyzed deep DNA sequence data from soil samples taken at a long-term snow fence experiment in Northern Alaska. Our results indicate that, in contrast with previously observed responses of plants to increased snow depth at the same experimental site, the ECM fungal community of the dry tundra was more affected by deeper snow than the moist tundra community. In the dry tundra, both community richness and composition were significantly altered while in the moist tundra, only community composition changed significantly while richness did not. We observed a decrease in richness of Tomentella, Inocybe and other taxa adapted to scavenge the soil for labile N forms. On the other hand, richness of Cortinarius, and species with the ability to scavenge the soil for recalcitrant N forms, did not change. We further link ECM fungal traits with C soil pools. If future warmer atmospheric conditions lead to greater winter snow fall, changes in the ECM fungal community will likely influence C emissions and C fixation through altering N plant availability, fungal biomass and soil-plant C-N dynamics, ultimately determining important future interactions between the tundra biosphere and atmosphere.

Deepened winter snow increases stem growth and alters stem δ13C and δ15N in evergreen dwarf shrub Cassiope tetragona in high-arctic Svalbard tundra

Deeper winter snow is hypothesized to favor shrub growth and may partly explain the shrub expansion observed in many parts of the arctic during the last decades, potentially triggering biophysical feedbacks including regional warming and permafrost thawing. We experimentally tested the effects of winter snow depth on shrub growth and ecophysiology by measuring stem length and stem hydrogen (δ2H), carbon (δ13C), nitrogen (δ15N) and oxygen (δ18O) isotopic composition of the circumarctic evergreen dwarf shrub Cassiope tetragona growing in high-arctic Svalbard, Norway. Measurements were carried out on C. tetragona individuals sampled from three tundra sites, each representing a distinct moisture regime (dry heath, meadow, moist meadow). Individuals were sampled along gradients of experimentally manipulated winter snow depths in a six-year old snow fence experiment: in ambient (c. 20 cm), medium (c. 100 cm), and deep snow (c. 150 cm) plots. The deep-snow treatment consistently and significantly increased C. tetragona growth during the 2008–2011 manipulation period compared to growth in ambient-snow plots. Stem δ15N and stem N concentration values were significantly higher in deep-snow individuals compared to individuals growing in ambient-snow plots during the course of the experiment, suggesting that soil N-availability was increased in deep-snow plots as a result of increased soil winter N mineralization. Although inter-annual growing season-precipitation δ2H and stem δ2H records closely matched, snow depth did not change stem δ2H or δ18O, suggesting that water source usage by C. tetragona was unaltered. Instead, the deep insulating snowpack may have protected C. tetragona shrubs against frost damage, potentially compensating the detrimental effects of a shortened growing season and associated phenological delay on growth. Our findings suggest that an increase in winter precipitation in the High Arctic, as predicted by climate models, has the potential to alter the growth and ecophysiology of evergreen shrub C. tetragona through changes in plant mineral nutrition and frost damage protection.

Long-term increases in snow pack elevate leaf N and photosynthesis in Salix arctica: responses to a snow fence experiment in the High Arctic of NW Greenland

We examine the influence of altered winter precipitation on a High Arctic landscape with continuous permafrost. Gas exchange, leaf tissue element and isotopic composition (N, δ13C, δ15N), and plant water sources derived from stem and soil water δ18O were examined in Salix arctica (arctic willow) following a decade of snow-fence-enhanced snow pack in NW Greenland. Study plots in ambient and +snow conditions were sampled in summer 2012. Plants experiencing enhanced snow conditions for 10 years had higher leaf [N], photosynthetic rate, and more enriched leaf δ15N. Enhanced snow did not influence stomatal conductance or depth of plant water use. We attribute the higher photosynthetic rate in S. arctica exposed to deeper snow pack to altered biogeochemical cycles which yielded higher leaf [N] rather than to enhanced water availability. These data demonstrate the complexity of High Arctic plant responses to changes in winter conditions. Furthermore, our data depict the intricate linkages between winter and summer conditions as they regulate processes such as leaf gas exchange that may control water vapor and CO2 feedbacks between arctic tundra and the surrounding atmosphere.

Growth and Reproductive Responses of Cassiope tetragona, a Circumpolar Evergreen Shrub, to Experimentally Delayed Snowmelt

While there has been a general trend of climate warming in the Arctic causing early snowmelt and prolonged growing season, climate change models for some areas in the High Arctic suggest increased snow accumulation, delayed spring melt, and consequently shorter growing season. We tested the vegetative and reproductive responses of Cassiope tetragona, an arctic shrub, to increased snow depth and delayed snowmelt using a snow fence experiment in Adventdalen, Norway (78°54′N, 18°01′E). We recorded seasonal shoot length, number of leaves, and capsules per shoot for three summers (2005–2007): two prior and one after the treatment began. Phenological events were recorded in 2007. The number of seeds per capsule was counted and seed germination was tested. Phenological development was significantly delayed behind fences, with lower shoot length, number of leaves, capsules, and seeds per shoot segment (16, 20, 54 and 11%, respectively) compared to controls, but there was no difference in these parameters between plots prior to fence establishment. Seed viability was unaffected by treatment. A delay in start of growing season due to delayed snowmelt decreased vegetative and reproductive performance of C. tetragona; earlier melt may therefore improve performance. This may lead to floral composition change in some high arctic locations.

The role of land use change on the development and evolution of the west coast trough, convective clouds, and precipitation in southwest Australia

[1] Land clearing for agricultural purposes in southwest Australia has created a landscape where a 750 km rabbit-proof fence separates 13 million hectares of croplands from the remnant native vegetation to the east. The Bunny Fence Experiment (BuFex) was conducted in the vicinity of the intended vermin-proof boundary in December 2005 and August 2007. The experiment examined the role of land cover change (LCC) on the preferential formation of clouds over the native vegetation that often terminates along the vermin-proof fence as well as the regional rainfall reduction observed in this region. Observations and numerical model analysis show that the formation and development of the west coast trough (WCT), which is a synoptic-scale feature that initiates spring and summertime convection, is impacted by land cover change and that the cloud fields induced by the WCT would extend farther west in the absence of the LCC. The surface convergence patterns associated with the wintertime WCT circulation are substantially altered by LCC, due to changes in both WCT dynamics and surface aerodynamic roughness, leading to a rainfall decrease to the west of the rabbit fence. Although this study focuses on only two events, it further illustrates that LCC has significant regional impacts in southwest Western Australia regardless of large-scale shifts in the climate system.

Habitat productivity influences root mass vertical distribution in grazed Mediterranean ecosystems

Herbivores are expected to influence grassland ecosystems by modifying root biomass and root spatial distribution of plant communities. Studies in perennial dominated grasslands suggest that grazing intensity and primary productivity may be strong determinants of the vertical distribution of subterranean biomass. However, no studies have addressed this question in annual dominated pastures. In this study we assess the effect of grazing and habitat productivity on the vertical distribution of root mass in an annual dominated Mediterranean pasture grazed by free-ranging sheep and wild rabbits. We evaluate the effects of grazing on total root mass and vertical root distribution (0–4, 4–8 and 8–12 cm depths) in two neighboring topographic sites (uplands and lowlands) with different productivity using a replicated fence experiment which excludes sheep and sheep plus rabbits. We found evidences that grazing affected root biomass and vertical distribution at lowlands (high productivity habitats), where places grazed by sheep plus rabbits exhibit more root mass and a higher concentration of it towards the soil surface than only rabbits and ungrazed places. In contrast, grazing did not affect root biomass and vertical distribution at uplands (low productivity habitats). We suggest that higher nitrogen and organic matter found in lowlands permit a plant adjustment for nitrogen acquisition by increasing biomass allocation to root production which would allow plant regrowth and the quick completion of the annual life cycle. Contrary, soil resources scarcity at uplands do not permit plants modify their root growth patterns in response to grazing. Our study emphasizes the importance of primary productivity in predicting grazing effect on belowground processes in Mediterranean environments dominated by annuals.

Effect of high density on the short term Calomys musculinus spacing behaviour: A fencing experiment

We studied the short term spacing behavioural responses of corn mice ( Calomys musculinus) with regard to population density in four 0.25 ha enclosures (two control and two experimental) in the 2007 breeding season. The goal of this research was to test the hypothesis that spacing behaviour only operates among C. musculinus adult females. We estimated 207 home ranges to study: 1) the home range size and the overlap degree of adult males and females in relation to population density; 2) the settlement distances of juveniles to the centre of activity of their mothers and the home range overlap proportion between them and their mothers in relation to population density. We found that home range size and overlap degree in C. musculinus adults were determined by sex and density. At high population density males had significant smaller and more exclusive home ranges, and this might reflect induced territoriality derived from social restrictions. Female home range sizes remained similar irrespective of population density, and they kept exclusive home ranges in both control and experimental enclosures. Thus, females maintained their territories independent of the population density values. The settlement distances of juveniles from their mothers and the overlap proportion between them and their mothers were independent of population density. We conclude that spacing behaviour only operates among C. musculinus adult females and it could have a role in regulating population abundances limiting the number of females that acquire breeding spaces.

Impact of fencing on the recovery of the ground flora on heavily eroded slopes of a deciduous forest.

This paper seeks to outline early stages in the recovery of forest ground flora on eroded slopes impacted by recreation activities and to suggest how these data might be applied in the formulation of management policies for forest recreation areas. Based on a fencing experiment in the Sonian Forest near Brussels, we investigated whether, over a 6-year period, the vegetation was able to recover after having been destroyed by recreation use. Short-term trends in overall species composition were already observable during this 6-year study. Species recovery on eroded hills was related to slope, aspect, and soil type. During the considered time scale, the proportion of hemicryptophytes and the number of ancient forest species increased significantly. A downward trend was detected for Ellenberg's nitrogen and temperature indexes and for the proportion of therophytes and pioneer plants of disturbed places. Changes in species' frequencies suggest six recovery strategies: early, late, expanding, disappearing, transient, and fluctuating species. Aside from seedling reproduction from overstory influences, Luzula sylvatica appeared to be the most resilient of the species identified in the study since this species has the highest global frequency in our sampling plots and has increased its cover during the study period. Study results indicate that (1) protection from recreation has initiated the recovery of species in the herb layer, but (2) it may take a long time before vegetation previously present in the ground flora may recover in both density and species composition.

Effect of sharp vegetation boundary on the convective atmospheric boundary layer

An abrupt change of turbulent fluxes and cloud formation were found during the Bunny Fence experiment in the southwestern Australia at a sharp boundary between agricultural and native vegetation. It is suggested that this is a direct consequence of distinct coherent structures formed within atmospheric sheared convective boundary layers. Using large-eddy simulations (LES), it was found that a boundary-attached roll circulation appears at the sharp vegetation boundary. Model calculations revealed significant horizontal heat and moisture flux from the agricultural to the native vegetation area. The additional moisture flux may explain the observed increased cloud formation over the native vegetation as well as contribute to increased rainfall over the native vegetation.