Search

Large areas of alpine meadow across the Qinghai-Tibet Plateau (QTP), western China, are undergoing degradation. Drawing upon field and remotely sensed data we develop a spatially explicit grid-based simulation model to explore the long-term dynamics of alpine meadow communities in this area. Our model represents the spatial dynamics of four plant functional types (PFTs) – sedges, grasses, forbs and weeds – together with disturbance from livestock grazing and small mammal activity. The model is most sensitive to parameters related to the reproductive rate and lifespan of the three plant functional types. Comparisons between model outcomes and field observations of vegetation composition suggest that the model produces plausible predictions of community dynamics. Simulation experiments indicate that grazing intensity is a fundamental control of plant community dynamics in these ecosystems. As grazing intensity increases there is a shift from a community dominated by grasses and forbs (very low grazing level) to grasses and sedges (low and moderate grazing levels) to degraded ground (high grazing levels). Severely degraded alpine meadows (locally termed ‘Heitutan’ or ‘black beach’) emerge after 370 simulation years under high levels of grazing pressure, but after only two decades under extreme grazing pressure. Under low to moderate intensity grazing regimes small mammals play an important role in maintaining meadow ecosystems. However, our model suggests that small mammal activity is no longer beneficial to the grassland ecosystem under high grazing pressures, increasing the rate of Heitutan formation. The time frame for a return to a sedge-dominated community is shortest under a moderate intensity grazing regime, but even in this instance it may take several hundred years.

The rent gap is the difference between the actual and potential ground rent of a site. This paper estimates the rent gap into two ways: gaps estimated using actual and potential ground rents at the...

Due to increasing habitat fragmentation and concern about its ecological effects, there has been an upsurge in the use of landscape connectivity estimates in conservation planning. Measuring connectivity is challenging, resulting in a limited understanding of the efficacy of connectivity estimation techniques and the conditions under which they perform best. We evaluated the performance of four commonly used connectivity metrics – Euclidean distance; least-cost paths (LCP) length and cost; and circuit theory’s resistance distance – over a variety of simulated landscapes. We developed an agent-based model simulating the dispersal of individuals with different behavioural traits across landscapes varying in their spatial structure. The outcomes of multiple dispersal attempts were used to obtain ‘true’ connectivity. These ‘true’ connectivity measures were then compared to estimates generated using the connectivity metrics, employing the simulated landscapes as cost-surfaces. The four metrics differed in the strength of their correlation with true connectivity; resistance distance showed the strongest correlation, closely followed by LCP cost, with Euclidean distance having the weakest. Landscape structure and species behavioural attributes only weakly predicted the performance of resistance distance, LCP cost and length estimates, with none predicting Euclidean distance’s efficacy. Our results indicate that resistance distance and LCP cost produce the most accurate connectivity estimates, although their absolute performance under different conditions is difficult to predict. We emphasise the importance of testing connectivity estimates against patterns derived from independent data, such as those acquired from tracking studies. Our findings should help to inform a more refined implementation of connectivity metrics in conservation management.

Landscape connectivity estimates are widely used to inform species conservation management. However, although many landscapes and species behaviours change over time, such as between seasons, the vast majority of studies view connectivity as static. Calls have therefore been made to use multiple connectivity estimates to adequately capture periodic (e.g., seasonal) temporal changes. These periodic temporal changes are also influenced by stochastic perturbations, such as extreme weather events, and this variability is likely to increase due to ongoing climate change. We aimed to investigate: 1) how variability interacts with periodic landscape changes to alter connectivity estimates, and 2) how these alterations change over longer time-periods. To achieve this we used a virtual ecology approach in which we simulated 100 landscapes. Each simulation ran for 20 seasons with the landscape experiencing regular seasonal changes. Each simulation was then rerun five times with increasing levels of variability super-imposed on the seasonal dynamics. Connectivity for each landscape and every season was calculated using least-cost paths modelling, and the differences between connectivity estimates calculated. We found that the variation in connectivity estimates between seasons increased with temporal variability. Differences in connectivity accumulated over time, meaning that as the variability increased connectivity estimates changed more quickly and by larger amounts. Our study demonstrates that increasing temporal variability will become a challenge for the successful use of static connectivity estimates.

There is a growing use of bottom-up simulation models to reconstruct past human-environment interactions. Such detailed representations pose difficult questions not only in their design (the generality-realism trade-off) but also about the inferences that are made from them. The historical sciences are faced with seeking to make robust inferences from limited, potentially biased and/or incomplete samples from uncontrolled systems, and as a result have sometimes employed narrative explanation. By contrast, simulation models can be used experimentally and can generate large amounts of data. Here, using an agent-based model of hunter-gatherer foraging in a previously unexplored ecosystem, we consider how narratives might be identified from the trajectories produced by simulations. We show how machine learning methods can isolate qualitatively similar types of model behaviour based on summaries of model outcomes and time series. We stand to learn from this approach because it enables us to answer two questions: (i) under what conditions (representations and/or parameterisations) do we observe in the model what is recorded in the archaeological and/or palaeoenvironmental record? and (ii) does the model yield unobserved dynamics? If so, are they plausible? Using models to develop narratives is a logical extension of the bottom-up approach inherent in agent-based modelling and has the potential, alongside conventional methods of model evaluation, to aid in learning from the rich dynamics of such simulations.

Invasive species can cause shifts in vegetation composition and fire regimes by initiating positive vegetation-fire feedbacks. To understand the mechanisms underpinning these shifts, we need to determine how invasive species interact with other species when burned in combination and thus how they may influence net flammability in the communities they invade. Previous studies using litter and ground fuels suggest that flammability of a species mixture is nonadditive and is driven largely by the more-flammable species. However, this nonadditivity has not been investigated in the context of plant invasions nor for canopy fuels. Using whole shoots, we measured the flammability of indigenous-invasive species pairs for six New Zealand indigenous and four globally invasive plant species, along with single-species control burns. Our integrated measure of flammability was clearly nonadditive, and the more-flammable species per pairing had the stronger influence on flammability in 83% of combinations. The degree of nonadditivity was significantly positively correlated with the flammability difference between the species in a pairing. The strength of nonadditivity differed among individual flammability components. Ignitability and combustibility were strongly determined by the more-flammable species per pair, yet both species contributed more equally to consumability and sustainability. Our results suggest mechanisms by which invasive species entrain positive vegetation-fire feedbacks that alter ecosystem flammability, enhancing their invasion. Of the species tested, Hakea sericea and Ulex europaeus are those most likely to increase the flammability of New Zealand ecosystems and should be priorities for management.

Fire regimes are powerful selective filters. In New Zealand, fire activity was rare before human settlement, and New Zealand’s indigenous woody flora shows little adaptation to frequent fire. One of the few woody indigenous species to show adaptation to frequent fire is Leptospermum scoparium (Myrtaceae), a Plio-Pleistocene immigrant from Australia. Leptospermum scoparium is a widespread early-successional shrub to small tree that frequently dominates post-fire successions, and shows geographically-variable pyriscent serotiny, a fire-adaptation in which seeds are retained in the canopy and open post-fire. Another widespread reproductive adaptation to fire is heat-stimulated germination. Some Australian Leptospermum respond positively to heat treatment, and observations in New Zealand suggest that the seeds of L. scoparium within thick-walled capsules open immediately after fire and successfully germinate, indicating at least no deleterious effect of heating. In some fire-prone systems such as Medi...

Ecologists fiercely debate the role of soil conditions and fire regimes in controlling forest – savanna boundaries. A prominent component of this debate centres on the plausibility and existence of fire-mediated alternative stable state dynamics (FMASS), a model first proposed by the Tasmanian ecologist WD Jackson in 1968. The FMASS model asserts that increased or decreased landscape fire activity, often due to human intervention, can overwhelm physical environmental (e.g. topography and edaphic factors) controls of forest and savanna mosaics, thereby creating landscape dynamism. Many FMASS models include fire-vegetation-soil interactions (FVS), in which changes in fire frequency can change soil fertility and hence tree growth. This interaction, in turn, affects the capacity offorests to recover from disturbance and long unburnt savanna to convert to forest. We evaluate support for the FMASS-FVS model in the context of the dynamics of the Tasmanian forests that have recently been drawn into wider, global debates surrounding the co-occurrence of tree and treeless vegetation states. We develop a simple spatial simulation model to illuminate the difficulties in analysing landscape pattern to draw inferences about the existence of FMASS-FVS. Our review of the Tasmanian evidence shows that FMASS-FVS cannot unambiguously explain all tracts of sedgelands in Tasmanian wet forests, and hence Tasmania should not be used as an exemplar of these theories globally. Our simulations highlight that soil sampling that targets forest boundaries risks erroneously concluding that the distribution of forest and savanna boundaries is decoupled from edaphic factors. We describe a structured methodological pathway that can identify the role of FMASS-FVS in Tasmanian forest dynamics, and elsewhere in the world. This approach use cross-scale temporal and spatial analyses, and targeted experimental tests, to illuminate the interplay of fire, vegetation and edaphic factors in controlling tree establishment and growth and forest boundary dynamics.

Individual-based models (IBMs) are increasingly used to explore ecological systems and, in particular, the emergent outcomes of individual-level processes. A major challenge in developing IBMs to investigate the movement ecology of animals is that such models must represent and parameterise unobserved behaviours occurring at multiple hierarchical levels. Approaches based on approximate Bayesian computation (ABC) methods have been used to support the parameterisation, calibration and evaluation of IBMs. However, a key component of the ABC approach is the use of multiple quantitative patterns derived from empirical data to exclude model structures and parameterisations that generate atypical or implausible patterns. We propose a modelling framework that integrates information derived from statistical inferential models, which are now widely used to describe the behaviour of moving animals, with ABC methodologies for the parameterisation and analysis of IBMs. To demonstrate its application, we apply this framework to high-resolution movement trajectories of the foraging trips of black petrels ( Procellaria parkinsoni ), an endangered seabird endemic to New Zealand. The outcomes of our study show that the use of inferential statistical models to summarise movement data can aid model selection and parameterisation procedures via ABC, and yield valuable insights into the modelling in movement ecology of animals.

Deterministic and stochastic factors interact to generate biogeographic patterns in fungal communities, challenging efforts to predict which fungal assemblages will develop in association with introduced plants. The coastal dune grass Ammophila arenaria has been moved around the world. We sampled A. arenaria roots in its native range in the United Kingdom, and in Australia and New Zealand where it is invasive. The co-occurring native dune grass Leymus arenarius was also sampled in the United Kingdom, and the native dune grass Spinifex sericeus was sampled in Australia and New Zealand. A. arenaria associates with root fungal communities in its invasive range that are richer than those in the United Kingdom, and richer than those in co-occurring native grasses, demonstrating that exotic plants are not necessarily depauperate in fungal associates. Most of the dominant OTUs associating with A. arenaria were shared among all three geographic regions. However, community composition varied among regions, with differences in temperature, nitrogen and pH correlating with community change. Within regions, community composition varied among dunes, although the majority of sequences obtained were in OTUs that were detected in multiple dunes. Host plant species identity influenced fungal community composition, although the majority of the data were in OTUs that were shared between hosts. While the roots sampled appeared healthy, many of the dominant fungi recovered are potentially pathogenic. This study demonstrates that many fungi are widely dispersed, and that introduced plants are likely to associate with functionally diverse fungal communities that include species from their native ranges. However root-associated fungal community composition is variable, even at regional and local scales in a common plant in a common environment.

Understanding the interactive effects of fragmentation and invasive species on forest dynamics requires a long-term perspective because they are difficult to assess in the medium- to long-term using observational or experimental data alone. In such settings ecological models have an important role to play. Here we describe the implementation of a spatially explicit individual-based model (SEIBM) representing the dynamics of small forest fragments in northern New Zealand based on empirical data collected in the region. In addition, we performed a baseline analysis to determine how well the model captured podocarp-tawa forest dynamics, and compared its performance with stand structure data obtained from an unfragmented forest in northern New Zealand. We used sensitivity analysis to determine how sensitive the model was to changes in the input parameters. In addition, we simulated different scenarios under diverse management conditions to explore the model’s potential as a management tool. The model captures the stand structural characteristics of the fragments reasonably well but under-predicts stand basal area, suggesting that it does not represent the long-term suppression of some canopy tree species adequately. Although some refinement is needed to improve its performance, we believe that the model presented here is a useful tool for management purposes and for the assessment of the long term viability of forest fragments. The model can help inform managers and decision-makers regarding the long-term persistence of podocarp-tawa forest patches.

Estimates of landscape connectivity are routinely used to inform decision-making by conservation biologists. Most estimates of connectivity rely on cost-surfaces: raster representations of landscapes in which cost values represent the difficulty involved with traversing an area. However, there is considerable uncertainty in the generation of cost-surfaces that have not been widely explored. We investigated the effects of four potential sources of uncertainty in the creation of cost-surfaces: 1) number of landscape classes represented; 2) spatial resolution (grain size); 3) misclassification of edges between landscape classes; and 4) cost values selected for each landscape class. Following a factorial design we simulated multiple cost-surface pairs, each comprising one true surface with no errors and one surface with uncertainty comprised of some combination of the four error sources. We evaluated the relative importance of each source of uncertainty in determining the difference between the least-cost paths (LCPs) costs and resistance distances generated for the true and erroneous cost-surfaces, using four model evaluation metrics. Errors in the underlying geospatial layers produced larger inaccuracies in connectivity estimates than those produced by cost-value errors. Incorrect grain size had the largest overall effect on the accuracy of connectivity estimates. Though the removal of an element class was found to have a large effect on the configuration of connectivity estimates, and the addition of an element class had a large effect on estimates configuration. Our results highlight the importance of minimising and quantifying the uncertainty inherent in the geospatial data used to develop cost-surfaces.

Lightning is a key component of the Earth's atmosphere and climate systems, and there is a potential positive feedback between a warming climate and increased lightning activity. In the biosphere, lightning is important as the main natural ignition source for wildfires and because of its contribution to the nitrogen cycle. Therefore, it is important to develop lightning climatologies to characterise and monitor lightning activity. While traditional methods for constructing lightning climatologies are suitable for examining lightning's influence on atmospheric processes, they are less well suited for examining questions about biosphere-lightning interactions. For example, examining the interaction between lightning and wildfires requires linking atmospheric processes to finer scale terrestrial processes and patterns. Most wildfires ignited by lightning are less than one hectare in size, and so require lightning climatologies at a comparable spatial resolution. However, such high resolution lightning climatologies cannot be derived using the traditional cell-count methodology. Here we present a novel geocomputational approach for analysing lightning data at high spatial resolutions. Our approach is based on probabilistic computational methods and is capable of producing a sub-kilometre lightning climatology that honours the spatial accuracy of the strike locations and is adaptive to underlying spatial patterns. We demonstrate our methods by applying them to the mid-latitude oceanic landmass of New Zealand, an area with geographic conditions that are under-represented in existing lightning climatologies. Our resulting lightning climatology has unparalleled spatial resolution, and the spatial and temporal patterns we observe in it are consistent with other continental and tropical lightning climatologies. To encourage further use and development of our probabilistic approach, we provide Python scripts that demonstrate the method alongside our resulting New Zealand lightning climatology. A sub-kilometre resolution lightning climatology is produced.Spatially adaptive probabilistic geocomputation methods are developed.Geocomputation workflow provided using an open-source Python approach.New Zealand has exceptionally low levels of lightning activity.

Summary 1.Habitat connectivity is a crucial determinant of population dynamics in fragmented landscapes. The corollary of the emphasis on maintaining connectivity to enhance the movement of organisms is that disrupting connectivity should minimise it. Here we evaluate the efficiency of an invasive species control strategy that targets the most connected habitats in a landscape. 2.A network (spatial graph) provides an intuitive representation of a landscape, and the topology of this network can be used to identify the most connected patches. We implemented a simulation model of the spread of an invasive species on a network and used it to evaluate whether targeting the better-connected components of the landscape enhances control effectiveness. 3.Control strategies based on network topology consistently outperformed both a null strategy of random habitat selection and one based on separation distance alone. The advantages of the connectivity-based strategy were strongest in the early phases of the invasion process, when a small number of habitats are occupied at low population density. However, if long distance dispersal events were common the advantages of the connectivity approach weakened. 4.The performance of the connectivity-based strategy is robust to habitat-level demographic stochasticity. In fact, connectivity-based targeting outperforms a strategy focussing on source habitats, with the additional benefit that it requires less information to be implemented. 5.Synthesis and applications. Our simulation model outcomes demonstrate that deliberately targeting the best-connected components of a landscape is an efficient control strategy for invasive species when long-distance dispersal is infrequent, and it is likely to be cheaper than other alternatives such as targeting population sources. Network scientists have developed a range of methods designed to identify the minimal set of nodes on a graph that will disrupt the network as a whole; these tools have potential to aid in the design of more effective control strategies. This article is protected by copyright. All rights reserved.

In some ecosystems a small suite of species can determine community-level patterns of species richness by acting as either ‘accumulators’ or ‘repellers’; that is, the richness of the immediate neighbourhood of such species departs from that expected on the basis of a given null model. Using the individual species-area approach, we evaluated community-level spatial pattern in four species-rich shrublands (two 40 × 40 m and two 30 × 30 m plots) by assessing the frequency of accumulator and repeller species and whether any such species were associated with specific life-history characteristics. Few species departed from the expectations of the null spatial model that we used, although, at three of the four sites, accumulators were more common than repellers. Departures from the null model we assessed were most prevalent within just 1 m of focal individuals and were not consistently associated with specific life-history traits. Model-based clustering suggests that there are distinct sub-communities in each of the four communities, but while internally spatially cohesive, these sub-communities intermingle and their membership is not predictable from the life-history traits of their constituent species. Comparable analyses in other species-rich systems have also detected a similar absence of spatial interactions. The disturbance regime in the shrubland communities we consider is markedly different from those in the forest ecosystems where previous studies have been concentrated (recurrent fire versus infrequent gap-phase dynamics) and resources are more limiting. Thus, our results provide further support for the generality of a model of stochastic geometry, likely underpinned by stochastic dilution effects, in species-rich plant communities.

Niche and neutral processes jointly influence species interactions. Predictions of interactions based on these processes assume that they operate similarly across all species. However, species characteristics could systematically create differences in the strength of niche or neutral processes for each interspecific interaction. We used national-level records of plant–frugivore interactions, species traits, biogeographic status (native vs. exotic), phylogenies and species range sizes to test the hypothesis that the strength of niche processes in species interactions changes in predictable ways depending on trophic generalism and biogeographic status of the interacting species. The strength of niche processes (measured as trait matching) decreased when the generalism of the interacting partners increased. Furthermore, the slope of this negative relationship between trait matching and generalism of the interacting partners was steeper (more negative) for interactions between exotic species than those between native species. These results remained significant after accounting for the potential effects of neutral processes (estimated by species range size). These observed changes in the strength of niche processes in generating species interactions, after accounting for effects of neutral processes, could improve predictions of ecological networks from species trait data. Specifically, due to their shorter co-evolutionary history, exotic species tend to interact with native species even when lower trait matching occurs than in interactions among native species. Likewise, interactions between generalist bird species and generalist plant species should be expected to occur despite low trait matching between species, whereas interactions between specialist species involve higher trait matching. Fil: Peralta, Guadalupe. University of Canterbury; Nueva Zelanda Fil: Perry, George L.W.. University of Auckland; Nueva Zelanda Fil: Vazquez, Diego P.. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Mendoza. Instituto Argentino de Investigaciones de las Zonas Áridas. Provincia de Mendoza. Instituto Argentino de Investigaciones de las Zonas Áridas. Universidad Nacional de Cuyo. Instituto Argentino de Investigaciones de las Zonas Áridas; Argentina Fil: Dehling, D. Matthias. University of Canterbury; Nueva Zelanda Fil: Tylianakis, Jason M.. University of Canterbury; Nueva Zelanda

New, more socially-acceptable technologies are being developed to suppress horticultural pests, because suppression is technically difficult with current technologies, especially in urban areas. One technique involves the release of sterile insects to prevent offspring in the next generation. This technology involves aerial or ground release systems, but this could also create issues for the public. This study investigated community perceptions of a recently-introduced response to codling moth control in New Zealand&mdash, Sterile Insect Technique (SIT). Community attitudes to SIT were examined in Hastings, New Zealand, in April, 2018. Eighty-six detailed interviews were undertaken with a random sample of households. This community was very willing (98% agreement) to host a sex pheromone trap in their gardens, and condoned regular visits to monitor traps. Attitudes to SIT were very positive (98% in favor). Once explained, the concept of using unmanned aerial vehicles to deliver sterile insects was also acceptable (98%) to the community. Use of unmanned aerial vehicles to release sterile insects during a hypothetical incursion response of an exotic fruit fly was also supported at 98% by respondent householders. Investigation of community attitudes can be valuable to guide practitioners in determining suitable technologies before an area-wide programme is launched.

Terrestrial plants and fire have interacted for at least 420 million years

Introduction Blood-based biomarkers of pathophysiological brain amyloid β (Aβ) accumulation, particularly for preclinical target and large-scale interventions, are warranted to effectively enrich Alzheimer's disease clinical trials and management. Methods We investigated whether plasma concentrations of the Aβ 1–40 /Aβ 1–42 ratio, assessed using the single-molecule array (Simoa) immunoassay, may predict brain Aβ positron emission tomography status in a large-scale longitudinal monocentric cohort (N = 276) of older individuals with subjective memory complaints. We performed a hypothesis-driven investigation followed by a no-a-priori hypothesis study using machine learning. Results The receiver operating characteristic curve and machine learning showed a balanced accuracy of 76.5% and 81%, respectively, for the plasma Aβ 1–40 /Aβ 1–42 ratio. The accuracy is not affected by the apolipoprotein E ( APOE ) e4 allele, sex, or age. Discussion Our results encourage an independent validation cohort study to confirm the indication that the plasma Aβ 1–40 /Aβ 1–42 ratio, assessed via Simoa, may improve future standard of care and clinical trial design.

Forest community assembly is usually framed in terms of sporophyte dynamics; however, the recruitment and maintenance of fern populations, frequently influential in forest composition and structure, are initially determined by gametophytes. Sporophytes of three Cyathea tree fern species show habitat partitioning along gradients of phosphorus and light; we asked whether gametophyte niche differences parallel this pattern. To compare niche characteristics among taxa we compared growth rates to a size threshold (≥3 mm) of gametophytes under controlled conditions using a multi-factorial, multi-level (3 × 4) experiment, varying irradiance (5.4 ± 4.4; 59.1 ± 44.3; 107 ± 74.1 µmol m −2 s −1 ) and orthophosphate concentrations (5, 10, 20, 40 mg kg −1 ). Gametophytes of the pioneer species C. medullaris developed to the size threshold across a broad range of phosphate and irradiance treatments (more than 20% of gametophytes in ≥ 7 of the 12 treatments), peaking at 20 mg kg −1 P and 60 µmol m −2 s −1 irradiance. The growth rates of the forest understorey species C. dealbata and C. smithii also peaked at 60 µmol m −2 s −1 but varied across treatments, suggesting niche differentiation along irradiance and orthophosphate gradients. Our analysis suggests that gametophyte development is strategically aligned to the ecological habits of sporophytes and that forest community assembly is likely strongly influenced by the independent gametophyte life-stage.

Dispersal is fundamental to population dynamics and hence extinction risk. The dispersal success of animals depends on the biophysical structure of their environments and their biological traits; however, comparatively little is known about how evolutionary trade-offs among suites of biological traits affect dispersal potential. We developed a spatially explicit agent-based simulation model to evaluate the influence of trade-offs among a suite of biological traits on the dispersal success of vagile animals in fragmented landscapes. We specifically chose traits known to influence dispersal success: speed of movement, perceptual range, risk of predation, need to forage during dispersal, and amount of suitable habitat required for successful settlement in a patch. Using the metric of relative dispersal success rate, we assessed how the costs and benefits of evolutionary investment in these biological traits varied with landscape structure. In heterogeneous environments with low habitat availability and scattered habitat patches, individuals with more equal allocation across the trait spectrum dispersed most successfully. Our analyses suggest that the dispersal success of animals in heterogeneous environments is highly dependent on hierarchical interactions between trait trade-offs and the geometric configurations of the habitat patches in the landscapes through which they disperse. In an applied sense, our results indicate potential for ecological mis-alignment between species' evolved suites of dispersal-related traits and altered environmental conditions as a result of rapid global change. In many cases identifying the processes that shape patterns of animal dispersal, and the consequences of abiotic changes for these processes, will require consideration of complex relationships among a range of organism-specific and environmental factors.

Fossilized gut contents suggest that seeds consumed by dinosaurs may have remained intact in their stomachs, and since seed dispersal distance increases with body-mass in extant vertebrates, dinosaurs may have moved seeds long distances. I simulated seed dispersal by dinosaurs across body-masses from 1 × 10 1 to 8 × 10 4 kg using allometric random walk models, informed by relationships between (i) body-mass and movement speed, and (ii) body-mass and seed retention time. Seed dispersal distances showed a hump-shaped relationship with body-mass, reflecting the allometric relationship between maximum movement speed and body-mass. Across a range of assumptions and parameterizations, the simulations suggest that plant-eating dinosaurs could have dispersed seeds long distances.

The non-random selection of fruits by frugivores can affect plant demography by changing the probability of seedling recruitment. We investigated whether the New Zealand pigeon (Hemiphaga novaeseelandiae) exhibits size-based preference of fruit in two indigenous tree species, Vitex lucens and Prumnopitys ferruginea, that have large fleshy diaspores containing single seeds. We also assessed if there was an advantage for seeds that were actively dispersed since birds may select fruits that have viable seeds (although the cue for such selection remains unclear) and whether a combined scarification–deinhibition advantage was conferred on germination due to seed passage through the gut. Using seed length as a proxy for fruit size, we compared the length of passively dispersed fruits in traps with the estimated fruit length of actively dispersed seeds in six trees of each species across three areas in one fruiting season. For both species, actively dispersed seeds were significantly longer than seeds fr...

Summary 1.A key rationale for pursuing de-extinction is the potential to restore lost processes and function to modern ecosystems. However, understanding and providing for the ecological requirements of the candidate species will also play a key role in determining the ultimate success of a de-extinction. This assessment is challenging for prehistoric extinct species, where empirical studies or observations on their ecology are not available or are incomplete. 2.A healthy, stable and self-sustaining population of a resurrected species needs to be embedded in an ecological interaction network that consists not only of interactions between a resurrected species and its external environment (e.g. habitat, diet), but also those where the resurrected species is the environment (e.g. microbiota and parasites). Palaeoecology can provide information on all of these interactions for extinct species, and this information can help guide and inform the selection of suitable de-extinction candidates or traits for resurrection. 3.Ecological interaction network analyses offer a complementary tool to existing frameworks for determining the suitability of de-extinction candidates, allowing palaeoecological information to be used to identify and quantify the potential implications of removing or adding resurrected species/functions to an ecosystem. 4.Although palaeoecological data and understanding clearly have an important role in informing and modelling the potential ecological functions provided by extinct species, they can only ever provide an incomplete picture, and therefore would only complement, rather than replace, observational or experimental data on the resurrected organisms themselves. This article is protected by copyright. All rights reserved.

Seed dispersal by avian frugivores is one of the key processes influencing plant spatial patterns, but may fail if there is disruption of plant-frugivore mutualisms, such as decline in abundance of dispersers, fragmentation of habitat, or isolation of individual trees. We used simulation model experiments to examine the interaction between frugivore density and behaviour and the spatial arrangement of fruiting plants and its effect on seed dispersal kernels. We focussed on two New Zealand canopy tree species that produce large fruits and are dispersed predominantly by one avian frugivore (Hemiphaga novaeseelandiae). Although the mean seed dispersal distance decreased when trees became more aggregated, there were more frugivore flights between tree clusters, consequently stretching the tails of the dispersal kernels. Conversely, when trees were less aggregated in the landscape, mean dispersal distances increased because seeds were deposited over larger areas, but the kernels had shorter tails. While there were no statistically meaningful changes in kernel parameters when frugivore density changed, decreases in density did cause a proportional reduction in the total number of dispersed seeds. However, birds were forced to move further when fruit availability and fruit ripening were low. Sensitivity analysis showed that dispersal kernels were primarily influenced by the model parameters relating to disperser behaviour, especially those determining attractiveness based on distance to candidate fruiting trees. Our results suggest that the spatial arrangement of plants plays an important role in seed dispersal processes – although tree aggregation curbed the mean seed dispersal distance, it was accompanied by occasional long distance events, and tree dispersion caused an increase in mean dispersal distance, both potentially increasing the probability of seeds finding suitable habitats for germination and growth. Even though low frugivore densities did not cause dispersal failure, there were negative effects on the quantity of seed dispersal because fewer seeds were dispersed. This article is protected by copyright. All rights reserved.

Tree ferns are a ubiquitous and often locally dominant element of wet southern temperate rainforests across Australasia, southern Africa and in regions adjacent to the tropics in South America. Published data on the ecology of tree ferns throughout these forest ecosystems is piecemeal, with the most comprehensive literature describing the ecology of tree ferns coming from New Zealand. Therefore using New Zealand forests as a model system, we review the ecology and importance of tree ferns for forest structure and composition. Most studies of the ecology and function of forest species in New Zealand have focussed on spermatophytes. Even though tree ferns (Cyatheaceae, Dicksoniaceae) can represent more than 50% of basal area and more than 20% of forest biomass they have been largely overlooked and quantitative information on their contribution to forest structure and function is relatively scarce. Here for the first time we synthesise information on NZ’s indigenous tree ferns published over the last 100 years and present new data on their ecology and potential ecosystem influences. Irradiance and nutrient (N, P) levels constrain development of tree fern gametophytes, with P limitation potentially influencing sporophyte production. Tree ferns establish during temporary removal of forest canopies across various spatio-temporal scales including after local disturbances, wind-throw openings of the canopy, and landslides. Members of the Cyatheaceae exhibit spatial differentiation along temperature and solar radiation gradients; the Dicksoniaceae species differentiate most strongly on their frost tolerance. Frequency/abundance of all nine understory tree fern species indigenous to New Zealand increases with total soil phosphorus, with some niche differentiation among species along a broader nutrient gradient. Tree ferns are prominent in early and mid-successional forest communities where they may persist for more than 250 years. Individually, and as a group, tree-ferns impact nutrient cycling, organic matter accumulation and ground-level irradiance, often shading out tree seedlings. Tree ferns also have long-term physical impacts on the regeneration niche of associated species, with epiphytism on tree fern trunks providing alternative establishment surfaces for many species. Domination of nutrient resources by tree ferns early in succession is likely a key factor influencing community establishment. We conclude with key recommendations for future research on fundamental unknown elements of the ecology and synecology of tree ferns including greater determination of their ecophysiology and influence on forest community assemblages.

Fencing forest fragments to limit the effects of herbivores is a common method for restoration, but recent studies have suggested that fencing alone is insufficient for fragment recovery. The endemic New Zealand canopy tree tawa (Beilschmiedia tawa (A.Cunn.) Kirk) is suffering recruitment failure in forest fragments throughout its northern range. We sought to determine the role of fragmentation and fencing on B. tawa seedling survival and to identify the factors driving recruitment failure in this species. The main factors that affected B. tawa seedling survivorship were desiccation and herbivory. Fragmentation had a negative effect on the survivorship and growth of potted B. tawa seedlings. Fencing neither influenced survivorship nor diminished mortality by desiccation. However, small roofed enclosures positively affected seedling survivorship, potentially by diminishing desiccation. Our results are consistent with other studies where desiccation has been deemed a key factor of tree mortality in fragmented forests. As a single management strategy, fencing appears to be insufficient to positively affect B. tawa regeneration and other complementary activities are suggested.

Ecological memory is central to how ecosystems respond to disturbance and is maintained by two types of legacies – information and material. Species life-history traits represent an adaptive response to disturbance and are an information legacy; in contrast, the abiotic and biotic structures (such as seeds or nutrients) produced by single disturbance events are material legacies. Disturbance characteristics that support or maintain these legacies enhance ecological resilience and maintain a “safe operating space” for ecosystem recovery. However, legacies can be lost or diminished as disturbance regimes and environmental conditions change, generating a “resilience debt” that manifests only after the system is disturbed. Strong effects of ecological memory on post-disturbance dynamics imply that contingencies (effects that cannot be predicted with certainty) of individual disturbances, interactions among disturbances, and climate variability combine to affect ecosystem resilience. We illustrate these concepts and introduce a novel ecosystem resilience framework with examples of forest disturbances, primarily from North America. Identifying legacies that support resilience in a particular ecosystem can help scientists and resource managers anticipate when disturbances may trigger abrupt shifts in forest ecosystems, and when forests are likely to be resilient.

Altered fire regimes in the face of climatic and land-use change could potentially transform large areas from forest to shorter-statured or open-canopy vegetation. There is growing concern that once initiated, these nonforested landscapes could be perpetuated almost indefinitely through a suite of positive feedbacks with fire. The rapid deforestation of much of New Zealand following human settlement (ca. 750 years ago) provides a rare opportunity to evaluate the feedback mechanisms that facilitated such extensive transformation and thereby help us to identify factors that confer vulnerability or resilience to similar changes in other regions. Here we evaluate the structure of living and dead vegetation (fuel loading) and microclimate (fuel moisture) in beech (Nothofagaceae) forests and adjacent stands that burned within the last 60–140 years and are dominated by mānuka (Leptospermum scoparium) or kānuka (Kunzea spp.). We show that the burning of beech forests initiates a positive feedback cycle whereby the loss of microclimatic amelioration under the dense forest canopy and the abundant fine fuels that dry readily beneath the sparse mānuka/kānuka canopy enables perpetuation of these stands by facilitating repeated burning. Beech regeneration was limited to a narrow zone along the margin of unburned stands. The high flammability of vegetation that develops after fire and the long time to forest recovery were the primary factors that facilitated extensive deforestation with the introduction of human-ignited fire. Evaluating these two characteristics may be key to determining which regions may be near a tipping point where relatively small land-use- or climatically driven changes to fire regimes could bring about extensive deforestation.

Transformations are underway in our ability to collect and interrogate remotely sensed data. Here we explore the utility of three machine-learning methods for identifying the controls on coastal cliff landsliding using a dataset from Auckland, New Zealand. Models were built using all available data with a resampling approach used to evaluate uncertainties. All methods identify two dominant landslide predictors (unfailed cliff slope angle and fault proximity). This information could support a range of management approaches, from the development of 'rules-of-thumb' to detailed models that incorporate all predictor information. In our study all statistical approaches correctly predict a high proportion (>85%) of cases. Similar 'success' has been shown in other studies, but important questions should be asked about possible error sources, particularly in regard to absence data. In coastal landslide studies sign decay is a vexing issue, because sites prone to landsliding may also be sites of rapid evidence removal. Machine-learning based approaches successfully identify coastal cliff landsliding controls.Possible management approaches vary from hazard mapping to heuristic tools.Difficult issues surround the nature of absence data.Sign decay may be a prevalent issue in earth-science applications.

Codling moth, Cydia pomonella (Lepidoptera: Tortricidae), is a phytosanitary pest of New Zealand&rsquo, s export apples. The sterile insect technique supplements other controls in an eradication attempt at an isolated group of orchards in Hawke&rsquo, s Bay, New Zealand. There has been no attempt in New Zealand to characterize potential sources of uncontrolled peri-urban populations, which we predicted to be larger than in managed orchards. We installed 200 pheromone traps across Hastings city, which averaged 0.32 moths/trap/week. We also mapped host trees around the pilot eradication orchards and installed 28 traps in rural Ongaonga, which averaged 0.59 moths/trap/week. In Hastings, traps in host trees caught significantly more males than traps in non-host trees, and spatial interpolation showed evidence of spatial clustering. Traps in orchards operating the most stringent codling moth management averaged half the catch rate of Hastings peri-urban traps. Orchards with less rigorous moth control had a 5-fold higher trap catch rate. We conclude that peri-urban populations are significant and ubiquitous, and that special measures to reduce pest prevalence are needed to achieve area-wide suppression and reduce the risk of immigration into export orchards. Because the location of all host trees in Hastings is not known, it could be more cost-effectively assumed that hosts are ubiquitous across the city and the area treated accordingly.

Human activity has resulted in the emergence of fire-induced forest-shrubland mosaics in many southern temperate forest landscapes. In New Zealand (NZ), this transformation was caused by the introduction of fire to previously very low fire-frequency ecosystems. High levels of seed and seedling predation by exotic mammals and the reduction of avian pollination and seed dispersal further influence vegetation change and can cause successions to become completely arrested. Analogous dynamics occur in many other Southern Hemisphere forest landscapes where changes in fire regimes and the introduction of mammalian herbivores and predators have resulted in forest-shrubland mosaics. We developed a spatially explicit state-and-transition model to disentangle the effects of altered fire regimes; invasive pyrophyllic shrub species; seed and seedling predation by exotic mammals; and lowered rates of long-distance seed movement on long-term landscape dynamics and state transitions. Our modelling highlights a threshold in fire frequency where forest systems move to a fire-induced early-successional shrubland state. Because early-successional vegetation in NZ and many other southern temperate forests is much more flammable than mature late-successional vegetation, this shrubland state will persist until fire is suppressed. Our modelling suggests also that in the absence of fire there are critical thresholds in seed predation beyond which succession may become completely arrested in a mid-successional state. Even moderate levels of predation slow regeneration and move the tipping point at which forests shift to early-successional shrublands in the direction of less frequent fire. The presence of invasive pyrophyllic shrubs in the landscape also influences the location of thresholds in fire frequency at which state shifts occur. Restoring these landscapes to their formerly forested condition will require the removal of fire as well as mammalian seed/ling predators and the reinstatement of pollination and dispersal services. Removal of seed predators may act as an effective component of fire control by allowing secondary successions to proceed more quickly, thereby reducing landscape-level flammability.

Predation by introduced species and anthropogenic climate change now are two major threats to avifaunae worldwide. We developed an age-structured, stochastic simulation model to perform a population viability analysis for the Blue Duck, or Whio (Hymenolaimus malacorhynchos), an endangered waterfowl endemic to the upland rivers of New Zealand. Demographic responses to six predator-control and flood-frequency scenarios were assessed to examine projected trends in long-term population dynamics and viability. A baseline analysis of abundance at current rates of predation by introduced mammals and mortality due to severe flood events indicated that the study population of Whio has only a limited probability of persistence during the next 100 years. Importantly, we found that the additive effects of population losses due to predation and a rising frequency of floods (as predicted by general circulation models) increased extinction risk by 69%. Our study demonstrates the urgent need for heightened control of invasive predators in Whio conservation areas, as well as informed consideration of future changes in regional climate patterns when planning management actions for Whio and other similarly threatened riparian species.

Prehistoric faunal extinctions dramatically reshaped biological assemblages around the world. However, the timing of such biotic shifts is often obscured by the fragmentary nature and limited temporal resolution of fossil records. We use radiocarbon-dating and ancient-DNA analysis of prehistoric (ca A.D. 1450–1834) Megadyptes penguin specimens to assess the time-frame of biological turnover in coastal New Zealand following human settlement. These data suggest that the final extirpation of the endemic Megadyptes waitaha, and subsequent replacement by the previously sub-Antarctic-limited Megadyptes antipodes, likely occurred within a narrow temporal window (e.g. a century or less). This transition represents one of the most rapid prehistoric faunal turnover events documented, and is likely linked to human demographic and cultural transitions during the 15th Century. Our results suggest that anthropogenic forces can trigger rapid biogeographic shifts.

The relative importance of people and climate in shaping prehistoric fire regimes is debated around the world, and this discussion has helped inform our understanding of past and present ecosystem dynamics. Evidence for extensive anthropogenic burning of temperate closed-canopy forests prior to European settlement is geographically variable, and the factors responsible for this variability are not well resolved. We set out to explain the differences in the influence of prehistoric human-set fires in seasonally dry forest types in the Pacific Northwest, New Zealand, and northern Patagonia by comparing the fire traits of dominant taxa, postfire vegetation recovery, long-term climate trends, and human activities that may have motivated burning. Our analysis suggests that ecological and climatic factors explain much of the differences in how these mesic–dry forests responded to prehistoric anthropogenic burning. Understanding past human–environment interactions at regional scales is an important step for assessing the impact of biomass burning at all scales.

The New Zealand conifers (20 species of trees and shrubs in the Araucariaceae, Podocarpaceae and Cupressaceae) are often regarded as ancient Gondwanan elements, but mostly originated much later. Often thought of as tall trees of humid, warm forests, they are present throughout in alpine shrublands, tree lines, bogs, swamps, and in dry, frost-prone regions. The tall conifers rarely form purely coniferous forest and mostly occur as an emergent stratum above evergreen angiosperm trees. During Maori settlement in the 13th century, these fire-sensitive trees succumbed rapidly, most of the drier forests being lost. As these were also the more conifer-rich forests, ecological research has been skewed towards conifer dynamics of forests wetter and cooler than the pre-human norm. Conifers are well represented in the pollen record and we here we review their late Quaternary history in the light of what is known about their current ecology with the intention of countering this bias. During glacial episodes, all trees were scarce south of c. 40°S, and extensive conifer-dominant forest was confined to the northern third of the North Island. Drought- and cold-resistant Halocarpus bidwillii and Phyllocladus alpinus formed widespread scrub in the south. During the deglacial, beginning 18,000 years ago, tall conifers underwent explosive spread to dominate the forest biomass throughout. Conifer dominance lessened in favour of angiosperms in the wetter western lowland forests over the Holocene but the dryland eastern forests persisted largely unchanged until settlement. Mid to late Holocene climate change favoured the more rapidly growing Nothofagaceae which replaced the previous conifer-angiosperm low forest or shrubland in tree line ecotones and montane areas. The key to this dynamic conifer history appears to be their bimodal ability to withstand stress, and dominate on poor soils and in cool, dry regions but, in wetter, warmer locations, to slowly grow thorough competing broadleaves to occupy an exposed, emergent stratum where their inherent stress resistance ensures little effective angiosperm competition.

Invasive species have become a major stressor in many ecosystems. Therefore, public-policy decision-makers desire ecologically informed risk assessments that characterise the likelihood and severity of potential adverse effects from invasive species. These risk assessments often take the form of maps, but risk maps have generally ignored uncertainty and avoided incorporating processes such as the risk of spread. One method that has been proposed to map risk of spread is a least-cost modelling based catchment area approach. In summary, using a raster cost-surface that represents the difficulty associated with traversing different parts of a landscape, least-cost catchments are calculated for all cells in a landscape, and the catchments’ areas are then visualised as a map of intra-landscape isolation. However, there are challenges with parameterising least-cost modelling in an ecological context which means it is particularly important that any estimates of isolation are coupled with associated estimates of uncertainty. Using an example of the common brushtail possum Trichosurus vulpecula on the Auckland isthmus New Zealand, we provide a demonstration of how the least-cost modelling based catchment area approach to mapping isolation can be applied to estimate isolation with uncertainty. We use a Python geoprocessing and geovisualisation framework to quantify isolation for four least-cost modelling based possum dispersal scenarios in order to produce a bivariate colour-by-alpha map that simultaneously visualises predicted isolation with associated uncertainty. The bivariate colour-by-alpha map clearly reinforces the need to consider uncertainty when producing catchment area isolation maps. While the overall pattern in isolation was a relatively simple function of the different isolation maps, the associated uncertainty had a complex spatial structure that would be difficult to understand without representing it explicitly. Providing policy decision-makers with bivariate maps that simultaneously include isolation and uncertainty will help support more robust invasive species risk assessments.

In many ecosystems ungulates have coexisted with grasslands over long periods of time. However, high densities of grazing animals may change the floristic and structural characteristics of vegetation, reduce biodiversity, and increase soil erosion, potentially triggering abrupt and rapid changes in ecosystem condition. Alternate stable state theory provides a framework for understanding this type of dynamic. In the Sanjiangyuan atop the Qinghai-Tibetan plateau (QTP), grassland degradation has been accompanied by irruptions of native burrowing animals, which has accentuated the loss of ground cover. Severely degraded areas of alpine meadows are referred to as ‘Heitutan’. Here, using the framework of alternate stable state theory, we describe the proximate and ultimate drivers of the formation of Heitutan on the QTP, and we assess prospects for recovery, in relation to the degree of biophysical alteration, of these alpine meadows. Effective rehabilitation measures must address the underlying causes of degradation rather than their symptoms. Heitutan degradation is not uni-causal. Rather it reflects different mechanisms operating at different spatio-temporal scales across this vast region. Underlying causes include overly aggressive exploitation of the grasslands (e.g. overgrazing), amplification of grazing and erosion damage by small mammals when outbreaks occur, and/or climate change. Given marked variability in environmental conditions and stressors, restorative efforts must vary across the region. Restoration efforts are likely to yield greatest success if moderately and severely degraded areas are targeted as the first priority in management programmes, before these areas are transformed into extreme Heitutan.

Some conceptual models seeking to explain the coexistence of multiple species in hyperdiverse settings predict that species will not be randomly distributed with respect to each other. In stark contrast the ‘ stochastic geometry ’ model assumes that a species fi ne-scale spatial distribution is independent of that of other species in the community. Empirical tests in temperate and tropical forests have provided support for both perspectives. Using point pattern analyses we assessed the prevalence of heterospecifi c associations between 10 500 pairs of species and 3400 pairs of plant functional types (PFTs) in four biodiverse shrubland communities in southwestern Australia. After controlling for fi rst-order eff ects, spatial associations between species and PFTs were rare, but were most prevalent at the least species-rich of the four sites considered. Individuals tended to have fewer species in their local neighbourhoods than expected under a null model of random relabelling, with this departure most pronounced at the site with fewest species. Th e consistency of neighbourhood composition experienced by individuals of the same species is, as a result, less than the average under random mixing. Our results demonstrate that the frequency of heterospecifi c spatial associations is both rare in speciose systems and declines with species richness, and provide further empirical support for the stochastic geometry assumption in species-rich communities.

Demographic and ecophysiological data for the endangered New Caledonian conifer, Araucaria muelleri (Araucariaceae), were analysed to: (i) evaluate population viability in relation to site conditions and human impacts; and (ii) advance our understanding of how stress-tolerator plant functional traits affect the population dynamic behaviour and conservation requirements of long-lived tree species. Growth, survivorship and recruitment in four A. muelleri populations were monitored across 9 years. Demographic rates were analysed using stage-based transition matrices. Leaf δ13C, %N and photosystem II stress (Fv/Fm) were measured for seedlings, saplings and trees, and leaf mass per area (LMA) for trees, and correlations among ecophysiological and demographic variables explored. Seedling, sapling and tree stem growth were among the slowest, and annual survivorship among the highest, reported for any tree. Transition matrix analyses yielded stable estimated population growth rates, λ, not significantly different from 1.0 for all populations. Leaf δ13C was positively correlated with seedling and sapling height growth, while daytime photosystem II stress (Fv/Fm) was high in seedlings, but low in saplings and trees. Ecophysiological measures suggest that individuals transition from moisture-limited growth at the seedling stage to nutrient-limited growth as adults. High levels of environmental stress result in slow stand dynamics, characterized by low recruitment and growth rates counterbalanced by equally low mortality rates. These dynamics result in populations with limited capacity to increase in size quickly, but potentially highly vulnerable to decline should the rate of adult mortality increase. This combination of traits is likely typical of extreme stress-tolerator woody species and suggests a conservation focus on the preservation of mature individuals.

Aim The increased incidence of large fires around much of the world in recent decades raises questions about human and non-human drivers of fire and the likelihood of increased fire activity in the future. The purpose of this paper is to outline a conceptual framework for examining where human-set fires and feedbacks are likely to be most pronounced in temperate forests world-wide and to establish and test a methodology for evaluating this framework using palaeoecological records.

The amount of palaeoecological information available continues to grow rapidly, providing improved descriptions of the dynamics of past ecosystems and enabling them to be seen from new perspectives. At the same time, there has been concern over whether palaeoecological enquiry needs to move beyond descriptive inference to a more hypothesis-focussed or experimental approach; however, the extent to which conventional hypothesis-driven scientific frameworks can be applied to historical contexts (i.e., the past) is the subject of ongoing debate. In other disciplines concerned with human-environment interactions, including physical geography and archaeology, there has been growing use of generative simulation models, typified by agent-based approaches. Generative modelling encourages counter-factual questioning (what if…?), a mode of argument that is particularly important in systems and time-periods, such as the Holocene and now the Anthropocene, where the effects of humans and other biophysical processes are deeply intertwined. However, palaeoecologically focused simulation of the dynamics of the ecosystems of the past either seems to be conducted to assess the applicability of some model to the future or treats humans simplistically as external forcing factors. In this review we consider how generative simulation-modelling approaches could contribute to our understanding of past human-environment interactions. We consider two key issues: the need for null models for understanding past dynamics and the need to be able learn more from pattern-based analysis. In this light, we argue that there is considerable scope for palaeocology to benefit from developments in generative models and their evaluation. We discuss the view that simulation is a form of experiment and, by using case studies, consider how the many patterns available to palaeoecologists can support model evaluation in a way that moves beyond simplistic pattern-matching and how such models might also inform us about the data themselves and the processes generating them. Our emphasis is on how generative simulation might complement traditional palaeoecological methods and proxies rather than on a detailed overview of the modelling methods themselves.

Why are the grassland resources in the Yellow River Source Zone in the Sanjiangyuan region on the Qinghai–Tibet Plateau important for eco-environmental protection in China? What are the key ecological issues in this area? This chapter provides a summary of the biological resources and protection measures in the Yellow River Source Zone. Desertification of Alpine-steppe and Heitutan formation of alpine meadows across the Yellow River Source Zone are assessed to identify the processes of grassland degradation. Impacts of grazing on grassland are related to the effects of climate change. Particular emphasis is placed on the potential role of small mammals in grassland degradation. Building on field analyses, findings from modelling work and key implications from the international literature, alternative stable states in the processes of grassland degradation and restoration are developed to frame management interventions. Management programmes that can be applied to restore Heitutan in the Yellow River Source Zone on the Qinghai–Tibet Plateau are outlined, highlighting implications for sustainable development strategies in the management of grassland resources.