Ecological Modelling最新文献

This study proposes a system of non-linear ordinary differential equations (ODEs) based on a SIR-type epidemic model to explore the role that humans and dairy cattle play in the contamination process and the effects of contaminated dairy products and environments in salmonellosis dynamics. The driving force behind using a system of non-linear ODEs lies in their ability to accurately capture biological complexity, such as population interactions, which are inherently non-linear. The positivity of model solutions is checked to ensure their mathematical and biological compatibility. The Lyapunov stability theorem is used to examine the global stability of the equilibria states based on the value of a threshold parameter $R_0$, also known as the basic reproduction number, which is formulated using the next generation matrix approach. Parameter estimation and model fitting are done using the least squares method along with the salmonellosis incidence data from 1990 to 2020. The uncertainty and global sensitivity analysis are performed using the Latin hypercube sampling and partial rank correlation coefficients methods. The dynamics of $R_0$ along with model parameters responsible for contamination are explored using contour plots. The proposed model is simulated with a statistical range of 95% confidence interval to assess the reliability and precision of the estimates derived from the real data. The simulations are performed to explore the complex scenarios and forecast the dynamics of salmonellosis. The results show that the bacteria growth rate, ingestion rates of Salmonella typhimurium bacteria in contaminated dairy products and environments, shedding rates of Salmonella typhimurium into the environments by the infected humans and dairy cattle, and the harvesting rate of contaminated dairy products are directly proportional to $R_0$, meaning that they are responsible for increasing the severity of salmonellosis whenever they are increased, while the decay rates of Salmonella typhimurium bacteria are inversely proportional to $R_0$. Control measures such as hand hygiene, food preparation cleanliness, cooking food thoroughly, isolating sick and new animals, and training farmers on symptom recognition and reporting cases to medical staff are recommended to mitigate the severity of salmonellosis.

The northward migration of several tree species ranges is likely to lag behind climate change due to slow demography, competitive interactions, and dispersal limitations. These will result in a colonization credit, where suitable climate envelopes are left unoccupied, and extinction debt, where tree stands persist at unsuitable climatic locations. While the underlying mechanisms explaining the delayed range shift of forest trees have been investigated, few studies have focused on how management could overcome this lag. Here we extend a forest community state model derived from the metapopulation theory and validated with over 40,000 forest inventory plots, to formulate how forest management can accelerate the response of the boreal-temperate ecotone under warming temperature. We first complete the model equations to represent how four types of forest management may affect the transitions between four forest states: Boreal, Temperate, Mixed and Regeneration. We then simulated the potential of forest management to reduce colonization credit and extinction debt using two complementary approaches to measure the resilience and range shift of the boreal-temperate ecotone in response to warming temperature. Our simulations reveal that paying the colonization credit by planting temperate trees in a stand in Regeneration or Boreal state are likely to i) reduce the return time to equilibrium, ii) increase forest resilience, and iii) move the ecotone towards colder temperatures. Surprisingly, harvesting boreal trees in stands in Boreal or Mixed state were not effective to reduce extinction debt and provide colonization opportunities for temperate trees. Our results suggest that forest management related to planting actions could help the boreal-temperate ecotone keep pace with climate change. Future experiments are required to test these theoretical expectations and make operational recommendations.

We develop a framework for incorporating biodiversity impacts into land use optimization models. Using the countryside species–area relationship, six approaches to incorporating biodiversity are evaluated: constraint versus objective function, and with characterization factors, piecewise linear, or non-linear implementation. We also consider the underlying ethical assumptions of different options for biodiversity evaluation considering taxa and ecoregions. To explore the six approaches, we use a case study that considers both climate change mitigation and the biodiversity impact due to habitat loss. We find that the characterization factor approach underestimates biodiversity loss, while the piecewise linear approach somewhat overestimates. Utilizing two approaches: biodiversity protection as the objective, and biodiversity as a constraint allows the construction of a Pareto frontier for cost versus loss of species. In addition to minimum cost solutions and minimum biodiversity loss solutions, the integrated model allows to identify (1) solutions with improved outcomes, those better for biodiversity and with small changes on costs, (2) mechanisms the model uses to achieve the improved solutions, (3) trade-offs and (4) challenges.

Endosymbiotic bacteria are being increasingly considered in novel methods of suppressing pest populations and/or reducing their potential to vector viruses. Here, we develop a generic modelling approach to explore the potential impacts of endosymbionts on pest populations. The model is presented as a modular R package, EndoSim. We use it to model demographic changes in the green peach aphid (Myzus persicae), which has recently been transinfected with the facultative endosymbiont, Rickettsiella. This novel transinfection has deleterious fitness effects following transfer to M. persicae. We consider a range of scenarios including different temperatures, thermal response curves, plant-based transmission dynamics, introduction dates and initial introduction sizes to investigate conditions under which Rickettsiella is expected to suppress M. persicae populations. We show that impacts on field pest populations are influenced by endosymbiont transmission dynamics, which are temperature-dependent, as well as aphid population processes. Our modelling approach suggests Rickettsiella could be used as a potential biological control to suppress M. persicae in natural settings. Furthermore, it highlights critical information that is required to accurately estimate the likely efficacy of novel endosymbiont transinfections in aphid pest control.

Temperature time series data are a composition of average trends and stochastic variability that together shape population dynamics. However, models of temperature-dependent species often overlook variability, focusing solely on growth rates under average conditions. When models omit variability, they can inaccurately predict the dynamics that underlie the establishment of invasive pests sensitive to temperature fluctuations. Here, we conduct a stochastic modeling study of the spotted lanternfly (Lycorma delicatula), a univoltine grape pest, which has invaded grape growing regions of the eastern U.S. due to human transport, leading to frequent establishment of populations in urban and suburban areas. As the spotted lanternfly continues to be transported to new grape growing regions and climate change alters variability, it is vital to predict its establishment potential. Although it overwinters as diapausing eggs, experiments suggest that diapause is plastic and not necessary for survival. We developed a deterministic stage-age-structured partial differential equation model of diapausing and non-diapausing populations. We derived a new metric quantifying population resistance to climatic variability defined as the level of stochasticity that leads to negative growth compared to average conditions. We simulated growth rates and resistance to variability across a range of average temperature conditions and stochasticity. We then analyzed how variability and diapause interact with survival, fecundity, and development to affect population dynamics. Finally, we estimated establishment potential across all U.S. cities. Diapausing populations were typically more resistant than non-diapausing populations because diapause enhances overwintering egg survival during winter cold waves, while allowing accelerated development and increased fecundity during summer and fall heat waves. Establishment potential is especially underestimated in important grape growing regions of California if models of diapausing populations omit variability. By quantifying population resistance to climatic variability, we gain a fuller understanding of invasive species establishment in today's stochastic and changing climate.

Simulating algae blooms using a hydrodynamic-water quality model is challenging because it requires a thorough understanding of physical and biological processes and involves numerous parameters. This study conducted a sensitivity analysis of the EFDC+ hydrodynamic and water quality model for simulating cyanobacteria growth, an important Harmful Algal Bloom (HAB) species in the Ohio River, USA. The sensitivity analysis assessed 23 model input parameters, divided into nine functional groups according to their characteristics. This assessment analyzes the impact of changing these input parameters on four water quality model outputs including algae (i.e., cyanobacteria), dissolved oxygen, total nitrogen, and total phosphorus. Light extinction parameters, maximum algal growth rate, and algal base metabolism were identified as the most sensitive parameters for simulating algal growth. Solar radiation required for algal growth was moderately sensitive. Currently, there are only a few studies that simulate HAB dynamics in riverine systems. This study deepens our understanding of HAB development in rivers with lock and dam structures that create a series of pools along the river. Future work will involve focusing on the sensitive parameters in model calibration.

Agricultural intensification is widely recognised as a primary driver of pollinator loss, but the success of land-management actions designed to remediate its impact is often mixed. Payments to farmers to increase habitat connectivity or the availability of floral and nesting resources may only result in short-term gains or even unintended consequences. The reasons may lie in changes to interaction networks or competition intensity that remain poorly understood. Models of pollination service typically implicitly assume pollinator population dynamics are regulated by nest-site availability, even though empirical evidence suggests nest-site occupancy is likely at least in part dependent on floral resource availability. To investigate the consequences of competition for floral resources in coarse-grained agricultural landscapes we extended an established model for bees combining optimal foraging and population dynamics, to include new functions for floral resource depletion and realistic colony dynamics. We find that intra-specific competition occurs late in the season forcing bees to forage underutilised sites situated further towards their foraging range limits. A lower rate of energy acquisition ultimately limits the size of the colony peak and delays its timing. Consequently, competition for floral resources can limit population size and distribution while at the same time contributing to a more stable and efficacious pollination service. Although competition was not found to be important in nest-site establishment success, the effect of a hunger gap early in the season on nest-site occupancy indirectly influences competition later in the season leading to complex outcomes.

The eastern oyster (Crassostrea virginica) provides numerous ecosystem services such as building reef habitat, clarifying the water by filtering seston, and reducing excess nitrogen when their biodeposits are denitrified. Some of these ecosystem services have been extensively studied using two-dimensional and three-dimensional (3D) models. Yet, the relationship between oyster abundance, their filtration and biodeposition rates, and associated water quality metrics has not been estimated with high-resolution models that include biodeposit resuspension as well as simulation of 3D processes in the sub-tributaries where oysters are abundant. To undertake these estimates, a 3D Regional Ocean Modeling System (ROMS) framework, comprised of a coupled hydrodynamic-water quality model with an oyster filtration and biodeposition sub-model, was implemented over a fine-resolution model grid (120–150 m) of the Choptank River on the eastern shore of Chesapeake Bay, U.S.A. After validation with data from 20 cruises between May and September 2010, the model was used to predict seven variables associated with water quality in simulations with zero, recent, and 50x recent abundances of oysters. Results indicated that improvement in the seven water quality variables differed in response to the abundance of oysters and between regions in the Choptank River. In line with expectations, the water quality metrics improved with increasing oyster abundance in the shallow and retentive sub-tributaries of Broad and Harris Creeks. In contrast, in the deeper and more flushed mainstem of the Choptank River, some water quality metrics deteriorated when recent abundances of oysters were added to the model compared to the simulation with zero oysters and the same metrics improved when oyster abundances were increased by 50x. Overall, model results indicated that the many complex physical and biogeochemical processes that influence the effect of oysters on water quality resulted in differing responses to oysters across different systems. In addition, this study introduces a high-resolution, predictive tool that could be used to estimate the number of oysters needed to meet water quality thresholds in specific systems in support of ecosystem management.

Habitat fragmentation and loss are major threats to species conservation worldwide. Studying species-habitat relationships is a crucial first step toward understanding species habitat requirements, which is necessary for conservation and management planning. However, some species inhabit a range of habitat types, potentially making the use of range-wide habitat models inappropriate due to non-stationarity in species-habitat preferences. The Mexican spotted owl (Strix occidentalis lucida) (MSO) is a species that inhabits both forests and rocky canyonlands, two habitats with large differences in environmental conditions. It is unclear whether the species uses habitat differently in these two habitat types or if previously-built habitat models for forest-dwelling owls can be used to understand MSO habitat use in rocky canyonlands. To explore this, we developed the first scale-optimized habitat suitability model for this subspecies of spotted owl in rocky canyonlands using an ensemble framework. We then compared our results with a previously-built habitat model for MSO in forested areas. In the rocky canyonland model, slope (800 m scale), cumulative degree days (1200 m scale), insolation (1000 m scale), and monsoon precipitation (100 m scale) were the most important environmental covariates. In contrast, in the forest model, percent canopy cover (100 m scale), percent mixed-conifer (5000 m scale), and slope (500 m scale) were the most important environmental covariates. The rocky canyonland model performed well, while the forest model performed poorly when projected to rocky canyonlands and predicted low suitability across the entire study area, including areas with known nesting locations. These results support the non-stationarity in habitat use for MSOs between rocky canyonland and forest habitats. Hence, when transferring habitat suitability models from one region to another, it is necessary to evaluate the transferability of the model by accounting for non-stationarity in species-habitat preferences.

This work compiles the conceptual and practical proposals of various research groups regarding interesting properties of mathematical models. Although different types of mathematical models are mentioned, phenomenological-based semi-physical models (PBSM) are the main focus of this paper. The partitioning of the process to be modeled, the scalability of the model, and the interpretability of its parameters are presented as important properties of the model that help to understand and use the mathematical model. The use of these properties is illustrated with examples, providing the respective bibliographical reference where an expanded presentation of each example can be reviewed. The importance of each model property presented is explained from the inherent condition of the model as a mathematical object and from the expectation of the final users of the model. The confidence in the model is increased when the user knows how the model is obtained. These properties provide valuable information to the end use of the model. Therefore, the modeler must understand and apply this knowledge during model deduction and write, in the model report, how the mentioned properties were tested.