Search

Spatial capture-recapture (SCR) models are now widely used for estimating density from repeated individual spatial encounters. SCR accounts for the inherent spatial autocorrelation in individual detections by modelling detection probabilities as a function of distance between the detectors and individual activity centres. However, additional spatial heterogeneity in detection probability may still creep in due to environmental or sampling characteristics. if unaccounted for, such variation can lead to pronounced bias in population size estimates. In this paper, we address this issue by describing three Bayesian SCR models that use generalized linear mixed modelling (GLMM) approach to account for latent heterogeneity in baseline detection probability across detectors with: independent random effects (RE), spatially autocorrelated random effects (SARE) with components of prior covariance matrix modelled as a decreasing function of inter-detector distance, and a two-group finite mixture model (FM) to identify latent detectability classes of each detector. We test these models using a simulation study and an empirical application to a non-invasive genetic monitoring data set of female brown bears (Ursus arctos) in central Sweden. In the simulation study, all three models largely succeeded in mitigating the biasing effect of spatially heterogeneous detection probability on population size estimates. Overall, SARE provided the least biased population size estimates (median RB: -9% – 6%). When spatial autocorrelation in detection probability was high, SARE also performed best at predicting the spatial pattern of heterogeneity in detection probability. At intermediate levels of autocorrelation, spatially-explicit estimates of detection probability obtained with FM were more accurate than those generated by SARE and RE. The empirical example revealed patterns consistent with the results from the simulation study. We found that ignoring spatial heterogeneity in detection probability led to at least 22% lower estimate of bear population size compared to models that accounted for it (i.e., SARE and RE models). When the number of detections per detector is low (≤ 1), the GLMMs considered here may require dimension reduction of the random effects by pooling baseline detection probability parameters across neighbouring detectors ("aggregation") to avoid over-parameterization. The added complexity and computational overhead associated with SCR-GLMMs may only be justified in extreme cases of spatial heterogeneity, e.g., large clusters of inactive detectors unbeknownst to the investigator. However, even in less extreme cases, detecting and estimating spatially heterogeneous detection probability may assist in planning or adjusting monitoring schemes. • Unaccounted spatial variation in detectability can lead to biased density estimates using spatial capture-recapture. • Generalized linear mixed modelling succeeded in mitigating the biasing effect. • Estimating variation in detectability assist in planning monitoring schemes. [ABSTRACT FROM AUTHOR]

Forest biomass is a renewable resource that is increasingly utilised for bioenergy purposes in Sweden, which along with the extraction of industrial wood may conflict with biodiversity conservation. The aim of this paper is to present a method for integrated sustainability assessment of forest biomass extraction, particularly from bioenergy and biodiversity perspectives. The landscape simulator LandSim was developed and linked with models for the assessment of biomass yields and habitat networks representing prioritised biodiversity components. It was applied in a case study in Kronoberg County in southern Sweden. Forest growth and management were simulated for the period 2010–2110, following two land zoning scenarios, one applying even-aged forest management on all forest land except for protected areas (EAF-tot), and one applying continuous cover forest management on parts of the forest land, combined with protected areas and an intensified even-aged management on the other parts (CCF-int). The EAF-tot scenario implied higher yields of biomass feedstock for bioenergy, the CCF-int scenario only giving 66% of that yield, while the CCF-int scenario performed substantially better when it came to the habitat network indicators, if habitat suitability was ensured. Conclusively, the case study confirmed that the modelling framework of the LEcA tool, linking the landscape simulator LandSim with the biomass yield assessment and the habitat network model can be used for integrating main policy concerns when assessing renewable energy options. [ABSTRACT FROM AUTHOR]

• We describe BAGRA_N, an extended version of the BASGRA model for grasslands. • This model simulates forage nutritive value, N leaching, NO and N 2 O emissions. • Parameterization and validation against field experimental data from Sweden and Norway. • Nutritive value attributes and dry matter were predicted satisfactorily. • An application to optimize N fertilization against yield and N losses is presented. The main objective of this paper is to present the new model BASGRA_N, to show how it was parameterized for grass swards in Scandinavia, and to evaluate its performance in predicting above-ground biomass, crude protein, cell wall content and dry matter digestibility. The model was developed to allow simulation of: (1) the impact of N-supply on the plants and their environment, (2) the dynamics of greenhouse gas emissions from grasslands, (3) the dynamics of cell-wall content and digestibility of leaves and stems, which could not be simulated with its predecessor, the BASGRA-model. To calibrate and test the model, we used field experimental data. One dataset included observations of biomass (DM) and crude protein content (CP) under different N fertilizer regimes from five sites in central and southern Sweden. The other dataset included observations of DM, and sward components as well as CP, cell wall content (NDF) and DM digestibility as affected by harvesting regime from one site in southwestern Norway. The total number of experiments was nine, of which three were used for model testing. When BASGRA_N was run with the maximum a-posteriori (MAP) parameter vector from the Bayesian calibration for the Swedish test sites, DM and CP were both simulated to an overall Pearson correlation coefficient (Rb) of minimum 0.58, Willmott's index of agreement (d) of minimum 0.69 and normalized root mean squared error (NRMSE) of maximum 0.30. Corresponding metrics for Norwegian test sites were 0.93, 0.96 and 0.27 for DM and >0.73, >0.61, <0.18 for DM digestibility, NDF and CP content, respectively. We conclude that BASGRA_N can be used to simulate yield and CP responses to N with satisfactory precision, while maintaining key features from its predecessor. The results also suggest that DM digestibility and NDF can be simulated satisfactorily, which is supported by results from a recent model comparison study. Further testing of the model is needed for a few variables for which we currently do not have enough data, notably leaching and emission of N-containing compounds. Further work will include application of the model to investigate greenhouse gas mitigation options, and evaluation against independent data for the conditions for which it will be applied. [ABSTRACT FROM AUTHOR]