Risk of invasion predicted with support vector machines: A case study on northern pike ( Esox Lucius , L.) and bleak ( Alburnus alburnus , L.)

The dimensionality reductions of environmental variables have a significant effect on the performance of species distribution models

How to effectively obtain species-related low-dimensional data from massive environmental variables has become an urgent problem for species distribution models (SDMs). In this study, we will explore whether dimensionality reduction on environmental variables can improve the predictive performance of SDMs. We first used two linear (i.e., principal component analysis (PCA) and independent components analysis) and two nonlinear (i.e., kernel principal component analysis (KPCA) and uniform manifold approximation and projection) dimensionality reduction techniques (DRTs) to reduce the dimensionality of high-dimensional environmental data. Then, we established five SDMs based on the environmental variables of dimensionality reduction for 23 real plant species and nine virtual species, and compared the predictive performance of those with the SDMs based on the selected environmental variables through Pearson's correlation coefficient (PCC). In addition, we studied the effects of DRTs, model complexity, and sample size on the predictive performance of SDMs. The predictive performance of SDMs under DRTs other than KPCA is better than using PCC. And the predictive performance of SDMs using linear DRTs is better than using nonlinear DRTs. In addition, using DRTs to deal with environmental variables has no less impact on the predictive performance of SDMs than model complexity and sample size. When the model complexity is at the complex level, PCA can improve the predictive performance of SDMs the most by 2.55% compared with PCC. At the middle level of sample size, the PCA improved the predictive performance of SDMs by 2.68% compared with the PCC. Our study demonstrates that DRTs have a significant effect on the predictive performance of SDMs. Specifically, linear DRTs, especially PCA, are more effective at improving model predictive performance under relatively complex model complexity or large sample sizes.

Fulfilling spawning flow requirements for potamodromous cyprinids in a restored river segment

The Baixo Sabor hydroelectric scheme was recently constructed (began operating in 2015), blocking fish migration immediately at the river mouth. As a compensation measure for spawning habitat loss the Vilariça River which confluences with Sabor River, immediately downstream from the Baixo Sabor scheme, was chosen for rehabilitation site as an attempt to replace the spawning grounds lost in Sabor River for the potamodromous cyprinid Iberian barbel (Luciobarbus bocagei). Restoration activities included: placement of submerged weirs with fish ramps to increase the lotic/lentic sequence, bank reinforcement, boulder placement and river water profile increase. Water from the nearest reservoir was channeled underground to the lower Vilariça River in order to increase the water flow. The most appropriate volume of water was determined by means of a 1D model coupled with the CASiMiR to determine the weighted usable area for barbel spawners. Nine spawning grounds were identified. Fish migrations began in mid-April when water temperature increased (c. 16°C) and water flow decreased. Barbel tend to spawn in coarse substrate with shallow water and moderate current. Ontogenetic differences in habitat use were reported, with barbel juveniles preferring weaker currents and deeper zones when compared to spawners. Females also display different preferences, which are more apparent after spawning. Both life-stages exhibit the maximum habitat suitability at c. 1.5m³/s. The habitat model used in this study was able to accurately predict the occurrence of spawning grounds, representing, therefore, a valuable decision-making support tool for river managers.

Assessing the suitable habitat for reintroduction of brown trout (Salmo trutta forma fario) in a lowland river: A modeling approach

Huge efforts have been made during the past decades to improve the water quality and to restore the physical habitat of rivers and streams in western Europe. This has led to an improvement in biological water quality and an increase in fish stocks in many countries. However, several rheophilic fish species such as brown trout are still categorized as vulnerable in lowland streams in Flanders (Belgium). In order to support cost-efficient restoration programs, habitat suitability modeling can be used. In this study, we developed an ensemble of habitat suitability models using metaheuristic algorithms to explore the importance of a large number of environmental variables, including chemical, physical, and hydromorphological characteristics to determine the suitable habitat for reintroduction of brown trout in the Zwalm River basin (Flanders, Belgium), which is included in the Habitats Directive. Mean stream velocity, water temperature, hiding opportunities, and presence of pools or riffles were identified as the most important variables determining the habitat suitability. Brown trout mainly preferred streams with a relatively high mean reach stream velocity (0.2-1 m/s), a low water temperature (7-15°C), and the presence of pools. The ensemble of models indicated that most of the tributaries and headwaters were suitable for the species. Synthesis and applications. Our results indicate that this modeling approach can be used to support river management, not only for brown trout but also for other species in similar geographical regions. Specifically for the Zwalm River basin, future restoration of the physical habitat, removal of the remaining migration barriers and the development of suitable spawning grounds could promote the successful restoration of brown trout.