Assessing Landscape Constraints on Species Abundance: Does the Neighborhood Limit Species Response to Local Habitat Conservation Programs?

Abstract

Landscapes in agricultural systems continue to undergo significant change, and the loss of biodiversity is an ever-increasing threat. Although habitat restoration is beneficial, management actions do not always result in the desired outcome. Managers must understand why management actions fail; yet, past studies have focused on assessing habitat attributes at a single spatial scale, and often fail to consider the importance of ecological mechanisms that act across spatial scales. We located survey sites across southern Nebraska, USA and conducted point counts to estimate Ring-necked Pheasant abundance, an economically important species to the region, while simultaneously quantifying landscape effects using a geographic information system. To identify suitable areas for allocating limited management resources, we assessed land cover relationships to our counts using a Bayesian binomial-Poisson hierarchical model to construct predictive Species Distribution Models of relative abundance. Our results indicated that landscape scale land cover variables severely constrained or, alternatively, facilitated the positive effects of local land management for Ring-necked Pheasants.

Effects of rotational grazing management on nesting greater sage‐grouse

“…This was evident in the relative magnitudes of the standard deviations for ranch and year effects (Table 1), which indicated variation among years was >2.5 times greater than among ranches. This is suggestive that factors such as weather or annually fluctuating predator abundance overwhelm the effects of grazing management, and is consistent with the emerging understanding that effects of local vegetation manipulation on prairie grouse are relatively minor in comparison to processes and patterns operating at broader scales, such as weather and landscape fragmentation (Jorgensen et al 2014, Hovick et al 2015, Fuhlendorf et al 2017. With only 6 years of observation, however, our estimate of s a is imprecise (Table 2).…”

Shorebird stopover habitat decisions in a changing landscape

“…Abundance differences between the RWB and the PPR may simply reflect intrinsic historical differences between the regions. Still, it is worth considering that the differences we find between the RWB and the PPR may also reflect the influence of regional land‐cover driving habitat selection at local stopover sites (Hutto , Buler et al ), and local species abundance in agricultural landscapes (Jorgensen et al ). Many studies have specifically noted the importance of dense networks of wetlands for sandpiper species (Skagen and Knopf ; Farmer and Parent ; Niemuth and Solberg ; Albanese and Davis , ).…”

A Bayesian method for assessing multi-scale species-habitat relationships

“…In a similar study of ring-necked pheasants in Nebraska, Jorgensen et al (2014) created a mixedscale model based on effect size selection from Bayesian hierarchical N-mixture models of predictors estimated separately at 2 scales (local management: 1 km radius, and landscape: 5 km radius). The authors chose to model predictors separately by scale because of autocorrelation, and avoided information criteria (i.e., AIC, BIC, DIC) because of their inappropriateness for hierarchical mixture models.…”

“…Backward stepwise selection was performed by fitting models first containing all scales of a single predictor at once (e.g., three full models, one for each predictor) and dropping the least important scale based on the largest 'p-value' calculated using a normal approximation of the posterior distribution until only a single scale was left for each predictor. Coefficient effect size-based model selection was performed by estimating models containing all predictors separately by scale (i.e., each model contained all predictors measured at the same single scale) and selecting the scale with the strongest estimated coefficient for each predictor individually (e.g., Jorgensen et al 2014). Information criterionbased model selection was performed by calculating AIC and WAIC for independent predictor by scale models (i.e., each model contained only 1 predictor measured at 1 scale for 12 total models) and selecting the scale model with the lowest AIC (Akaike and Hirotogu 1998) or WAIC (Watanabe 2013;Gelman et al 2014) for each predictor (e.g., Kirol et al 2015).…”

“…While potentially informative, the notion of a single 'characteristic' scale to which a species may associate fails to consider that, in general, the processes that define a species' niche often interact, with the possibility of multiple explanatory variables acting at different spatial scales (e.g., Cushman and McGarigal 2002). However, few studies evaluate the explanatory performance of different variables collected at multiple scales simultaneously (e.g., Grand and Cushman 2003;Pearson et al 2004;Jorgensen et al 2014), but such studies have the ability to provide insights into the hierarchical nature of spatial processes (Urban et al 1987).…”