Ecological monitoring through harmonizing existing data: Lessons from the boreal avian modelling project

Barker, Nicole K. S.; Fontaine, Patricia C.; Cumming, Steven G.; Stralberg, Diana; Westwood, Alana R.; Bayne, Erin M.; Sólymos, Péter; Schmiegelow, Fiona K. A.; Song, Samantha J.; Rugg, David

doi:10.1002/wsb.567

Cited by 22 publications

(30 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For species for which we had BBS EDR estimated, we extracted corresponding off-road boreal EDR estimates derived from the Boreal Avian Modelling (BAM) Project database (http://www.borealbirds.ca) using the same analytical approach (Sólymos et al 2013, Sólymos 2016. BAM uses off-road point count data from across the boreal to generate habitat-and species-specific EDR estimates for 75 species (Sólymos et al 2013, Barker et al 2015, Sólymos 2016. BBS EDR estimates were generated for 51 species, but we only report values for those with ≥20 counts during the BBS surveys to assure reliable estimates.…”

Section: Discussionmentioning

confidence: 99%

Potential benefits of augmenting road-based breeding bird surveys with autonomous recordings

Mahon¹,

Haché²,

Sólymos³

et al. 2017

ACE

Self Cite

View full text Add to dashboard Cite

ABSTRACT. The North American Breeding Bird Survey (BBS) is one of the longest annual avian surveys and has the greatest spatiotemporally extensive coverage in the Western Hemisphere. Although this important survey provides trend estimates for more than 400 species, it has limited coverage in the boreal forest and biases in representation and detectability that complicate inference. Thus, there is a need to evaluate the potential of new technologies and analytical approaches to increase coverage and improve monitoring efficiency. We documented variation in counts between BBS surveys (hereafter "human BBS") and different on-road and forest-edge surveys using autonomous recording units (ARUs) from 3 routes in the Northwest Territories, Canada. Specifically, we quantified percent differences (i.e., bias in counts) in species richness, abundance indices of birds, and species-specific variation in counts between human BBS and ARU-based surveys conducted on-road and at the forest edge at different dates and times of day. We also generated on-road effective detection radius (EDR) estimates for 15 species and tested for species-specific differences in EDR to explain bias in counts between on-road and forest-edge ARU surveys. Overall, species richness and abundance indices in human BBS surveys were higher than forest-edge ARU surveys conducted simultaneously and when similar forest-edge ARU surveys were conducted at sunset and a week earlier in June. However, there was no difference when comparing values from human BBS with on-road ARU BBS and forest-edge ARU surveys conducted at sunrise. Extracting the maximum count per species from 4 types of 3-minute forest-edge surveys increased counts by 62% and 64% for species richness and abundance indices, respectively, relative to human BBS, but the importance of this bias differed considerably among the 10 most common species in the study area. Our results suggest that false-negative bias in species detection could be corrected with appropriate methods, and ARUs deployed at the forest edge near BBS stops could be used to increase data quality of on-road surveys. When combined with appropriate correction factors to adjust for surveys done at the forest edge, ARUs could also be used to increase the geographic coverage of boreal surveys by allowing inexperienced volunteers to collect BBS data along winter or secondary roads in remote locations. Avantages

show abstract

Section: Discussionmentioning

confidence: 99%

Potential benefits of augmenting road-based breeding bird surveys with autonomous recordings

Mahon¹,

Haché²,

Sólymos³

et al. 2017

ACE

Self Cite

View full text Add to dashboard Cite

show abstract

“…We used point-count survey data compiled by the Boreal Avian Modelling Project (Cumming et al 2010, Barker et al 2015. In a point-count survey, observers in the field stand at a point and record signals of individual birds over a specified time interval within a circle of given radius (Ralph et al 1993, Matsuoka et al 2014.…”

Section: Datamentioning

confidence: 99%

Phylogeny and species traits predict bird detectability

et al. 2018

Self Cite

View full text Add to dashboard Cite

Avian acoustic communication has resulted from evolutionary pressures and ecological constraints. We therefore expect that auditory detectability in birds might be predictable by species traits and phylogenetic relatedness. We evaluated the relationship between phylogeny, species traits, and field‐based estimates of the two processes that determine species detectability (singing rate and detection distance) for 141 bird species breeding in boreal North America. We used phylogenetic mixed models and cross‐validation to compare the relative merits of using trait data only, phylogeny only, or the combination of both to predict detectability. We found a strong phylogenetic signal in both singing rates and detection distances; however the strength of phylogenetic effects was less than expected under Brownian motion evolution. The evolution of behavioural traits that determine singing rates was found to be more labile, leaving more room for species to evolve independently, whereas detection distance was mostly determined by anatomy (i.e. body size) and thus the laws of physics. Our findings can help in disentangling how complex ecological and evolutionary mechanisms have shaped different aspects of detectability in boreal birds. Such information can greatly inform single‐ and multi‐species models but more work is required to better understand how to best correct possible biases in phylogenetic diversity and other community metrics.

show abstract

“…Changing climatic conditions will affect the distribution of many landbirds. Continued efforts to combine data from independent studies at a continental scale can leverage existing data to understand the combined effects of climate change, industrial activity, protected areas, and ecological processes on avian populations (Bled et al 2013, Barker et al 2015, Roy et al 2019).…”

Section: Discussionmentioning

confidence: 99%

Landbird trends in protected areas using time‐to‐event occupancy models

et al. 2019

View full text Add to dashboard Cite

Global populations of wildlife are affected by human activity, land cover change, and climate change. Long-term monitoring programs across large spatial scales are required to understand how these and other factors affect wildlife populations. Occupancy models are frequently used to monitor changes in species distribution while accounting for imperfect detection. Occupancy surveys can be expensive because they typically require multiple surveys to estimate the probability of detection. Time-to-detection models provide a promising approach for estimating occupancy because they require just one visit; however, few studies have tested or applied these models to wildlife data. We ran a simulation study to assess biases of time-to-event occupancy models for standardized avian point-count surveys and then applied the models to 10 yr of data. Time to first detection occupancy models had minimal bias and almost nominal coverage for species with a mean time to first detection <8 min on surveys with 10 min of sampling. Biases and root mean squared error increased with increasing time to first detection. We applied a single species, multiyear occupancy model to 34,665 detections of 77 landbird species collected across 500 km of latitude in five protected areas along the Rocky Mountains. Models from 64 species converged and had mean times to first detection <8 min. Average time to first detections was 3.2 min, which reflected a cumulative probability of detection of 0.96. Occupancy rates increased, decreased, and remained unchanged for 53%, 9%, and 38% of species, respectively. Overall, occupancy rates increased in 2015 and 2016 for short-and long-distance migrants and decreased slightly for winter residents. Average decadal temperature and precipitation were important predictors for almost half of the species, while annual changes in spring temperature and precipitation affected 23% of species. Our studies demonstrate that time to first event occupancy models provide an efficient method for monitoring changes in distribution so long as encounter rates are much shorter than the survey duration. Our stable to increasing trends and strong responses to spring temperature and precipitation highlight the value of long-term monitoring for understanding how changing climatic conditions affect wildlife.

show abstract

Ecological monitoring through harmonizing existing data: Lessons from the boreal avian modelling project

Cited by 22 publications

References 45 publications

Potential benefits of augmenting road-based breeding bird surveys with autonomous recordings

Potential benefits of augmenting road-based breeding bird surveys with autonomous recordings

Phylogeny and species traits predict bird detectability

Landbird trends in protected areas using time‐to‐event occupancy models

Contact Info

Product

Resources

About