Roxborough Park Community Wildfire Evacuation Drill: Data Collection and Model Benchmarking.

Wildfires are increasing in scale, frequency and longevity, and are affecting new locations as environmental conditions change. This paper presents a dataset collected during a community evacuation drill performed in Roxborough Park, Colorado (USA) in 2019. This is a wildland-urban interface community including approximately 900 homes. Data concerning several aspects of community response were collected through observations and surveys: initial population location, pre-evacuation times, route use, and arrival times at the evacuation assembly point. Data were used as inputs to benchmark two evacuation models that adopt different modelling approaches. The WUI-NITY platform and the Evacuation Management System model were applied across a range of scenarios where assumptions regarding pre-evacuation delays and the routes used were varied according to original data collection methods (and interpretation of the data generated). Results are mostly driven by the assumptions adopted for pre-evacuation time inputs. This is expected in communities with a low number of vehicles present on the road and relatively limited traffic congestion. The analysis enabled the sensitivity of the modelling approaches to different datasets to be explored, given the different modelling approaches adopted. The performance of the models were sensitive to the data employed (derived from either observations or self-reporting) and the evacuation phases addressed in them. This indicates the importance of monitoring the impact of including data in a model rather than simply on the data itself, as data affects models in different ways given the modelling methods employed. The dataset is released in open access and is deemed to be useful for future wildfire evacuation modelling calibration and validation efforts. Supplementary Information: The online version contains supplementary material available at 10.1007/s10694-023-01371-1.

Eastern Canada Flocks: images and manually annotated bird positions.

The Eastern Canada (ECA) Flocks data set consists of manually annotated images from the Common Eider (COEI, Somateria mollissima) Winter Survey and the Greater Snow Geese (GSGO, Anser caerulescens atlanticus) Spring Survey. The images were taken in eastern Canada using fixed-wing aircraft and manually annotated with ImageJ's Cell counter plugins. We selected and annotated the ECA Flocks images in order to test the precision of the CountEm flock size estimation method. ECA Flocks includes 179 COEI and 99 GSGO single flock images. We cut each image manually to a rectangle that excluded large parts of the image with no birds. Both versions (original and cut) of each image are available in the data set. We manually annotated 637,555 (124,309 COEI and 514,235 GSGO) bird positions in the cut images from both surveys. Each bird has an associated "Type," which refers to species and/or sex. Sex identification was only possible for adult common eiders, because females and immature males are brown birds, whereas adult males have mainly white plumage. In the COEI images 64,484 males and 58,029 females, as well as 1,796 birds of other species, were identified. In the GSGO images 504,891 Snow Geese and 9,344 birds of other species were labeled. A .csv file including all annotated bird positions and types is available for each image. The COEI and GSGO photos of the ECA Flocks data set were taken in the years 2006 and 2018 and 2016-2018, respectively. We selected these photos in order to include images with different quality and resolution. COEI and GSGO flock sizes range from 6 to 4,154 and from 43 to 36,241 respectively. There is high variability in light conditions, backgrounds, and number and spatial arrangement of birds across the images. The data set is therefore potentially useful to test the precision of methods for analyzing imagery to estimate the abundance of animals by directly detecting, identifying, and counting individuals. We release these data into the public domain under a Creative Commons Zero license waiver. When you use the data in your publication, cite this data paper. Should ECA Flocks be a major part of the data analyzed in your study, you should consider inviting the ECA Flocks originators as collaborators. If you plan to use the ECA Flocks data set, we request that you contact the ECA Flocks core team to learn whether updates are available, and whether similar analyses are already ongoing.

Correction: Simplified procedure for efficient and unbiased population size estimation.

[This corrects the article DOI: 10.1371/journal.pone.0206091.].

Simplified procedure for efficient and unbiased population size estimation.

Population size estimation is relevant to social and ecological sciences. Exhaustive manual counting, the density method and automated computer vision are some of the estimation methods that are currently used. Some of these methods may work in concrete cases but they do not provide a fast, efficient and unbiased estimation in general. Recently, the CountEm method, based on systematic sampling with a grid of quadrats, was proposed. It offers an unbiased estimation that can be applied to any population. However, choosing suitable grid parameters is sometimes cumbersome. Here we define a more intuitive grid parametrization, using initial number of quadrats and sampling fraction. A crowd counting dataset with 51 images and their corresponding, manually annotated position point patterns, are used to analyze the variation of the coefficient of error with respect to different parameter choices. Our Monte Carlo resampling results show that the error depends on the sample size and the number of nonempty quadrats, but not on the size of the target population. A procedure to choose suitable parameter values is described, and the expected coefficients of error are given. Counting about 100 particles in 30 nonempty quadrats usually yields coefficients of error below 10%.