Calibration of visually estimated distances to migrating seabirds with radar measurements

Abstract

Censusing seabirds from coastal areas requires reliable estimates of bird numbers and the distances of the birds from the coastline. Logistical constraints make visual estimation of distances the only feasible method in many studies. We tested the accuracy of visually estimated offshore distances of six migratory seabird species in the Strait of Gibraltar using simultaneous measurements obtained by radar. Most birds (91%) were detected within 3 km of the coast and we truncated our calibration at this distance. We found a strong correlation between radar and visual estimates (R 2 = 0.83, P < 0.0001). The magnitude of errors in visual estimates was moderate and ranged from 0.08 to 0.20 for different distances and observers. Among the factors potentially affecting the accuracy of visual estimates of distance to seabird in our study were observer identity, bird species, bird behavior, and weather; the most parsimonious model in our study included observer identity as the only predictor, and no model with more than one predictor had a smaller Akaike's information criterion value. Radar can be used to help train observers and to reduce biases in visual estimates of distances by means of calibration. When no other methods are available to accurately measure distances to seabirds, visual estimates of distances, recorded by experienced observers and once calibrated with radar (or other ground-truthing methods), may be acceptable for different species under a wide range of environmental conditions. RESUMEN. Utilización de medidas de radar para la calibración de distancias a aves marinas migratorias estimadas visualmenteCensar aves marinas desdeáreas costeras requiere estimaciones fiables del número de aves y de la distancia de estas a costa. Restricciones logísticas hacen que a veces las estimaciones visuales de las distancias sean elúnico método factible para llevar a cabo estos estudios. Pusimos a prueba, en el Estrecho de Gibraltar, la exactitud de las estimaciones visuales desde costa, utilizando seis especies de aves marinas, concurrentemente, con medidas obtenidas por radar. La mayoría de las aves (91%) fueron detectadas dentro de unárea de 3 km de la costa y ajustamos nuestra calibración a esta distancia. Encontramos una fuerte correlación entre las estimaciones visuales y las del radar (R 2 = 0.83, P < 0.0001). La magnitud de los errores visuales al estimar las distancias fue moderada y variaron de 0.08 a 0.20 para diferentes distancias y observadores. Entre los factores que afectaron a la exactitud de las estimaciones visuales, se encuentran el observador, la especie de ave, la conducta exhibida por el ave y las condiciones meteorológicas. El modelo con mayor parsimonia incluyó la identidad del observador como elúnico vaticinador y ningún otro modelo, con más de un vaticinador, tuvo un valor de AIC menor. El radar puede ser utilizado para entrenar al personal y reduce el sesgo en las estimaciones visuales mediante la calibración. Cuando no hay disponible otros métodos para medir con exactitud la distancia a que se...

Bird Radar Validation in the Field by Time-Referencing Line-Transect Surveys

“…for mitigating bird collisions with turbines and aircraft). Further limitations of the method are related to the capabilities of field observers to correctly categorise the different distance and altitude classes, which may be difficult in the absence of visual landmarks [28], but is achievable by experienced observers [29].…”

“…Through storing the raw radar signal during the validation and reprocessing the track extraction with different parameters or tracking algorithms, the link algorithm can be run repeatedly on the same visual observation data set, allowing comparison of validation results for different tracker settings. Second, the method may be used to study the distance estimation capabilities of field observers [29], whose accuracy is highly relevant for survey techniques such as distance sampling [27]. Finally, time-referencing transect counts may be used as a general strategy for routinely linking large numbers of radar tracks to their respective species identity, thereby allowing species-specific studies using surveillance radars.…”

“…Although determining the position of animals in the field is generally difficult and prone to estimation errors [27]–[29], the moment of crossing a line transect is one of the few types of positional information that can be quantified routinely and accurately. Line transects can be easily defined in the field by observers looking towards fixed visual landmarks near the horizon, such as towers, trees, buoys or wind turbines.…”

Distance Sampling: Methods and Applications

“…Mateos et al 2010), bat or insect (e.g. Chapman et al 2002;Reynolds and Riley 2002) radar surveys for which the radar is placed on or near the ground and animals exhibit a strong non-uniform distribution with height.…”

“…The use of radar was proposed by Harmata et al (1999) to detect bird movement in the context of impact assessment. It has been used to estimate distances to birds (Petersen et al 2006), and in the context of a distance sampling survey to calibrate observer estimates of distanceto-bird measurements (Mateos et al 2010). Shamoun-Baranes et al (2014) propose the use of these systems to evaluate bird migrations at a continental scale.…”