Search

Environmental impact assessment (EIA) follow-up monitoring in energy projects, such as wind farms, solar facilities and power lines, frequently encompasses bird fatality surveys based on regular carcass searches. Since a large number of bird carcasses are often not found, field experiments to determine carcasses persistence (CP) and searcher efficiency (SE) biases in fatality surveys are typically performed for every single project. These data are, however, rarely used beyond the scope of each original project. We compiled data from CP and SE trials performed under 36 independent monitoring programs aiming to assess bird mortality at transmission lines in Portugal, whose results were until now unavailable or dispersed in grey literature. We used survival analysis and generalized linear mixed-effects models to investigate the ecological and methodological factors influencing CP times and SE rates. Bird carcass size was a key driving factor of both CP times and SE rates of human observers, which were consistently out-performed by scent detection dogs. Season and habitat interactions had also an important role in CP patterns, while variations in SE rates by human observers were largely influenced by ground visibility (i.e., a combination of ground cover and vegetation height). Our results reinforce previous studies indicating that CP and SE biases are site-specific and determined by a wide range of ecological and methodological factors not always accounted in standard trial designs. Overall, our study demonstrates that data routinely collected under bird monitoring programs from multiple projects can be combined to identify broad ecological patterns, limitations of current studies and, ultimately, improve EIA follow-up practice. • Data from EIA follow-up monitoring is rarely used beyond each original project. • We compiled bias-correction data of bird fatality surveys in multiple transmission lines. • Bias-corrections were affected both by ecological and methodological factors. • Despite the broad patterns, bias-correction values were often site-specific. • Monitoring data from multiple projects can be combined to inform EIA practice. [ABSTRACT FROM AUTHOR]