Rhian Trueman’s first year as an Environmental Data Analyst at CSX has seen us making a breakthrough in turning bird acoustic data into measurable and monitorable biodiversity assessments. CSX’s acoustic monitoring program evidences changes in land management and supports nature recovery and Biodiversity Net Gain (BNG) reporting.
Our acoustic monitoring program enables us to gather baseline and monitoring data across sites. And for Rhian the CSX program also serves personal interest, allowing nature enthusiasts to explore and understand local bird diversity.
Bird biodiversity can be seen as a crucial indicator of ecosystem health. As the demand for biodiversity data is increasing, there is a need for repeatable and scalable bird surveys. At CSX, we measure and monitor bird biodiversity using acoustic monitoring devices, that record bird songs and calls. Once these devices are deployed in the field, they collect audio data which minimises any disturbance to habitats and species.
Acoustic monitors are deployed for 4 weeks and record at regular intervals throughout the day and night to ensure we are getting a full representation of the site at that time.
To develop the systems and service Rhian began with processing the extensive raw acoustic data into measurable and meaningful insights. This data came from different acoustic monitoring devices we were trialling, to test each of their capabilities. That analysis process allowed us to decide upon which of the sensors we are to use to ensure the best suited devices are in the field.
To convert the raw data into a user-friendly format, Rhian utilised advanced machine learning algorithms to identify species-specific bird songs and calls from the raw audio files. This allows us to measure any changes in species abundance and diversity over time. To ensure accurate and reliable species identification, Rhian tested various filters for the data to remove any human-induced noise and noise from factors including wind and rain. Removing this background noise allows clarity of the bird songs and calls clearly within the recordings and therefore increases the confidence of the results.
Schedule testing was a large part of the process to ensure that we maximise the time acoustic monitors are out gathering data in the field but also making sure the batteries and storage were sufficient. I created a schedule that corrects itself for sunrise and sunset each day according to the location to ensure that we are capturing times when bird activity is expected to be high.
As part of the data analysis, we calculate several acoustic indices which are a quantitative way of monitoring trends and changes in bird activity, species richness and species diversity. Acoustic indices add value to the data as they provide further information on the soundscape and ecosystem health on site.
Once I had refined the data processing and analysis methods, the next challenge was commercialisation to enable landowners and businesses to start evidencing their impact on biodiversity. This involved internal processes to efficiently tailor our acoustic monitoring program to different sites as they are all unique. Building a plan for deployment where there are different sized land parcels, different habitats and elevation levels was important as these are some of the factors that affect the detection distance of the acoustic monitors.
This journey has highlighted how technology is transforming biodiversity monitoring, and I am excited to see the future opportunities it brings to help CSX provide evidence to land and business owners of their contributions to conservation. CSX look forward to introducing our next metric, as we progress from taking bird acoustic data into measurable and monitorable biodiversity assessments and widen the ways in which we can monitor the results of Nature Based Solutions.