JRBP Bird Banding Station

Neotropical migratory birds connect biological systems across continents as they travel between North America, where they reproduce, and Central and South America, where they winter. Dramatic declines in these species in recent decades have been attributed to habitat fragmentation and loss (Robbins et al., 1989) and climate change (Holmes, 2007). At the Jasper Ridge Biological Preserve the riparian woodland provides breeding habitat to a disproportionate number of these species as compared to the other ecosystems within the preserve. Not only is the number of species it supports high, but the density of breeding landbirds in the JRBP riparian woodland is one of the highest ever recorded in North American (DeSante 1971 and 1972; Fee et al, 1996). 

 

Despite their importance for breeding landbirds, riparian ecosystems account for less than 0.5% of the landcover in California (Smith, 1977; RHJV, 2000). The loss of riparian woodland both in California and across Western North America is linked to the decline of more species in this region than any other anthropogenic factor (DeSante and George, 1994). The biological and anthropogenic dynamics of such ecosystems make the riparian woodland at JRBP a critical place to study bird populations. Additionally, the gradient of riparian woodland fragmentation that surrounds JRBP sets up conditions to study the effects of such fragmentation. 

 

This project aims to achieve the following goals: 

 

 

Banding involves the live capture, marking, and release of wild birds. An array of mist nets, each measuring 12 meters across and 3 meters high, are set up approximately 30 minutes before sunrise. These nets passively capture birds that fly into them. Nets are monitored by our research team and birds are removed upon capture for processing. After the species has been identified we place a small aluminum band around the bird’s leg. 

 

Each band has a unique 9-digit number issued by the Bird Banding Laboratory of the USGS. Once the band is on the bird additional data is taken, such as the age and sex, along with standard body measurements. After processing the bird is safely released.

 

 

 

 

Fig 1. An aluminum band with a unique 9-digit number issued by the USGS is placed around the leg of a Wrentit with specialized pliers. 

 

During the process of banding a large amount of data can be collected about each captured bird. At the most basic level, the species, age, sex, date of capture, and dates of re-capture can provide critical insights into the demographic processes underlying population growth, stability, and decline. 

 

During each morning of banding there is a certain proportion of a species’ population that will be captured. In between this first capture period and the next capture period there is some probability of survival among those captured birds. If a bird survives between two capture periods, we may recapture it. However, just because a bird is banded does not mean it will be subsequently recaptured. This presents the first challenge of quantifying survival: we need to distinguish the probability of recapture from the probability of survival. Luckily, there are many modeling methods that have been developed by population ecologists to deal with this challenge.

 

 

Fig 2. Graphical illustration of the key population parameters involved in CJS models. 

 

Once we have multiple periods of capture, we can create what is called a capture history for each banded bird. A capture history is made up of 1s and 0s where a 1 signifies a period in which the bird was recaptured and a 0 signifies a period in which the bird was not recaptured. For example, if we capture and recapture a bird in the first and second capture periods, respectively, and then do not recapture it in the third nor fourth capture period its capture history would be “1100”. 

With enough capture histories over a long enough time we can begin estimating demographic parameters like survival. There are many types of models that have been developed to use capture histories to estimate such parameters. One of the most commonly used models is called the CJS model. This model uses what’s called maximum likelihood, where it tests values between 0 and 1 to see which value is “most likely” to produce the observed capture history. 

You can see that the species identification and date of capture alone can be used in powerful ways. We can add complexity to our model by adding in covariates such as age, sex, body weight, etc. This allows us to look at how survival might vary across age classes (ie how does juvenile overwintering survival compare to that of adults?). 

Summary of Research Effort

Since 2018, over 2000 birds of 51 species have been captured at the JRBP banding station. The ten most commonly banded species are Bewick’s wren (221), song sparrow (213), chestnut-backed chickadee (201),  Wilson’s warbler (168), ruby-crowned kinglet (135), bushtit (129), wrentit (110), spotted towhee (103), common yellowthroat (97), and golden-crowned sparrow (68). A seasonal summary of captures for 2019 can be found below. The table is interactive and is organized with drop-down menus for each represented taxonomic family.