 |  |  |  |  |  |  |  |
Guidelines
Our survey is very simple. It is based on the repeat-survey methodology of MacKenzie et al. (see citations and additional
details at the bottom of this page). It is vital that everyone follow the same procedures so that data can be compared.
We are searching primarily for two species of owls in the region (eastern screech owls and barred owls), and secondarily
for two additional species (great-horned and saw-whet).
Briefly, a single survey is a matter of playing recorded owl calls for 10 minutes while listening for responses. These
species will almost always respond to a call because they are quite territorial. You then record whether or not you get a
response, that's it.
Here are exact instructions on how to conduct a survey:
So, we also ask all of our participants to do their best to spread out their surveys and, if you detect an owl on your
property, please do not start calling them in just for fun. There's no hard line as to spreading them out, just leave at least a
few days in between each one. Obviously a limit of 1 survey a month per species is arbitrary and a few more will
probably not harm the birds, but there is a point where calling will affect the owls and we would like to err on the side of
caution.
Below are additional guidelines for your surveys:
Thank you!!
Some Details on our Methods
If you are interested in the details of our analysis, please see:
OCCUPANCY ESTIMATION AND MODELING by MacKenzie, Nichols, Royle, Pollock, Bailey, Hines
or
Occupancy/Presence website
Here is a brief explanation:
Many times, biologists are looking for a certain species in a number of locations. So they go out and look for it, in any
number of ways appropriate to catching or observing that species (e.g., cameras, traps, nets, tracks, or call surveys).
Usually they find the species in some sites, and don't find it in others. However, in those places that they are not found, the
species could really be there but the biologist failed to find it. This means that whatever method they used (traps, etc.) is
imperfect and has a rate of failure.
Scientists can deal with imperfection and failure, if they can estimate what the rate of failure actually is. By doing
repeated surveys in a set of locations, you can estimate not only the places where the species definitely lives, but the
places where the species probably lives even if you didn't find it. This gives you a better estimate of the proportion of the
total sites you looked that have the species (e.g., "60% of the places you looked are likely occupied"). You can also
estimate how effective your method is (called the detection probability), e.g., "you have a 80% chance of finding the
species using call surveys if it is actually there."
Once these two measurements are estimated, you can relate both to habitat or other variables. Did your detection rate go
up (i.e., was the species easier to find) when it was warm/at night/etc.? Were they more often found in the woods or in the
city? Or, were they just easier to find in the city than in the woods? In our study we are also interested to see if the two
species are found together or do not overlap.
Repeated surveys at every location are necessary, even if you find the species right away. Without the repeated surveys,
you could not estimate detection probability.
Our survey is very simple. It is based on the repeat-survey methodology of MacKenzie et al. (see citations and additional
details at the bottom of this page). It is vital that everyone follow the same procedures so that data can be compared.
We are searching primarily for two species of owls in the region (eastern screech owls and barred owls), and secondarily
for two additional species (great-horned and saw-whet).
Briefly, a single survey is a matter of playing recorded owl calls for 10 minutes while listening for responses. These
species will almost always respond to a call because they are quite territorial. You then record whether or not you get a
response, that's it.
Here are exact instructions on how to conduct a survey:
- First, please register as a member of the program by just sending us an email with your name, and address at
owlcall@mianus.org This will allow us to keep in touch with you and send you any important information and
updates. Your address will be kept confidential.
- Download the call tracks for both species on this website and burn them to a CD or onto an ipod. Play them via
a boombox. The boombox you are using should be powerful enough to be used in an outdoor setting (any portable
players that take C or D batteries are fine). Note some older boom boxes can't play burned CD's.
- Set up your CD player so that it projects outward into an open area (e.g., not pointing at a wall). Use the same
spot for all 3 surveys. Facing a woodlot or other greenspace is better than facing into the road, but if you live in a
densely populated area, we are still interested in your results. Remember, these owls are all generally nocturnal, so
only do your surveys at night. Anytime after dark is fine.
- When you play the recorded calls, any owls in the area will likely respond to them. The wild owls will sound
very similar, if not identical, to the calls you are playing on your CD player. Occasionally, owls respond to calls
of another species, so you might want to listen to the tracks beforehand to get an idea of what to listen for.
- Volume should be set loud, but not so loud as to cause distortion in the speakers (if distortion is occurring you will
hear buzzing or "resonating" in the speaker output).
- Submit your data via the form linked on this website. If you have trouble with the form, give us a call.
- Only survey for a single species in one night. Even though (foolish?) owls sometimes respond to other species'
calls, it is much more likely that they will stay quiet when they hear another type of owl because the bigger species
often prey on the little ones (see below).
- If you detect an owl on your property, please do not start calling them in just for fun. Too much stimulation by
artificial calls can change the behavior of these birds. This is why we are limiting the number of surveys to 3 per
season per species. However, more frequent calling will disrupt the owl's behavior and the effectiveness of the
method itself. An owl who thinks its territory is being constantly "invaded" may leave the area or spend too much
time defending it and not enough time finding food or some other necessary behavior. It can also disrupt pair-bonds
and courtship. Try your best to spread out your surveys across the season.
So, we also ask all of our participants to do their best to spread out their surveys and, if you detect an owl on your
property, please do not start calling them in just for fun. There's no hard line as to spreading them out, just leave at least a
few days in between each one. Obviously a limit of 1 survey a month per species is arbitrary and a few more will
probably not harm the birds, but there is a point where calling will affect the owls and we would like to err on the side of
caution.
Below are additional guidelines for your surveys:
- Surveys for each species must be done on different nights
- Barred owls and great-horns sometimes eat the smaller screech and saw-whet owls, so if you play small owl
calls after large owl calls, the small owls will usually stay quiet so they are not found by the larger owl they
think is there.
- Similarly, barreds and great-horns may not call after they hear smaller owl calls because they are looking for the
small owl they think is there.
- We would also recommend separating your cross-species surveys by 2 or so days, for the same reasons.
- Listen for the entire 10 minutes of the track. In our experience, owls tend to respond around the 7 - 8 minute mark.
- You should conduct 3 surveys for each species each season. Even if you find an owl on the 1st or 2nd try, you
must complete all 3. This is a vital part of the analysis and helps us determine how often we tend to miss owls that
are really there (see details below).
- Use the same location for all 3 surveys. Again, in order to determine how often we miss owls that are really there,
the same location must be surveyed multiple times.
- The project is divided into 3 month seasons. Please perform 3 surveys for each species in one season. This adds
up to just 60 minutes per season. The seasons are Spring (March - May), Summer (June - August), Fall (September to
November), and Winter (December - February).
Thank you!!
Some Details on our Methods
If you are interested in the details of our analysis, please see:
OCCUPANCY ESTIMATION AND MODELING by MacKenzie, Nichols, Royle, Pollock, Bailey, Hines
or
Occupancy/Presence website
Here is a brief explanation:
Many times, biologists are looking for a certain species in a number of locations. So they go out and look for it, in any
number of ways appropriate to catching or observing that species (e.g., cameras, traps, nets, tracks, or call surveys).
Usually they find the species in some sites, and don't find it in others. However, in those places that they are not found, the
species could really be there but the biologist failed to find it. This means that whatever method they used (traps, etc.) is
imperfect and has a rate of failure.
Scientists can deal with imperfection and failure, if they can estimate what the rate of failure actually is. By doing
repeated surveys in a set of locations, you can estimate not only the places where the species definitely lives, but the
places where the species probably lives even if you didn't find it. This gives you a better estimate of the proportion of the
total sites you looked that have the species (e.g., "60% of the places you looked are likely occupied"). You can also
estimate how effective your method is (called the detection probability), e.g., "you have a 80% chance of finding the
species using call surveys if it is actually there."
Once these two measurements are estimated, you can relate both to habitat or other variables. Did your detection rate go
up (i.e., was the species easier to find) when it was warm/at night/etc.? Were they more often found in the woods or in the
city? Or, were they just easier to find in the city than in the woods? In our study we are also interested to see if the two
species are found together or do not overlap.
Repeated surveys at every location are necessary, even if you find the species right away. Without the repeated surveys,
you could not estimate detection probability.
| "Who's Whoo-ing" In Your Backyard? A Citizen Science Study on Suburban Owls |