PRISSM butterfly surveys were initiated in February 2015 in order to track local butterfly in different habitats populations over time. The bio-monitoring protocol for butterflies was designed to collect data on butterfly phenology or relative densities throughout the season (the number of individuals recorded within an interval of space and/or time) as well as species richness. Over multiple years, we expect that this approach will provide an inventory of butterfly species at each site. In addition, the protocol allows a researcher to identify habitat preferences of the various species.
Currently only the Bernard Field Station is conducting monthly butterfly surveys, but we hope that other sites participate in the future.

When PRISSM was initiated, PRISSM workshop participants wanted to include an invertebrate component in the California Sage Scrub (CSS) monitoring program. In considering the variety of methods and the myriad invertebrate taxa that might be used to monitor invertebrate communities in CSS fragments, PRISSM workshop participants specified the following criteria for taxon selection:

  1. Members of the taxon can generally be easily identified to species.
  2. The taxon can be sampled using methods that do not include extensive processing efforts (e.g., pitfall traps).
  3. The taxon can engage broad public participation.

Because of these criteria, some invertebrate groups commonly used in bio-monitoring programs were excluded. Native bees, for example, are hyper-diverse and require trapping and considerable expertise to identify to species, and ground-dwelling invertebrates, such as ants and carabid beetles, typically require intensive processing and a taxonomic specialist. PRISSM consequently decided to monitor butterfly diversity for four main reasons:

  1. Butterflies are appealing to students and the general public.
  2. Butterflies have been used as indicator species in multiple systems (Daily & Ehrlich 1996), despite it being unclear if patterns can be generalized across taxa (Ricketts et al. 2002).
  3. There are good data on Southern California butterfly species, and many species are easy to distinguish from one another by sight.
  4. Butterflies can easily be monitored by modified “Pollard Walk” – a protocol used by many butterfly monitoring programs. For example, PRISSM’s approach is nearly identical to the protocol for the Illinois Butterfly Monitoring Program (

PRISSM Butterfly Monitoring Protocol:
It is critical that the protocol for butterfly monitoring is strictly followed to provide comparable data on relative abundances among CSS fragments.

  • Habitat designation:
    Prior to implementation, managers/researchers need to establish a permanent census route (see Pollard & Yates 1993) that travels through each of the major habitat types present at the site. The portions of the trail that are in different habitat types are called “transects”.

    The habitat type of each transect needs to be defined a prioi by the researchers (e.g., intact CSS, degraded CSS, non-native grassland, burned CSS, Chaparral). Classification of portions of the trail into different transects will require discussion by experts prior to initiation of butterfly surveys and may require some plant surveys.

    Classification of a particular habitat type typically uses the minimum requirement that 25% or more of the plants detected along transect/s are typically associated with a particular habitat type (Sawyer and Keeler-Wolf 1995). If more than 25% of plants are associated with two habitat types, this should be considered an ecotone.

    Much of Southern California has experienced some level of disturbance or invasion by exotic grasses (Keeley 2005, Wolkovich et al. 2010); consequently, habitats are only designated as non-native grassland if no other vegetation type constitutes 25% or more of the linear vegetation of a transect, typically determined by the point-intercept method (see Matsuda et al. 2011). Using this system we can also classify CSS sites by level of non-native grass incursion using broad categories:

    • Non-native grassland = >75% non-native grass
    • Heavily degraded CSS = 75-50% non-native grass
    • Degraded CSS = 50-25% non-native grass
    • Intact CSS = < 25% non-native grass


  • Route selection:
    In addition to traversing major habitats, the path chosen should take advantage of existing paths when possible, be easy to locate and follow by all participants, and pass close to important host and nectar plants. The path chosen should take at least 1 hour and no more than 3 hours to complete. Transitions from one habitat type to another and from intact to degraded areas should be clearly delineated.

  • Timing of surveys:
    Surveys are carried out once a month. Because butterflies fly nearly year-round in Southern California, we recommend that surveys be conducted between February through October when possible. Additional months can be added if additional resources are available. Surveys should be carried out at approximately the same week each month.

    Surveys should initially be conducted between 8 AM and 3 PM, depending on season and weather. Ideally, surveys are conducted when temperatures are warm (≥65°F), but less than 95°F, with little to no wind, no precipitation, and <50% cloud cover. We realize, however, that weather patterns and observers’ schedules may mean that ideal conditions cannot be achieved in a particular month. It is better to carry out the survey in less than ideal conditions than to not conduct the survey at all.

  • Observers:
    Each monthly census requires two individuals: (1) a monitor, who spots and identifies butterfly species, and (2) an assistant who records data, helps spot butterflies, and may help with identification. All lead monitors will need to spend the appropriate time to familiarize themselves with the local butterfly fauna. If any other participants join the survey, they must be “silent” and not contribute observations to the count.

  • Data collection:
    The observers record the time they begin the survey, as well as the times when the observers transition into a different transect (i.e., habitat type), and the time of the end of the survey. They should proceed along the route at a slow, steady pace (approximately 1 mph). All butterflies within 6 m (approximately 20 ft) to either side of the census route are counted. Butterflies outside this corridor are not included in the survey. When the observers spot a butterfly, they should identify the butterfly to the to the lowest taxonomic level of which they are confident. If an observed butterfly cannot be identified to species with high (95%) certainty, then the observation should be recorded to the next highest taxonomic level. For example, a butterfly might be recorded as “funereal or mournful duskywing” or “unidentified sulphur”. With experience and repeated field verification, observers should be able to recognize butterfly species from other more subtle characteristics (hue, flight pattern, etc.) than detailed diagnostic field marks. One or both observers should carry binoculars that allow for close-up focus to aid in identification.

    When a butterfly is spotted and identified, record in the “Notes” whether it was flying, on the ground, or on a plant. If it’s on a plant, record the plant species and whether it is nectaring, ovipositing, or resting. Also note caterpillars.

  • Photography and specimen collection:
    Photography is encouraged, and PRISSM recommends that observers photograph as many individuals as possible to confirm species identification. Collection of specimens should be minimized, but in some instances, it may be necessary for the monitor to collect the individual for identification either in the field or further identification in the lab.

Additional Materials:
Click on these links for:


  • Daily G.C., & P.R. Ehrlich 1996. Nocturnality and species survival. Proceedings of the National Academy of Sciences 93: 11709–11712.
  • Keeley, J.E. 2005. “Fire as a threat to biodiversity in fire-type shrublands.” United States Department of Agriculture Forest Service Gen. Tech. Rep. PSW-GTR-195.2005.
  • Matsuda, T., G. Turschak, C. Brehme, C. Rochester, M. Mitrovich, & R. Fisher. 2011. Effects of large-scale wildfires on ground foraging ants (Hymenoptera: Formicidae) in Southern California. Environmental Entomology 40: 204-216.
  • Pollard, E., & T.J. Yates. 1993. Monitoring Butterflies for Ecology and Conservation. Chapman & Hall, London, UK.
  • Ricketts, T.H., G.C. Daily, & P.R. Ehrlich 2002. Does butterfly diversity predict moth diversity? Testing a popular indicator taxon at local scales. Biological Conservation 103: 361–370.
  • Sawyer, J. O., & T. Keeler-Wolf. 1995. A manual of California vegetation. California Native Plant Society, Sacramento, USA.
  • Wolkovich, E.M. 2010. Grass invasion causes rapid increases in ecosystem carbon and nitrogen storage in a semiarid shrubland. Global Change Biology 16: 1351-1365.