Sedgwick Reserve

01 Jul, 03 AM - 30 Jun, 09 AM

The proposed research on Figueroa Creek is part of the three-reserve study of the populations of the stream-dwelling water strider, Aquarius remigis on small, isolated natural streams. The goal of the study is to assess the ?effective? population size (Ne) of these natural populations and determine its relationship, if any, to the intensity of sexual selection occurring in these populations. The ?effective? population size is a measure of the potential of the populations to persist and adapt to possible environmental changes. Effective population sizes are generally much smaller than census population sizes, and the extent of this difference is roughly proportional to the variance in reproductive success among breeding individuals and the magnitude of fluctuations in population size from generation to generation. We will assess Ne, variance in mating success, census population sizes, and standing levels of genetic variation over six consecutive generations in three populations, each on a separate UC reserve. Previous assays of sexual selection and genetic variance in similar Californian populations suggest that these small, isolated populations may be at risk of long-term loss of genetic variation and eventual extinction. The proposed research will test this hypothesis. The basic sampling design will follow that used successfully to estimate population parameters in eastern populations of A. remigis. We will survey a 100 m stretch of stream selected to represent the typical mix of habitats (e.g. pools, riffles, waterfalls) for that stream. On our preliminary sampling trip in April 2002, we identified a suitable section of the east branch of Upper Figueroa Creek, extending upstream from its confluence with the west branch. A standard capture-recapture protocol will be used to estimate adult population size (N) and operational sex ratio (OSR) along this stretch of stream. A team of 3 - 4 persons will use hand nets to capture all extant adults in the population on every visit. Our daily capture success using this technique has averaged 75% in previous studies. Thus, the probability capturing a given individual at least once in four successive sampling days should be 1 - 0.254 = 0.996. Such a high capture probability yields very precise estimates of population parameters, particularly when survival of individuals from one sample to the next is also high. For example, with a daily capture probability of 0.75 and survivorship of 0.9, the standard error of N estimated using the standard Jolly-Seber model is expected to be less than 2% of the parameter value. Our sampling plan for each year comprises a series of four well-spaced sampling periods, each sampling a different generation of seasonal stage. All adult animals within the study area will be captured each day, and any unmarked animals will be individually marked by painting numbers on their thoraxes with enamel paint. Extensive evaluation of this technique in the laboratory and in natural populations has revealed no adverse effects, and the marks are permanent. The sex and wing morph of each individual will be recorded, and all individuals will be released where captured. The sampling periods are designed to sample the post-diapause breeding population (late March), the F1 breeding population (late May), and the F2 pre-diapause population (August and October). Dates are given only for 2004 but a similar schedule will be repeated up to August 2006. In addition to the capture-recapture work, we will estimate the mating frequency of individual males by scan sampling. These data will be used to estimate the standardized variance in mating success (Vk/k) and the opportunity for sexual selection (Vk/k2; where k = mean number of matings and Vk = variance in number of maintgs). The basic protocol is to simply walk slowly up and down the stream bank, recording which marked animals males are mating and which are single. During intromission, males remain mounted on the back of their mates and the mating pair typically remains in tandem for several hours. The daily capture sequence will provide four additional estimates of mating success. Thus, mating activity will be recorded seven times during each sampling visit: once on sampling day 1, and twice on sampling days 2, 3, and 4. To estimate Ne and standing neutral genetic variation, we will type A. remigis for 15-20 microsatellite loci. To minimize the impact of the genetic sampling on the subsequent population parameters, we will collect 75 third-fourth instar nymphs during the summer sampling visit each year. We expect densities to be near their maximum at this time of year, at the beginning of the diapause period and before food becomes critically limiting. Our removal sample will thus represent only a very small portion of the extant nymphal population. Further, because the nymphs compete with other nymphs and adults for limited food, and cannibalism is common, nymphal survival through the summer and fall is expected to be density-dependent. Thus, we expect that the slight increase in nymphal loss caused by our sampling will be compensated for by increased survival of competing nymphs. Nymphs will be preserved in 99% ETOH and shipped to the University of Windsor for genetic analysis. We will assess standing genetic variation for quantitative genetic traits using a standard half-sib mating design, and four morphological measurements. To obtain parents for the half-sib experiment, we will collect eggs from 50 females from each population during the spring sampling trip each year. On the last day of capture-recapture sampling, 50 females will be selected at random from the marked population and placed singly in plastic buckets anchored to the stream bottom (an average of 5 buckets within each10 m sampling area). The buckets have screened openings that allow passage of stream water, but are covered to prevent escape of the striders. The females will be fed previously-frozen Drosophila melanogaster and provided with scored foam cups for oviposition sites. The buckets will be left in place for 48 h, after which time all females will be returned to their populations where caught. We hope to collect a minimum of 10 eggs per female. If egg yields are insufficient in any sample, females will be left in the buckets for an extra day. The foam cups with attached eggs will be returned to the laboratory, and the eggs will be reared at 20C and 14hL:10hD. Previous experience indicates an 85% survival rate from egg to adult under these conditions, and we therefore expect a minimum of 8 F1 adults per female (= 300 per population). Parents for the half-sib experiment will be selected from these F1 adults. As for the sampling for microsatellite analysis, this method of obtaining sires and dams for quantitative genetic analysis minimizes the demographic effects on the source population. In summary, the proposed research consists of four bouts of capture-recapture sampling combined with daily scan samples, and periodic samples of eggs or nymphs for genetic analysis. The sampling scheme, spread over three years and six generations, is predicted to have no effect on the long-term dynamics or genetic structure of the population of A. remigis on Figueroa Creek, or upon any other aspect of the creek community. The project will benefit the UC Reserve system by providing detailed information about the ecology and population genetic structure of A. remigis, and highlighting the importance of the reserve system for conservation of endangered populations in the fragmented California landscape.

Visit #1575 @Sedgwick Reserve

Approved

Under Project # 1296 | Research

Assessing sexual selection and effective population size in natural populations of Aquarius remigis.

faculty - University of California, Riverside

Reservation Members(s)

Group of 2 Research Assistant (non-student/faculty/postdoc) Jul 1, 2003 - Jun 30, 2004 (366 days)
Group of 2 Faculty Jul 1, 2003 - Jun 30, 2004 (366 days)

Reserve Resources(s) | Create Invoice

Studio Apartment 4 Jul 1, 2003 - Jun 30, 2004