Occurrences from Kenai National Wildlife Refuge’s Long Term Ecological Monitoring Program, 1999–2006

Study Extent

The geographic coverage included all USDA Forest Inventory and Analysis plots on the Kenai National Wildlife Refuge.

Sampling

Morton et al. (2006) and Morton et al. (2009) provided details of sampling methods. Vegetation data on forested plots were collected as part of the USDA Forest Inventory and Analysis Program (FIA) by the USDA Pacific Northwest Research Station (PNWRS). Vegetation data on non-forested plots were collected by Kenai National Wildlife Refuge (KNWR). Bird occurrence data on all plots were collected KNWR.

Method steps

1. Vegetation sampling of forested sites by FIA: The PNWRS sampled vegetation on 176 forested plots following field protocols used in the Coastal Alaska Inventory (USDA Forest Service, Pacific Northwest Research Station, & USDA Forest Service, Region 10 - Alaska, 2003). Consistent with national Phase 2 protocols, each FIA site consists of four subplots to measure within-site variance. Each subplot consists of three nested fixed-radius circular plots: 7.3-m tree plot, 5.64-m horizontal-vertical (HV) vegetation profile, and a 2-m seedling/sapling plot. Three 11.28-m transects intersect the center subplot to estimate downed wood. National and regional FIA field protocols were generally similar except for relatively minor differences in measurement units and location of subplot sampling. The one obvious and significant discrepancy was the exclusive sampling of HV plots as part of the Coastal Alaska Inventory. Horizontal-Vertical (HV) Plot: A 5.64-m radius (horizontal distance) circular plot was established at point center. The HV plot was used to determine the horizontal and vertical distribution, density, diversity, and composition of plants and non-living material. Vegetation was classified into strata or layers starting at ground level such that this plot represented a column with a base of 100 m2. Vertical dimensions were estimated using the natural layer breaks observed on the plot. One HV profile plot was established for each polygon type that had a point meeting the following criteria: the polygon type at point center is vegetated (i.e., shrub, herbaceous, forest); the HV plot falls entirely within the polygon at point center (i.e., does not straddle polygon types); the polygon type at point center covers ≥ 90% of the area; and the point center does not fall in an inclusion. Data collected on HV plots include shrubs, grasses, forbs, lichens, mosses, tree seedlings (trees < 2.5 cm DBH), and in some cases shrub-like trees not recorded on the tree plot. For example, Krummholz and other stands where trees were growing in a twisted/stunted shrub-like form, were treated as shrub land. The HV record also included arboreal lichens, mosses, forbs and shrubs regardless of their substrate. The data we used here we had downloaded circa 2004 from the Forest Inventory and Analysis Database (Miles et al., 2001; Breazeale et al., 2008).
2. Vegetation sampling of non-forested sites by KNWR: Vegetation Sampling of Non-forested Sites by KNWR: We did not use the FIA cluster sampling design, where four subplots are surveyed at a site. Instead, we sampled one plot, centered on the plot center and used sampling methods which better characterize non-forested vegetation. Our objectives were to quantify the relative frequency of ground cover and species within different height strata at each site. Similar to the HV plot sampled by the FIA, we also recorded all vascular and nonvascular flora on a 100 m2 circular plot with a 5.64 m radius from plot center. Voucher specimens or unknown plant material were collected at the site, but from locations > 10 m from plot center. We also took two stereo digital photographs on the north-south axis of each plot similar to PNWRS protocols for the HV plot. Herbaceous and woody vegetative cover within the first 2 m above ground were sampled using a modified point-intercept technique. A version of this sampling approach has been adopted by Denali National Park. Four 10-m long transects were laid out in the cardinal directions from plot center using tightly-woven, braided nylon cord marked at 0.5 m intervals (n = 20 points per transect). A 2 m long, 13 mm diameter steel rod, marked at 1 m to separate the vertical column into two strata, was used as a sampling pin that is held vertically at each sampling point. Each plant taxon that touches the sampling pin ≥ 1 time was recorded within stratum at each point. Only one hit per point per stratum per taxon was recorded at each of the 80 sampling points. Consequently, a minimum of 40 tallies were recorded per stratum; the number of tallies recorded can exceed this value considerably depending on species richness.
3. Bird sampling by KNWR: We sampled landbird abundance and occurrence using variable circular plot methods during the last three weeks in June. We adopted (with some modifications) the protocols used in the Alaska Landbird Monitoring Survey (Handel & Cady, 2004), where horizontal distances to each bird were estimated at 1-min increments during a 10-min sampling interval using auditory or visual cues. Surveys were conducted 30 min after sunrise during the first 4 - 5 hrs of the morning, but only under specified conditions of good visibility, little or no precipitation, and light or no winds. We used a laser range finder to help with distance estimation and recorded unknown or questionable songs/calls with a Sony digital Walkman™ MZ-N10 minidisc recorder and a Saul Mineroff Electronics™, Inc., ATR55 mini-microphone boom. Weather and noise: We measured wind speed (m/sec averaged over 30 sec), temperature (°C), and relative humidity with a Kestrel 3000 Pocket Weather™ meter at ground level. We used the Beauford wind scale to record wind speed at canopy level. We used the Larson Davis Model 720 sound meter to measure sound levels (Leq, Lmax and Lpeak) over a 5-min interval while birds were being counted. The sound meter was mounted on a tripod, oriented with the microphone pointed skywards, and placed ≥ 3 m from either observer to reduce recording of incidental noise.
4. Reformatting data for GBIF: Occurrence data were downloaded from Morton et al. (2021), reviewed, edited, and reformatted using an R markdown script (Bowser, 2022).

Vascular plants and birds were generally identified to species. Some non-vascular plants were also included.

1. Plantae
   common name: plants rank: kingdom
2. Aves
   common name: birds rank: class

The geographic coverage included all USDA Forest Inventory and Analysis plots on the Kenai National Wildlife Refuge.