Back
Predicting Nitrogen Deposition to Forests in the Los Angeles Basin
using Lichen Communities
Sarah Jovan1, Jennifer Riddell2 and Pamela E. Padgett2
Forests in the Los Angeles Basin receive the highest known levels of nitrogen (N) deposition in the
United States. Excess N is implicated in a wide variety of detrimental ecological impacts to both
terrestrial and aquatic systems, leading to shifts in vegetation communities that favor invasive
species, elevated nitrate (NO3) runoff, soil acidification, decreased frost-hardiness in trees, and so
on. In 2008 we surveyed the epiphytic (tree-dwelling) lichen communities of Quercus kelloggii
forests at 23 sites across the San Bernardino Mountains, the Palomar Mountain area (Cleveland
National Forest), and the Sawmill Mountains (Angeles National Forest). We employed gradient
analysis to determine how lichen community patterns relate to N measurements collected at some
or all of our survey sites; these include throughfall N (kg ha-1 yr-1), seasonal averages of NH3, NO2,
and HNO3 from Ogawa passive monitors (µg m-4), modeled site N deposition (kg ha-1 yr-1) from the
Community Multi-scale Air Quality model (CMAQ), and assays of NO3- accumulated on twig
surfaces (µg cm-2). With non-metric multidimensional scaling ordination (NMS) we resolved a
gradient explaining almost half the variability in lichen communities (r2 = 0.48), which clearly
reflected a community-level response to N. Lichen community scores along the gradient correlated
exceptionally well with throughfall N (r2= 0.94), an N measure that captures the hydrologic flux of
ammonium (NH4+) and NO3- ions from the tree canopy to the forest floor. We then used simple
linear regression (SLR) on community scores to predict throughfall N at all sampled sites. The twophase
model combining NMS and SLR yielded N predictions sufficiently accurate (error: ± 4.57 kg
N ha-1 yr-1) for use in the Basin where throughfall N spans 6.1 to 71.1 kg ha-1 yr-1. The ability to
make reasonably accurate N predictions based solely on lichen community information marks a
significant utilitarian advancement in the lichen-bioindication field. In the highly N-compromised
Basin, land managers and air quality regulators may use lichen estimates as a surrogate or
justification for implementing more costly campaigns that measure N directly. For instance, lichens
predict throughfall N in excess of 30 kg ha-1 yr-1 for most sites in Palomar and the Sawmill Mountain
Ranges, an amount that far exceeds all published N critical loads for California forest ecosystems,
yet N is not actively monitored in either of these forests.
1USDA Forest Service, Forest Inventory and Analysis Program 620 SW Main, Suite 400, Portland, OR,
97205, 503-808-2070;
2USDA Forest Service, Pacific Southwest Research Station, 4955 Canyon Crest Drive, Riverside, CA 92507