Summary of 2012 data

Students at Hanshew and Central middle schools, East and West High schools collected data within plots of Anchorage area urban forests.  They counted available browse and bites for desirable and undesirable browse species, and classified browse intensity based on commonly used tree architecture classifications (broomed, browsed and unbrowsed).    The goal of data collection is to determine if Anchorage area moose are eating wild bird cherry trees (Prunus padus or P. Virginiana) in comparison to trees considered desirable browse and alders.  Data used in this analysis were collected at 9 sites and include 134 trees.  Some data were inadvertently not collected by students often in the category of browse intensity.  A small sample size (5) of aspen were observed in all plots and were not included in analysis.  Table 1 shows the sample size (n) by species.  In class students graphed and compared means to interpret differences in browse of tree species by moose.  The summary below expands on that using an Analysis of Variance to determine if the observed differences are significant. 

SpeciesBrowsed ratio (n)Browse Intensity (n)
Alder15  15
Birch3217
Cottonwood2525
Cherry2624
Willow3326

Anchorage Moose seem to prefer traditional browse species over wild cherry trees based on proportion browsed and browse intensity (aka architecture) data.

There is a significant difference in browse/availability ratio between species (ANOVA Kruskal-Wallis p= 0.0009).  The significant differences are between Willow and Alder (Tukey HSD p=0.0036) and Willow and Cherry (Tukey HSD p=0.0060).  All other comparisons do not show significant differences of the browse/availability ratio between species.

Browse intensity is defined as; 0=unbrowsed, 1=browsed and 2=broomed.  There is a significant difference in browse intensity between species (ANOVA Kruskal-Wallis p< 0.0000).  The significant differences (calculated using Tukey HSD) are between Willow and Alder (p=0.0039), Willow and Birch (p=0.0234), Cottonwood and Cherry (p=0.0082), Willow and Cherry (p=0.0002).  All other comparisons do not show significant differences of the browse intensity between species.

Even though there appears to be a preference for certain species over others based on the proportion of browse removed and the browse intensity there is no significant difference between species based on the number of bites taken per species (ANOVA Kruskal-Wallis p=0.1207 see figure below).  No difference in number of bites by browse species indicates that moose are biting “less” desirable browse (alder and cherry) when available.  Care should be taken to not over assume this means moose diet contains as much cherry as other species because quantity consumed is best determined using bite diameter data to calculate mass consumed and analyzing content of fecal pellets.  While bite diameters were measured in this project they were not used because the data was not measured reliably, and some calipers failed 

From the data collected in 2012 it appears that moose do indeed eat wild invasive cherry trees within the sample area.  While moose eat less cherry than highly preferred browse species (willow and cottonwood), moose do eat the cherry.  Personal observations of moose browsing cherry outside of the Anchorage bowl indicate that the observed browse is not localized to a few Anchorage moose “learning” to eat the tree.  Moose in general are likely eating cherry to fulfill their daily dietary intake requirements, and eating less of it than other species because it is novel and/or contains cyanogens.  Further work will seek to determine how the concentration of cherry trees in a plot affects browse behavior of moose.  Toxicological and nutritional analyses of Anchorage area trees should be completed to explain moose browse behavior with changing concentrations of cherry. 

Back to Top