Investigating Vegetation in the Phoenix Basin

A final goal to address concerns the significance of changes in biomass between drier and wetter years. The image above shows the difference between 1986 (dry) and 1992 (average) biomass values. The differences here are smaller than those obtained by comparing either of these years to 1993. However, I think that what differences there are between these two years is significant. Perhaps most importantly, there are some non- obvious differences being detected that vary spatially. Some stretches of the rivers are scoured of vegetation in linear patterns. Some vegetation in the mountains as well as in flat areas is actually less healthy in the wetter year, cross-cutting vegetation communities.

As a further exploration of the significance of inter-annual biomass changes, I derived a map of biomass stability, based on the variance for the three years in each cell of the study area. The figure above shows the results. The virtually unchanging green areas are primarily riparian vegetation, with some representation in low lying areas and along the ridges of the Estrella Mountains. West-facing slopes of all mountains, but especially in the Maricopas, are marked by extreme variation in biomass levels during the years examined.

Application

A final test is of the significance of this research. I intend for the results to be useful for examining landscape ecological issues, especially within an archaeological context. Although the results are still preliminary, and I have yet to extend my spatial scale from the pilot area to the CAP-LTER study area, I thought it might be interesting to examine a simple archaeological application. The map in the image below is a combination of biomass magnitudes (productivity) and biomass change over time (predictability). The white circles on the map are archaeological site locations, extending from early to late prehistoric times. If the results of this study are at all accurate, then site patterning should show a preference away from unproductive, unpredictable areas, and towards productive and predictable areas. The histogram below the map shows the number of archaeological sites in each of three levels of "attractiveness" described in the legend. There appears to be a preference for productive and/or predictable locations on the landscape. While 10% of the study area is classed as having healthy and stable vegetation, 20% of the sites are located in those areas. This is in spite of most of this class being located in river beds, where sites are unlikely to be located or found. The results of this project so far seem to be realistic, significant and useful. In the coming months, I will be focusing on expanding the spatial scale of investigation, develop a means of predicting riparian biomass, and find a way to deal with the severely urbanized portions of the study area.