98th ESA Annual Meeting (August 4 -- 9, 2013)

OPS 4-9 - Imputation of Statewide FIA and nearby data for developing local forest and ecological inventory detail

Thursday, August 8, 2013
Exhibit Hall B, Minneapolis Convention Center
Alan R. Ek, David C. Wilson and John Zobel, Forest Resources, University of Minnesota, St. Paul, MN
Background/Question/Methods

Developing and improving the detail of forest inventories for public agencies and large landowners is expensive.  This is especially true for management inventories that seek to describe tree and stand conditions, especially those conditions of an ecological nature such as provided by ecological classifications.  Given that, efforts to update and improve those databases, especially for ecological descriptions, often falls off the action list.  The result is often outdated and simplistic databases that are suboptimal for numerous forest management operations and local to forest wide planning.

These management inventories typically consist of two components: (1) mapping the forest stand or ecological unit and (2) observation of tree data on field plots—usually the most expensive component.  However, in Minnesota the USDA Forest Service Forest Inventory and Analysis (FIA) program regularly measures more than 5,000 field plots and these are measured and classified in a very similar manner, but more detailed, than most operational inventory plots.  Additionally, significant portions of the state have been mapped by their ecological classification and these data can be linked to FIA and/or operational inventory data.   Given the similar plot and stand description data for the FIA and operational inventories, this study examines the effectiveness of imputation of FIA plot and ecological classification data to supplement and/or replace operational field plot data thereby greatly reducing the cost, frequency and time of inventory field plot measurement efforts. 

Results/Conclusions

Results to date suggest considerable savings are possible with little loss of precision for numerous forest management operations and to address local to forest wide and long term planning.  Finally, examples of imputation and data usage in operations and planning are presented for sample counties in Minnesota.