Go to NPS Home
Sequoia and Kings Canyon National Parks

Fire Information Cache
Navigation Bar

BACK TO CACHE HOME

Fire Mgmt. Policy

  • SEQ./KINGS CANYON
  • NAT. PARK SERVICE
  • THREE RIVERS FIRE SURVEY
  • SEKI FIRE MGMT. PLAN & EA

    Fire and Park Resources

  • ONLINE PAPERS
  • BIBLIOGRAPHY
  • CURRENT RESEARCH
  • RESEARCH NEEDS

    Annual Fire Reports

  • INTRODUCTION
  • 2000
  • 1999
  • 1998
  • 1997
  • 1996
  • 1995

    CURRENT FIRE INFORMATION

    FIRE MAPS

    GIS INFO

    SNEP REPORT

    FIRE GLOSSARY

    FIRE LINKS


    BACK TO CACHE HOME

    SEKI NAT. RESOURCES HOME

    SEKI FIRE MGMT

    MAIN VISITOR CENTER

    NPS HOME


    FireNet Home Page National Park Service Home Parkwide Natural Resources Site, National Park Service
    get acrobat

  • Twentieth-Century Fire Patterns on Forest Service Lands


    Kevin S. McKelvey
    U.S. Forest Service
    Redwood Sciences Laboratory
    Arcata, California

    Kelly K. Busse
    U.S. Forest Service
    Redwood Sciences Laboratory
    Arcata, California

    ABSTRACT: Maps of twentieth-century fires on Forest Service lands were analyzed. Time trends showed no overall trend in acreage, but human-caused fires decreased and lightning fires increased. The increase in lightning fire was dominated by two recent years (1987 and 1990), but more subtle trends prior to 1987 indicated that lightning fires were following a trajectory separate from that of human-caused fires. Landscape-level analysis indicated a strong and stable elevation gradient in burn frequency, and this allowed the development of an accurate descriptive model.

    An analysis relating fire frequency to vegetation type showed that certain types of vegetation burn more than expected given their elevation, but that burning within these types followed the general trend, with higher-elevation types burning less frequently. An analysis of reburn patterns showed that, given a particular risk zone, fire location is nearly random. Acreage that burned more than three times had a greater burn frequency than would be expected if the fires were random, but the total area with multiple burns was tiny. The location of multiple-burn sites indicated that they were associated with special features such as busy roads.

    Fire correlations with general weather indices were weak, but more area burned in hot, dry years. Perhaps more importantly, all of the extreme fire years occurred when it was hot and dry. Short-term (1979-89) analyses of drought patterns indicated that drought decreased with increasing elevation, paralleling the decrease in fire frequency. Periods of drought were highly synchronized between weather stations, increasing the window during which extreme fire events could occur at all elevations.

    The strong and stable elevation trends in fire frequency indicated that future risk could be inferred from twentieth-century fire patterns. The largely random location of fires within each risk zone indicated that a general zonal strategy for fire control would be most effective. The fire acreage patterns over time indicate that while suppression and possibly education can reduce human-caused fires, large lightning fires will continue to occur. Overall risk will vary with weather; but we can expect that large fires will occur during future droughts. If the weather in the twenty-first century is similar to weather in the twentieth century, we might reasonably expect 40% to 60% of the foothills zone to see fire at least once in the next 100 years.


    ________________________________

    Fire Management Office Main Visitor Center NPS Home