TY - JOUR
T1 - CFSv2 Monthly Forecasts of Tornado and Hail Activity
AU - Lepore, Chiara
AU - Allen, John
N1 - Funding Information:
Acknowledgments. This work was partially supported by a Columbia University Research Initiatives for Science and Engineering (RISE) award, NOAA’s Climate Program Office Modeling, Analysis, Predictions and Projections Program Award NA14OAR4310185, and the Willis Research Network.
Funding Information:
This work was partially supported by a Columbia University Research Initiatives for Science and Engineering (RISE) award, NOAA's Climate Program Office Modeling, Analysis, Predictions and Projections Program Award NA14OAR4310185, and the Willis Research Network.
Publisher Copyright:
© 2018 American Meteorological Society.
PY - 2018/9/24
Y1 - 2018/9/24
N2 - AbstractClimate Forecast System, version 2, predictions of monthly U.S. severe thunderstorm activity are analyzed for the period 1982–2016. Forecasts are based on a tornado environmental index and a hail environmental index, which are functions of monthly averaged storm relative helicity (SRH), convective precipitation (cPrcp), and convective available potential energy (CAPE). Overall, forecast indices reproduce well the annual cycle of tornado and hail events. Forecast index biases are mostly negative and caused by environment values that are low east of the Rockies, although forecast CAPE is higher than the reanalysis values over the High Plains. Skill is diagnosed spatially for the indices and their constituents separately. SRH is more skillfully forecast than cPrcp and CAPE, especially during December–June. The spatial patterns of forecast skill for CAPE and cPrcp are similar, with higher skill for CAPE and less spatial coherence for cPrcp. The indices are forecast with substantially less skill than the environmental parameters. Numbers of tornado and hail events are forecast with modest but statistically significant skill in some NOAA regions and months of the year. Skill tends to be relatively higher for hail events and in climatologically active seasons and regions. Much of the monthly skill appears to be derived from the first 2 weeks of the forecast. El Niño–Southern Oscillation (ENSO) modulates forecasts and, to a lesser extent, forecast skill, during March–May, with more activity and higher skill during cool ENSO conditions.
AB - AbstractClimate Forecast System, version 2, predictions of monthly U.S. severe thunderstorm activity are analyzed for the period 1982–2016. Forecasts are based on a tornado environmental index and a hail environmental index, which are functions of monthly averaged storm relative helicity (SRH), convective precipitation (cPrcp), and convective available potential energy (CAPE). Overall, forecast indices reproduce well the annual cycle of tornado and hail events. Forecast index biases are mostly negative and caused by environment values that are low east of the Rockies, although forecast CAPE is higher than the reanalysis values over the High Plains. Skill is diagnosed spatially for the indices and their constituents separately. SRH is more skillfully forecast than cPrcp and CAPE, especially during December–June. The spatial patterns of forecast skill for CAPE and cPrcp are similar, with higher skill for CAPE and less spatial coherence for cPrcp. The indices are forecast with substantially less skill than the environmental parameters. Numbers of tornado and hail events are forecast with modest but statistically significant skill in some NOAA regions and months of the year. Skill tends to be relatively higher for hail events and in climatologically active seasons and regions. Much of the monthly skill appears to be derived from the first 2 weeks of the forecast. El Niño–Southern Oscillation (ENSO) modulates forecasts and, to a lesser extent, forecast skill, during March–May, with more activity and higher skill during cool ENSO conditions.
UR - https://doi.org/10.1175/WAF-D-18-0054.1
M3 - Article
SN - 0882-8156
VL - 33
SP - 1283
EP - 1297
JO - Weather and Forecasting
JF - Weather and Forecasting
IS - 5
ER -