NCEP’s EMC provides a CFSv2 monthly forecast climatology that matches the start times of operational seasonal forecasts.
Operational forecasts of 2-m temperature that were initialized in January and February of 2017 are used to illustrate the effect of the EMC forecast climatology bias on forecast anomalies. Operational CFSv2 seasonal forecasts began in early 2011 and are initialized every day at 6-hourly intervals, which means that there are no hindcasts with matching start times for most operational CFSv2 seasonal forecasts (e.g., there are no seasonal hindcasts initialized on 2–5 January). CFSv2 seasonal hindcasts have initializations on every fifth day starting from 1 January (not counting 29 February in leap years) at 6-hourly intervals (0000, 0600, 1200, and 1800 UTC all start times here are UTC, and we omit the time zone). We use 2-m temperature hindcast data from 1982–2010 and SST hindcast data from 1999–2010. The CFSv2 variables examined here are 2-m temperature and the Niño-3.4 index computed from sea surface temperature (SST). A summary and conclusions are given in section 5. We propose two alternative methods for computing the CFSv2 monthly forecast climatology that depend on fewer parameters and that better fit the hindcast data. Section 4 examines bias in the EMC monthly climatologies of near-surface temperature and the Niño-3.4 index and relates those biases to the fitting procedure. The EMC forecast climatology appears to assume periodic dependence on start time. Section 3 describes least squares estimation of forecast climatologies with linear, locally linear, and periodic dependence. A standard method for computing a forecast climatology is to fit the hindcast data to some specified dependence on start time, forecast target, and lead time. Details of the CFSv2 hindcasts, forecasts, and forecast climatology are provided in section 2. A fitting method is required to compute the CFSv2 forecast climatology because the naive estimate cannot be used for initialization times that are not available in the hindcasts. (1) of the forecast climatology is equal to the hindcast average and therefore has no bias. By bias, we mean the difference between the forecast climatology and averages of the corresponding hindcasts. 2014), provided by the NCEP Environmental Modeling Center (EMC). Here we examine issues that lead to biases in the monthly forecast climatology of the NCEP Climate Forecast System, version 2 (CFSv2 Saha et al. Fitting methods can estimate forecast climatologies for start times that are not in the hindcast (interpolation) and can reduce the errors due to sampling variability (smoothing). Curve (or surface) fitting methods are an alternative to the naive method. The naive approach may be problematic when a forecast climatology is required for start times that are not present in the hindcasts (e.g., when the operational forecast schedule differs from that of the hindcasts) or when the number of hindcasts is relatively small in comparison with the forecast anomaly variance. In particular, the variance of the naive estimate is, where is the variance of the forecast anomaly. The accuracy of the naive estimate depends on the number of years in the hindcast as well as the variability of the quantity being estimated. Where f i are hindcasts whose start time, lead time, and target period match those of the forecast climatology being estimated. The proposed methods more accurately fit the hindcast data and provide a clearer representation of the CFSv2 model climate drift toward lower Niño-3.4 values for starts in March and April and toward higher Niño-3.4 values for starts in June, July, and August. Two alternative methods for computing the forecast climatology are proposed and illustrated. A further undesirable consequence of this fitting procedure is that the EMC forecast climatology varies discontinuously with lead time for fixed target month. Biases in the monthly Niño-3.4 forecast climatology are also largest for start times near calendar-month boundaries.
Biases in the monthly near-surface temperature forecast climatology reach 2☌ over North America for March targets and start times at the end of January. These biases, which are unexpected, are primarily due to fitting harmonics to hindcast data that have been organized in a particular format, which on careful inspection is seen to introduce discontinuities. Here the Climate Forecast System, version 2 (CFSv2), monthly forecast climatology provided by the NCEP Environmental Modeling Center (EMC) is shown to be biased in the sense of systematically differing from the hindcasts that are used to compute it. Forecast climatologies are computed from hindcasts by various averaging, smoothing, and interpolation procedures.
Forecast climatologies are used to remove systematic errors from forecasts and to express forecasts as departures from normal.
0 kommentar(er)
