HIGHLIGHT: The seasonal mean NINO3.4 anomaly was about -0.37 C for DJF2014, a continued slightly –ve value.

The CA method, acting on global DJF2014 seasonal means as latest input, but using a string of global SST going back one year, has all of its members staying negative in the neutral zone for Nino34 thru MAM2014. By summer 2014 a majority of members is neutral, but now on the positive side, and a few grow >+0.5 later in 2014. But two members stay slightly –ve for all of 2014!

Weights have changed from last month, see also ‘explanation’ below. The number of +ve weights larger that 0.10 has increased to nine. 1988 (thru December 2012 that particular past situation was labeled 1986, and thru last month it was labelled 1987) has now been holding on to a large negative weight for two years: since spring 2012. A remarkable case of determinism.

Explanation: Weights have changed from last month for three reasons. First, weights are labelled by the year of the middle month of the latest 3 mo period. Since we are changing from NDJ to DJF, we see, for instance, December 1987 turn to January 1988, so weights (even if unchanged) move forward by one year in the table below. Secondly, one year of data has been added, so the larger co-variance matrix (now 58X58, was 57X57) will possibly respond differently to the co-linearity present. Lastly, there will be the normal changes one sees every month relative to the previous month.

OCN flavor is weak. The last 13 years have generally +ve weight, but now four years are mildly –ve. 1981-2000 is mainly negative, but with exceptions (1991).

Table 1. Inner products (IP; scaled such that the sum of absolute values is 100; not shown) and weights (Wgt multiplied by 100.) of each of the years to construct an analogue to the sequence of 12 consecutive overlapping 3-month periods defined as the base (currently the 12 overlapping seasons string JFM2013 thru DJF2013/14). Years in the table are labeled by the year of the middle month of the most recent predictor season. Weights for 1956 and 2013 are less (by 50%) because only half the members use these years ("early" members use 1956, "late" members 2013). Data currently thru Feb 2014.

Year	IP	Wgt	Year	IP	Wgt	Year	IP	Wgt	Year	IP	Wgt	Year	IP	Wgt	Year	IP	Weigt
1956	xx	-4	1966	xx	-5	1976	Xx	-5	1986	xx	4	1996	xx	4	2006	xx	7
1957	xx	11	1967	xx	-1	1977	Xx	-2	1987	xx	-8	1997	xx	3	2007	xx	-2
1958	xx	7	1968	xx	2	1978	Xx	-2	1988	xx	-14	1998	xx	-9	2008	xx	4
1959	xx	11	1969	xx	-5	1979	Xx	4	1989	xx	-1	1999	xx	1	2009	xx	13
1960	xx	-5	1970	xx	-2	1980	Xx	10	1990	xx	7	2000	xx	-13	2010	xx	8
1961	xx	-7	1971	xx	2	1981	Xx	-8	1991	xx	13	2001	xx	11	2011	xx	-3
1962	xx	8	1972	xx	-1	1982	Xx	-10	1992	xx	-4	2002	xx	-1	2012	xx	10
1963	xx	4	1973	xx	15	1983	xx	-6	1993	xx	-9	2003	xx	11	2013	xx	9
1964	xx	-3	1974	xx	3	1984	Xx	-2	1994	xx	-6	2004	xx	-3		xx	Xx
1965	xx	-8	1975	xx	0	1985	Xx	4	1995	xx	-3	2005	xx	9		xx	Xx

Weights are high +ve (>+0.10) for 2012, 2009, 2003, 2001, 1991, 1980, 1973, 1959, 1957

Weights are high -ve (< - 0.10) for 2000,1988,1982

Ens ave Nino34 anomaly forecasts, for leads -3, -2, -1, 0 etc: in plain numbers (ensemble mean):

-0.37	-0.37	-0.34	-0.25	-0.1	0.07	0.19	0.14	0.12	0.2	0.37	0.4	0.34	0.32	0.29	0.3	0.3	0.28	0.3	0.32	0.34
	JFM14		MAM		MJJ		JAS		SON14		NDJ		JFM15		MAM		MJJ		JAS

One month ago:

-0.13	-0.17	-0.21	-0.24	-0.16	0.01	0.17	0.32	0.29	0.34	0.43	0.6	0.61	0.54	0.43	0.39	0.35	0.3	0.25	0.26	0.28
NDJ		JFM14		MAM		MJJ		JAS		SON		NDJ		JFM15		MAM		MJJ		JAS

Two months ago:

-0.13	-0.11	-0.12	-0.13	-0.12	-0.04	0.13	0.27	0.37	0.3	0.3	0.35	0.46	0.42	0.32	0.25	0.22	0.22	0.17	0.1	0.05
	NDJ		JFM14		MAM		MJJ		JAS		SON		NDJ		JFM15		MAM		MJJ

Three months ago

-0.14	-0.14	-0.12	-0.17	-0.19	-0.16	-0.04	0.14	0.27	0.32	0.25	0.22	0.27	0.38	0.36	0.26	0.18	0.17	0.19	0.18	0.11
SON13		NDJ		JFM14		MAM		MJJ		JAS		SON14		NDJ		JFM15		MAM		MJJ

Four months ago

-0.23	-0.25	-0.26	-0.27	-0.32	-0.31	-0.25	-0.09	0.15	0.34	0.4	0.32	0.3	0.42	0.61	0.65	0.55	0.41	0.36	0.34	0.25
	SON13		NDJ		JFM14		MAM		MJJ		JAS		SON14		NDJ		JFM15		MAM

Anomalies relative to 1981-2010 climo.