*3.2. Oceanic Indices*

Because ENSO is very much a phenomenon associated with ocean-atmosphere interaction, a logical parameter for monitoring its behaviour is sea surface temperature (SST) as initially recognised by Bjerknes [12] and later fully explored by Rasmussen and Carpenter [23]. The three main oceanic indices used are based on SST anomalies for a number of ocean regions distributed along the equator. These are named after numbered shipping routes as Niño 1, 2, 3, and 4, because a vast array of ships crossing the Pacific for operational reasons recorded SST over a number of decades (Figure 5). Based on studies of SST variability in relation to ENSO events, Niño 1 and 2 were combined into region Niño 1 + 2; a new region, Niño 3.4 that straddles Niño 3 and 4, is now used (Figure 5).

**Figure 5.** Niño sea surface temperature regions (see Table 2). Sourced and redrawn from Barnston [21] https://www.climate.gov/news-features/blogs/enso/why-are-there-so-many-ensoindexes-instead-just-one.

Niño 1 + 2 is the smallest Niño region. It sits directly off the coast of South America and tends to have the largest variation in SST when compared with the other Niño SST regions (Figure 6). Initially, Niño 3 was favoured as the key region for observing and forecasting El Niño; however, it was realised that in terms of critical ENSO related ocean-atmosphere interactions, an area further to the west, Niño 3.4, had greater diagnostic power [24]. Niño 3.4 anomalies capture the average equatorial SSTs across the Pacific from around the dateline to the South American coast (Figure 6). It is one of two official NOAA ENSO indices used for classifying ENSO events: when the 5-month running mean of Niño 3.4 SST anomalies exceeds +0.4 ◦C (−0.4 ◦C) for six months or more, an El Niño (La Niña) is defined to have occurred. Complementing the Niño 3.4 index is the Oceanic Niño Index (ONI), the other official index, and the one used for operational definitions of ENSO events by NOAA. While the ONI uses the same SST region as the Niño 3.4 index, it classifies ENSO events differently. A 3-month running mean is used, with "fully-fledged" El Niño (La Niña) events defined when SST anomalies exceed +0.5 ◦C (−0.5 ◦C) for at least five consecutive months. The ONI is also used for defining El Niño (La Niña) onset. When the Niño 3.4 anomaly exceeds +0.5 ◦C (−0.5 ◦C) for a 3-month period El Niño (La Niña) onset is declared. Niño 4 covers the central equatorial Pacific. It displays the least SST variance of all the Niño regions (Figure 6) and is infrequently used in ENSO analyses.

In addition to the Niño regionally-based oceanic indices, the Trans-Niño Index (TNI) was developed by Trenberth and Stepaniak [24], who suggest that the TNI be used in tandem with the Niño 3.4 index. The TNI is defined as the difference in the standardised SST anomalies between Niño 1 + 2 and Niño 4 regions. The physical justification is that it captures the SST gradient between the central and eastern Pacific, and thus may be useful for identifying El Niño Modoki events, as they arise when the central to eastern Pacific SST gradient is steep, for example with sizeable positive (negative) SST anomalies in Niño 4 (Niño 1 + 2) regions. However, as noted by Hanley et al. [25], the TNI has non-consistent lag correlations with the Niño 3.4 index related to the transition of the Pacific Ocean from a cool to warm PDO phases in the mid-1970s. Accordingly, Hanley et al. [25] do not include the TNI as an index for identifying individual events and comparison of ENSO years.

**Figure 6.** Time series of SST anomalies (◦C) for NIÑO regions (see Table 2 and Figure 5) 1997–2016. The strong positive SST anomalies are especially apparent for the 1997–1998 and 2015–2016 El Niño events. Notice also how the SST anomalies generally decrease in magnitude to the west. Sourced and redrawn from the Climate Diagnostics Bulletin http://www.cpc.ncep.noaa.gov/products/CDB.
