*3.4. Water Vapour Changes during the 2015–16 El Nino*

WV has major consequences for the radiative and heat transport in the atmosphere. Even very small changes in the LS WV could affect the surface climate [6]. It also plays an important role in the distribution of O3 in the LS as an important contributor for long term change in the LS temperatures [4,5,55]. Debate is still going on the impact of El Niño 2015–16 and QBO disruption in 2016 on the WV concentrations in the LS [7,9,25]. However, in the present study we do not focus on which one is having more impact on changes in the WV concentrations in the LS. Instead, we focus only on the quantification of WV concentrations with respect to the background climatology within the tropical UTLS region. Recently, Avery et al. [7] clearly demonstrated the increasing of WV in the LS over the WP region along with warm CPT-T in December 2015. In the present study we tried to examine the zonally averaged changes in the WV at different pressure levels in the tropical UTLS region. Figure 9 shows zonally averaged WV mixing ratios at 146, 100 and 82 hPa over different latitudes. Black colour line represents climatology and red colour line represents year 2015. Strong increase of WV at 146 hPa was clearly noticed from July to December over all the latitudes (Figure 9a–c). But at 100 and 82 hPa, strong enhancement was observed from October to December 2015. There was a little drop in the WV compared to the background climatology during the summer months (June–August) of 2015 over the equator (Figure 9e,f). From the Figure 9, it is clear that the enhancement of the WV in 2015 started from October and it continues after that. The zonal mean anomalies reveal that the enhancement is high over equatorial latitudes as compared to the other latitudes.

**Figure 9.** Same as Figure 2 but for the monthly mean of water vapor.

The WV enhancement in December 2015 at 82 hPa was little high over the equator (averaged over 10◦S–10◦N) compared to the NH and SH. (Figure 10). Strong El Niño events like 2015–16 El Niño cause the warming even up to the cold point tropopause allowing more WV to enter the stratosphere [26]. In a recent paper, Garfinkel et al. [26] suggested that the impact of ENSO events on the LS temperature and WV is nonlinear in boreal spring whereas linear in boreal winter. They also clearly mentioned that the strong El Niño events led to warming over Indo-WP region that subsequently warms the CPT and moistens the tropical LS [26]. The observed zonal mean anomalies clearly indicate the strong decrease

of WV in the LS in 2016. At 82 hPa, the decrease was quite higher over equator compared to NH and SH. This record loss of WV was reported well in an earlier study reported in 'State of the Climate 2016' [56]. However, in their study, it is reported that the WV anomalies were found at 82 hPa only. In the present study, we tried to see the changes in the tropical UTLS region by estimating the WV anomalies at different pressure levels based on different latitude bands. The loss of WV in the LS in 2016 is strongly correlated with the large negative anomalies of CPT-T [25]. From the study by Tweedy et al. [25], it is evident that the decrease in global WV in the LS in December 2016 is the lowest in the record (1992–2016) [Figure 7 of [25]]. Our results also matched well with their results, except that the higher drop in the WV is observed over equatorial latitudes (>1 ppmv).

**Figure 10.** Zonal mean water vapour anomalies observed (**a**) over northern hemisphere (averaged over 11◦N–20◦N) (**b**) over equator (averaged over 10◦N–10◦S) and (**c**) over southern hemisphere (averaged over11◦S–20◦S). Different colours indicate different pressure levels.
