Composites of Anomalies

I believe there were some issues inherent to the way I was previously defining two day events in the IWVF field. For example, in a situation where there were four days straight above 90th percentile, then my code would have registered 3 events. To combat this flaw I went through and defined a two day event as:

2 consecutive days above 90th percentile
for a zone where event day - 1 must not meet criteria

This appears to have helped out very well, and thus, I will show new and improved (and shiny, I might add) results.

Note: Since MATLAB disagrees with you about how large certain subplots should be when you start adding in colorbars, I will simply state data ranges here:

DWP Anomalies: -2:5
IWV Anomalies: -3:8
Precip Anomalies: 0:5

These restrictions are based on observed values in the data and which value range best described the data.

(1) Event for SAZ Zone
The figure includes IWVF and anomalies of Dewpoint, Precipitable Water and Precipitation. Anomalies also helped to weed-out patterns present in these fields that were not easily decernded from the basic field values. We clear see a northward flux of dewpoint, precipitable water and precipitation in relation to IWVF events. Precipitable water shows the most concurrent signature with IWVF, which should be expected as these fields are based on similar data. We can now see the slightly delayed, by one or two days, flux of precipitation and dewpoint into the AZ, NM and UT domains. As described more fully in a previous post, this is a feature we expect to see. So, things are looking good.

(2) Event for SAZ and NAZ Zones

Features are similar in this figure as with (1) except that we are seeing a stronger signal in all fields compared to (1). This should be expected since we are further refining our events to cases where IWVF moved more northward, thus a strong flux.

(3) All Zones (1 Day Criteria)

So, the final figure is for an event that shows up in all three zones on a given day. This event is not restricted to 2 consecutive days criteria as with above examples. Most of the signatures appear before event day for this criteria. What we are pretty much looking at in this 'event day' is the height of activity, i.e., furthest northward flux. If we were defining events of 2 consecutive days for SAZ and NAZ then it would probably show up on Day-1 here. Again, signals are very amplified here, as we should expect.

Well, I am pretty to call events defined. So, Events = Defined:

Weak Event = 2 consecutive days of IWVF at or above 90th percentile for SAZ
Moderate Event = 2 consecutive days of IWVF at or above 90th percentile for SAZ and NAZ
Strong Event = 1 day of IWVF at or above 90th percentile for SAZ, NAZ and SUT

-jamie

CompositesI

I will be showing composites of three different surge criteria:

(1) SAZ Zone IWVF above 90th ( 342 Total: 12 per Year: 3 per month)
(2) SAZ Zone IWVF above 90th for 2 consecutive days ( 146 Total: 5 per Year: 1.3 per month)
(3) All Zones IWVF above 90th ( 47 Total: 1.7 per year: 0.4 per month)

These criteria were chosen empirically, and simply based on which results I found interesting or best told a story.

All composites will be shown for 3 Days Before through 3 Days After an event.
====================================================================
First, let's look at plot (1)
I suppose my choosing this criteria was slightly less empirical than the other two simply because I felt as though looking at results for southern Arizona might provide some strong signals. The strongest signal composite appears to be between IWVF and precipitation. A clear northward flux of precipitation over the Great Basin can be seen to correlate temporally with Day -1 through Day +1. Precipitable Water values do appear larger during this same time period, but since the general spatial pattern of values for this parameter does not appear to be dynamic I am wondering if IWV results should be suspect. Having tried to use IWV in the past for defining surges and looking at other composites this issue has come up before. So, I believe that these values might be a little shifty in the model output. Again, there does seem to be a clear signal in the precipitation field in relation to these IWVF events.

====================================================================
Next, I wanted to look at events of two consecutive days of the above criteria. I felt that by further restricting my dataset, i.e., number of events, that I would get a clearer picture of patterns in the system.
Patterns here are very similar to what was seen in the previous composite. There is still a strong temporal agreement between IWVF and precipitation. Something I found interesting in this plot though is that on Day +1 there seems to be a more intense region of precipitation further north into southern Utah and Colorado. I believe that this is indicative a stronger surge event, which would agree with our criteria of needing two consecutive days to reach or exceed the 90th percentile in SAZ.

====================================================================
Finally, I looked at events of all three zones meeting or exceeding their 90th percentile on the same day. This criteria should allow us to look at the very strong surge events, and hopefully show us heightened signals compared to the previous two plots.
Indeed, we are seeing signals that one would suspect to be associated with strong surges of the NAM. Precipitation appears to be moving quite far northward now, and it also seems augmented compared to previous plots. At first, I felt as if my calculation might have been off based on the fact that precipitation seemed so heavy in Arizona for Day -1, and even Day -2. I don't believe this is an error, but is simply the first stages of the northward flux of moisture. If we looked at events for SAZ we would see that the majority of them are centered around Day -1 for this criteria, in place of the event day. This is shown in Figure 1 from 5 Sept 2009.

Again, the IWV values seem rather questionable. They do show some sign of being heigthened on Day -3 through Day -1, but I still have my reservations about trusting this data. One feature I have noticed in all plots is the elevated values of IWV in southeastern Arizona. I realized after looking at numerous plots that this region of elevated values of IWV is spatial correlated to a region of very low, for the region, precipitation values. This only makes me question this data more as I feel like this just does not make solid physical sense. The only connect I could make between the two is that the model is representing moisture that has not been precipitated out yet in this region. I am baffled.

You may have noticed that I left out dewpoint from this figures. Plots of dewpoint for all figures were ineffective at showing any association with the larger pattern or with IWVF events. I assumed, again, that my calculations were off, but after looking back through codes I found that dewpoint was calculated no different than any other parameter, and thus should not be showing signs of bad data manipulation. Once again, I am baffled.

I feel that the strong composites between IWVF and Precipitation is good, and expected, sign. This is showing that we are not only picking up events of IWVF, but that those events are associated with elevated precipitation patterns for the desert southwest. Seeing further northward progression of precipitation appear as we further restricted IWVF events to tougher criteria is also a good sign as it shows we are able to classify the stronger surges through IWVF values.

-jamie

Moisture Surge Events

Building onto work that was shown on my last blog post, I will present my current work on describing poleward moisture surges of the NAM system.

I am still working with the sub-domains, or zones, that were defined in the previous post. First, I will show a figure of IWVF values (blue line) from 1984 for each of the three zones. Again, these values are the aggregated value for each zone for each day of the monsoon season (1 June through 30 Sept). On top of each zones IWVF will be plotted the 95th (magenta), 90th (cyan) and 80th (red) percentile values for each specific zone. The quantitative value for each threshold level will be indicated in the title of each subplot.

Interesting to note is the fact that the 95th percentile is larger for Southern Utah (SUT) compared to Northern Arizona (NAZ), 10.5 and 10.2 respectively. The percent difference is only 2.9% which is a neglagiable difference and doesn't necessarily constitute or reflect bad data or bad calculations.

Event Correlation Between Zones:

Lead/Lag correlations between zones will provide an understand of northward progression of moisture surges, as well as a hint at surge intensity. My calculation is based on events in Southern Arizona (SAZ) and looks at the correlation between the other two zones for Days -2 through Days +2 relative to the event in SAZ. Therefore, a 0.44 value for NAZ at Day +1 would mean that 45 percent of the events in SAZ had an event in NAZ the next day. Below will be three figures: The first will be 80th percentile threshold calculation for events, the second will be 90th and the third will be 95th.

There appears to be quite good correlation between all through zones on event days for SAZ. All figures also appear to have higher percentage values for post-event days, which makes physical sense when thinking about the typical northward progression of these surge events. At first I was a little surprised to see the left side of each probability curve having such high values. Once I spent a little time thinking about this I came to the conclusion that given the close temporal proximity of many of the surge events during active years, some of these high values could be attributed to surge events on previous days and not the one defined for the correlation.

Event Classification Scheme:

Onwards to defining what a surge event 'is' for this study. Based on these results, I propose the following classification algorithm for surge events in this study.

Foundation to this Criteria: 90th percentile threshold for each zone

Further refinements:
As has been shown in many recent papers, there is a push towards defining not only 'surge events' but also surge intensity, i.e., weak, moderate and strong.

Weak Surge: Only SAZ shows an event for a given day
Moderate Surge: SAZ and NAZ both show an event for a given day
Strong Surge: tSAZ, NAZ and SUT all show an event for a given day

With this criteria, I get the following numbers:
Weak Surge: 342 events over 28 yrs; avg 12.2 per yr; avg 3.1 per month
Moderate Surge: 217 events over 28 yrs; avg 7.8 per yr; avg 2.0 per month
Strong Surge: 150 events over 28 yrs: avg 5.4 per yr; avg 1.4 per month

Therefore, 63% of surges are 'moderate' and 44% are classified 'strong.'

I feel confident about this scheme, or as confident as I will ever feel about defining such dynamic weather phenomena.

-jamie