Easterly Waves

After much effort and little success with classifying easterly waves with the methodologies of Knippertz (2003) I have decided to classify these features with the criteria of Stensrud et al. (1997). I think the issue with using Knippertz methods is that the easterly waves typically have shorter wavelengths and are often tilted horizontally at angles that could prove to be difficult for identification with his strictly zonal calculations. Probably a more important factor is that these waves are identified at a lower level, usually 600 or 700 hPa, which makes topography an issue. Either way, I plan on coming back to this issue at a later time. For now, let's see what results I have found with Stensrud et al. (1997).

This identification methodology uses time-dependent changes in 600 or 700 hPa meridional winds to classifty easterly waves. They classified passage of an easterly wave when a northerly wind on Day 1 was followed by a southerly wind on Day 2, as reference to trough axis passage, at 110 W. They used Hovmoller diagrams to show how this methodology 'looked.' So, below is an example of 600 hPa for 1 June through 24 Aug 1987. Winds are at 22.5 N and 145 to 60 W. Warm (cool) colors correspond to southerly (northerly) flow.
Their study used winds that were averaged over 10 to 22.5 N. Douglas and Leal (2003) showed that high amplitude waves over western Mexico typically migrate across the region between 10 and 30 N. Therefore, I chose to limit my domain to 20 (~Mexico City, MX) to 25 N (~Brownsville, TX) so that I would capture features inline with their findings as well as ones more associated spatially with the central and northern portions of the NAM. Many recent papers have started to use winds at 600 hPa in lue of 700 hPa so that an influences from the topography of Mexico does not affect the signatures of these wave features. I performed analysis on both levels for comparison and my own interest in their correlation.

Results:
Fuller and Stensrud (2000) using Stensrud et al. (1997)methods -- 85 easterly waves over 14 years (only July and August)
=> 6.07/year and 3.03/month
My work -- 260 (243) easterly waves at 600 (700) hPa over 29 years (only July and August)
=> 8.9/year and 4.4/month(600 hPa)
=> 8.4/year and 4.2/month(700 hPa)

How did the two levels compare with one another? Well, 161 waves (62% and 66% of waves at 600 and 700 hPa, respectively) were identified at both levels on the same day. I am happy with these results in the context of how they compare with previous work, but I believe we can definitely improve the validaty of these results by using a more robust identification scheme like those in Knippertz. In the case where unforeseen events come into play, at least we have results to use in our reserach that are in line with previous research.

-jamie

Surge Classification

After talking with Mike last week I decided to go back to my surge classification work. Along with thresholds for dewpoint and winds, I have now included precipitable water. Before I get to how I have taken into account precipitable water, I first want to mention how I have come to my threshold values for dewpoint and winds.

Many of the Stensrud papers have used wx station data from Yuma, AZ to define surges. Since this is the present benchmark for surge classification, I have worked to judge my results to theirs, i.e., roughly the same average monthly surge events. All of the papers that I have read so far have found an average of ~3.0 surge events per month for July and August (Note: Almost all of these papers only look at data for these two months). Mike, though admitting that describing an average for something as dynamic and variable as the NAM can be tricky, said that this value seemed quite legitimate. As a simple test of the methods of Stensrud, I ran his threshold algorithm on our SAO data for the Yuma station and got very similar results. After this, I used data from the closest NARR grid point to Yuma. With some trial and error I found that a 70th percentile threshold level for dewpoint, wind speeds and precipitable water gave me comparable results. From this successful result I moved forward and defined a larger domain of interest to test these thresholds over a larger, and more appropriate, area.

I am using a domain of 31 to 35N and 108 to 115W. An example of this domain is shown below for precipitable water, in mm, on 19 August 1983.
I calculated an aggregated daily mean for each variable, as well as a daily 70th percentile value for each variable. Events of each are defined when the daily value is equal to or greater than the 70th threshold. I defined 4 different combinations of these criteria:

(1) Wind event on Day 1; PW event on Day1 or Day 2; DWP event on Day 1
(2) Wind event on Day 1; PW event on Day1 or Day 2; 2 consecutive days of DWP event from Day 1
(3) Wind event on Day 1; PW event on Day1 or Day 2; 3 consecutive days of DWP event from Day 1
(4) Wind event on Day 1; PW event on Day1 or Day 2; DWP event on Day 1 or Day 2

The average monthly number of surge events in a given July or August for each criteria were 2.85 (1), 2.79 (2), 2.2 (3) and 3.86 (4). Next, I wanted to compare these results with the results from using the Stensrud surge criteria for the NARR grid point near Yuma. I did this by finding days were an event was classified by a specific criteria about, i.e., 1 through 4, and also classified by the one grid point Stensrud criteria for Day-1, Day or Day+1. The results for percentage of agreement is below for each criteria:

(1) 59.8%
(2) 64.7%
(3) 70.7%
(4) 56.9%

There does appear to be rather good agreement between these criteria and Stensrud's results, but it is not perfect. The criteria with the lowest average monthly surge events, (3) with 2.2, showed the highest correlation with Yuma. In general, I believe there is much room for improvement over the classification scheme of Stensrud since he has used such a limited data source of only one station. Branching out from what is tried and true could raise a lot of questions of our work, but I imagine that is typical of anything new within research.

Monthly Average Trough and Ridge Occurrences

Below is a graph of the average monthly occurrences of troughs, left column, and ridges, right column, over the rough extent of the United States. Our dominate monsoonal ridge is well represented in the June-Aug data. July and August also show a well defined separation between troughs off the east and west coasts of the US. A large blocking pattern, like the one associated with the monsoonal ridge, would produce a pattern very similar to this one.

June and September both show a larger extent of both trough and ridge occurrences. Since these months can be are more closely related to the more dynamical Spring and Fall weather patterns of the northern hemisphere than July and August, we could expect to see more transient features and thus a larger area experiencing these features during June and September.

Trough Identification

Following Knippertz (2003), I ran calculations on 500 hPa height data for 1 Jun through 30 Sept from 1980 to 2008.

This identification scheme uses calculated zonal geopotential height gradient data to detect occurrences of troughs. For an in-depth methodology refer to the above paper. My domain covers 10 to 60N and 145 to 65W. Setting these domain boundaries, specifically the longitudinal boundaries, produces data for trough occurrences located between 130 to 80W, which is of importance because of its influence on the North American monsoon system.

I ran a moving-grid box calculation across the domain for each day of each year in order to locate troughs within the region. The grid-box calculation consisted of three boxes. Z2, the central of the three, was a 3x4 grid domain. Z1 and Z3, the boxes westward and eastward adjacent, respectively, were comprised of 3x5 grid domains. For a given day the mean of the averaged Z1 and Z3 was compared to the mean of Z2. The resulting value, P, comparable to the zonal geopotential height gradient, was then assigned to a central point within Z2. Values of P greater than 25 are indicative of a location experiencing troughiness. Values of 100 or greater relate to more extreme, a.k.a. deep, troughs.

Below is an example of results for one day, 1 Jun 2001, from these calculations. The 500 hPa height field is shown in black, solid lines. We can see a clear pattern of troughiness off the western coast of the United States and the Mississippi River Delta. A well defined ridge is located between these two locations. The results from the P calculations are colored-coded, values ranging from 25 to 100, and plotted under the 500 hPa height field. The location of the P values that are associated with troughs are indeed co-located along the trough axis. After running loops of one week to one month, it seems clear that the calculations for P have worked quite well, and can be considered accurate, at least empirically.
Next, ridges will also be defined. This will be a simple task though, as I will use the same P calculations that have already been done, expect I will look for large negative values to indicate ridge occurrences.

-jamie

IWV and IWVF Events

As referenced in an earlier post, I have been working towards quantitatively describing surges with IWV and IWVF. I have tried many different approaches to this problem, such as:

(1) the aggregated value of the domain being greater than 95th, 90th, 80th, 70th or 50th percentiles
(2) sustained values of (1) for 2 and 3 consecutive days

Below are the 12 combinations of the above two critera for the year 1983. Titles for each plot are given above the respective plot.
Note: You can click on any graph to open up the full resolution version of the image in a new tab.

To test these 12 thresholds for accuracy, I compared them to a Hovmoller for precipitation (averaged over -116 to -107 longitude) for the same domain and same year (This figure will be shown below). From the precipitation Hovmoller, I chose 5 suspected events on days ~38, ~50, ~70, ~78 and ~120. I compared the above plots to see which, if any, produced events on roughly these same dates. From that, I observed that the IWVF thresholds produced better results than the IWV, note plots 1 and 2 (1 day of >=90th percentile for IWVF and IWV, respectively). Overall, it appeared that 1 day of IWVF >=95th percentile, plot 3, and two consecutive days of IWVF >= 90th percentile, plot 5, produced the best results. These two plots will be highlighted below and compared to the Hovmoller of precipitation mentioned earlier.
It is my judgment, that for this year the two consecutive days of >=90th percentile threshold produced better results. To provide a better view of exactly what this threshold is describing, below is a comparison of 'events' in this criteria to a Hovmoller of IWVF for the same time period and domain.

It seems clear that this threshold limit is picking-out the more statistically significant periods of IWVF. Now, to show the relationship between IWVF and precipitation, below are IWVF and precipitation Hovmollers for same temporal and spatial extents.

There does seem to be a rather nice agreement between this two parameters (precipitation surges seem to be associated with high values of IWVF). So, I got that going for me...which is nice.

Hopefully, this will be an accurate and clean way of describing this "Gulf Surges" in the North American monsoon.

-jamie