Caution:
-The daily precipitation value is the average from of hundreds of stations in districts in and surrounding Iowa, so some areas may not have gotten any rain though the value is high and conversely, some areas may have gotten a substantial amount of rain though the value is low.

Upon examination of a 1958 to 1999 climatology of warm season subweekly 850 hPa meridional geostrophic wind speed over Texas (hereafter; 850vg or RI) with respect to subweekly Midwestern United States warm season precipitation, it is shown that monitoring 850vg in real time can be used to anticipate changes in the submonthly trend of Midwest precipitation.

Basic Usage

About 80% of the times that the meridional wind 10-day running average was at a relative maximum (minimum), the 5-day running average trend break and subsequent crossover indicated the next 7 to 10 days would indeed be relatively dry (wet).  For many of these times, about a 2-day lag occurred between the beginning (ending) of precipitation days and the meridional wind running average minimum (maximum).  Short term operational NWP models would likely have no problem forecasting the 2-day lag, but the indication of precipitation persistence out to 10 days might have advantages over operational medium range forecasts.  For drier years, when the 5-day trend break and crossover occurred, the lag between precipitation change and the meridional wind change was generally longer.  This is important because an increasing meridional wind would allow justifiable anticipation of a break in a dry spell.

Notes:    The begin day needs to be lower than the end day.
                To compare segments to the examples, keep the interval at 30 days.
                Date labels become cluttered when segments are over 60 days.
                Precipitation is the average of hundreds of Midwest stations.
 

Definition

                           g                   Zeast  -  Zwest
RI = 850Vg =  ---------------------------------------------
                            f    (distance between Zeast and Zwest)

where 850Vg is the meridional component of the geostrophic wind at the 850mb pressure level,
g is gravitational acceleration, f is the Coriolis parameter, and Z is the geopotential height at the 850 mb pressure level.

Note:  g / f (distance) ~ 0.14.
Note:  Meridional means the north south component of the horizontal wind.

The ability of changing 850vg, which is essentially a gauge of the changing strength of the low level southerly flow from the Gulf of Mexico, to indicate changing precipitation arises from selecting the 850vg parameter so that it is positioned some distance upstream from the Midwest.  The positions produce a time lag, such that the predictive capability depends on the flow maintaining its moisture transport until it reaches the Midwest and such that the precipitation is proportional to the flow. Alone, the usefulness of 850vg to evaluate Midwest precipitation is limited to precipitation trend changes because it represents only a slice of the moisture flux into the Midwest and also does not account for the moisture flux out of the Midwest, Midwest evaporation, and Midwest atmospheric storage.

The calculated wind acts as a low-level flow index and somewhat represents both anomalous cyclones in the lee of the Rocky Mountains and westward extension of the Bermuda High.  The circulation associated with the subtropical ridge is an explanation for large scale forcing of low-level southerly flow into the United States (Wexler 1961).  Although the Atlantic subtropical high is mentioned in a causal context from time to time (e.g., Helfand and Schubert 1995; Mo et. al. 1997; Walters 2001), Mitchell et. al. (1995) make the observation that the LLJ tends to occur in the warm sector of the extratropical cyclone and summarize how a major contribution to the LLJ by the subtropical high contradicts findings by Uccellini (1980) and Chen and Kpaeyeh (1993) that lee side troughing and surface cyclogenesis are key factors.  Schubert et. al. (1998) associate LLJs on different time scales with interaction between various scale troughs and the Rocky Mountains.

Definition Details
 The vicinity of Fort Worth, Texas and the area south is generally the location with highest northward moisture flux (e.g., Helfand and Schubert 1995; Mo et. al. 1997; Higgens et. al. 1997).  Because of the substantial moisture flux and the upstream position (upstream of important southerly flow) from the Midwest, the east central Texas region is the place focused on for finding reliable wind information for the desired period.  Thus, a first consideration of a low-level flow parameter, for first establishing climatology and then for monitoring, would be the radiosonde winds reported at Fort Worth or stations south.  However, it is not uncommon for data to be missing for several days, so in the interest of improving consistency, it would be beneficial to involve more than one station.  Except for a station at Stephensville, Texas from 1974 to 1994, no other consistent stations are close to and south of the Fort Worth station for the selected time period.  Station spacing is such that there seems to be a large region in east central Texas with no radiosonde stations.  The stations at Del Rio, Texas (DRT; 29.4 N, 100.9 W), Midland, Texas (MAF; 31.9 N, 102.2 W), Corpus Cristi, Texas (CRP; 27.7 N, 97.1 W), and Lake Charles, Louisiana (LCH; 30.1 N, 93.2 W) are quite far from Fort Worth, Texas (FWD; 32.8 N, 97.4 W) and it does not seem logical to guess the wind data from those stations would have strong ties to the FWD wind data.  Therefore, using measured winds was dismissed upon consideration of calculating the average geostrophic wind over the east central Texas area.  Radiosonde stations at DRT, MAF, LCH, and Jackson, Mississippi (JAN; 32.3 N, 90.1 W) have provided reasonably consistent data for the 1958 to 1999 period and look to be stable in upcoming years.  For these four stations, 850 hPa radiosonde heights were assembled from Radiosonde Data of North America 1946 to 1992 (1993), Radiosonde Data of North America 1994 to 1997 (1998), http://raob.fsl.noaa.gov/, and http://weather.uwyo.edu/upperair/sounding.html.  Heights from DRT and MAF (LCH and JAN) were averaged into one western (eastern) height to be used in the equation to calculate the meridional 850 hPa geostrophic wind.  When the height was missing from a particular station and reporting time, the average height was derived as a function of the non-missing height.  The function was determined by linear regression of all average heights that were computed and the corresponding individual station height data.  This methodology provided a calculated 850 hPa meridional geostrophic wind data set that was quite consistent.  Data set gaps at this time include one day in May 1998 and three days in July 1998 where both western stations' heights were missing.

Basic Usage Details
"The moving average convergence-divergence trading method (usually abbreviated to MACD) was originally developed in 1979 by Gerald Appel as a stock market timing device. ... The basic MACD signal is the crossover.  Buy signals are generated when the faster line crosses the slower line from below, and sell signals are just the opposite....One very interesting way of using a MACD is to get a jump on a crossover signal by drawing a trendline on the MACD itself and then trading when the trendline is broken, rather than waiting for the crossover.  A break in an MACD trendline can precede an important break in the market, and it serves as an early warning signal that a market is turning.  MACD crossovers that are preceded by or in conjunction with a trendline break tend to have much more technical importance than MACD crossovers alone....Remember, if you trade based solely on a break in the trendline without waiting for the crossover, the trade will have little justification if the crossover fails to occur in the near future. "(LeBeau and Lucas 1992)
Instead of strictly applying the specifics of Appel's (1985) system, a simple dual moving average analysis was performed on the low-level flow parameter.  For the longer term average, a 10-day moving average was selected while a 5-day moving average was selected for the short term.  Not surprisingly, the 10-day precipitation and the 10-day meridional wind followed each other closely as did the daily precipitation and meridional wind.  Quite often the same information could be attained from the precipitation trend as the trend of the meridional wind.
 

Appel, G., 1985: The Moving Average Convergence/Divergence Trading Method— Advanced Version.
    Scientific Investment System. ??pp.
Chen, T.-C., and J. A. Kpaeyeh, 1993: The Synoptic-Scale Environment  Associated with the Low-Level Jet of the Great Plains.
    Mon. Wea. Rev.,  121, 416-420.
Helfand, H. M., and S. D. Schubert, 1995: Climatology of the Simulated Great  Plains Low-Level Jet and
    Its Contribution to the Continental Moisture Budget of the United States.
    J. Climate, 8, 784-806.
Higgins, R.W., Y. Yao, E. S. Yarosh, J. E. Janowiak, and K. C. Mo, 1997: Influence of  the Great Plains Low-Level Jet on
    Summertime Precipitation and Moisture  Transport over the Central United States.
    J. Climate, 10, 481-507.
LeBeau, C., and D. W. Lucas, 1992: Technical traders guide to computer  analysis of the futures market.
    Business One Irwin. 234 pp.
Mitchell, M. J., R. W. Arritt, and K. Labas, 1995: A Climatology of the Warm  Season Great Plains Low-Level Jet Using Wind Profiler Observations.
    Wea. Forecasting, 10, 576-591.
Mo, K. C., J. N. Paegle, and R. W. Higgins, 1997: Atmospheric processes  associated with summer floods and droughts in the central United  States.
    J. Climate, 10, 3028-3046.
Radiosonde Data of North America 1946-1992. Version 1.0, 1993: CDROM Vol III
    Forecast Systems Laboratory and National Climatic Data Center.
Radiosonde Data of North America 1994-1997. Version 2.0, 1998: CDROM Vol IV
    Forecast Systems Laboratory and National Climatic Data Center.
Schubert, S. D., H. M. Helfand, C.-Y. Wu, and W. Min, 1998: Subseasonal  Variations in Warm-Season Moisture Transport and
    Precipitation over the  Central and Eastern United States.
    J. Climate, 11, 2530-2555.
Uccellini, L. W., 1980: On the role of upper tropospheric jet streaks and  leeside cyclogenesis in the development of low-level jets in the Great  Plains.
    Mon. Wea. Rev., 108, 1689-1696.
Walters, C. K., 2001: Airflow Configurations of Warm Season Southerly Low- Level Wind  Maxima in the Great Plains.
    Part II: The Synoptic and  Subsynoptic-Scale Environment.
    Wea. Forecasting, 16, 531-551.
Wexler, H., 1961: A boundary layer interpretation of the low level jet.
    Tellus, 13, 368-378.

http://weather.uwyo.edu/upperair/sounding.html
http://raob.fsl.noaa.gov/