Mississippi Floods 1993
Disturbance is, of course, an LTER core area. All LTER sites are engaged in studies of disturbance in or on ecosystems. Hopefully, this contribution will help draw people to the workshop in droves, get everyone thinking in new, out-of-the-cat-box ways, and force us to consider a network-wide synthesis on perturbations. Now, that is a tall order, but tall orders don't get to be person-sized until you begin to work on them.
The upper Mississippi usually wins its flood waters from the spring melt of winter snows. The spring flood is the collective product of many snowfalls over the span of the winter months.
This year was different. Reports of the spring floods were restricted to the local tabloids. Dan Rather had to find other places to go in his field-reporter khakis. Long after the winter's burden of Minnesota snows had passed through the contained Mississippi in New Orleans, the thunderstorm season began. Normally, thunderstorms flood the tributaries of the tributaries and leave the mighty Mississippi more muddy but not swollen. In 1993, we had a long run of big-time thunderstorm-systems tormenting the basin of the Upper Mississippi. This was not a disturbance event but rather a long run of disturbance events. Are runs of weather disturbances? We need to distinguish between these two kinds of disturbance.
On the upper Mississippi this year, persistence ruled the day and the day after and the days thereafter. A rather winter-like trough (southward dip) in the jet stream was centered about 90 W longitude with a companion tending-to-stay-in-one-place ridge was over the Northern Rockies. The summer-typical westward and northward expansion of the North Atlantic Subtropical Anticyclone (the Bermuda High) was happily in place and persistently so. It was like a New Age harmonic of big-time weather convergences.
Ideas about the ecological consequences of the Mississippi Flood of 1993 are rumbling around academic palaces and research proposals are, no doubt, in the making. Rightly so! Ecological systems are integrative. Runs of cloudy days have different impacts than runs of clear sunny days. A spring with rains every Monday is quite different than a spring with the same amount of rain cluster in a single week. As savers and consumers of data of interest to disturbance scientists, we have our faults. We tend to focus on the day-night and lunar cycles and save and sum-up our data in daily and monthly increments. We add the data up, divide by N and smooth out the extremes. We all know that averages hide almost everything of interest. We lose information, perhaps critical information, in the quest to understand disturbance from the perspective of the average.
I think that the Flood of 1993 will legitimize the study of runs and spells of weather and climate over a wide range of time scales. (That did not happen – so much for my vision of the future.) Meteorologists and climatologists need some fresh topics to consume their intellects and grease the wheels of their bandwagons. Publications on runs and spells of weather (bad and good) have a long and but marginalized history in the atmospheric sciences. In my view, runs and spells of weather and climate should especially impact systems that have an integrative response to weather, like ecosystems. These runs should be seen as disturbances in ecosystems.
My title for this section includes reference to the Noah and Joseph Effects. Floods are often viewed as events and aliased with nom-de-plumes such as the flood-of-the-century or a "200-year flood" based on 50 years of data. It is not unusual to hear of two 100-year floods or two 100-year storms in the same decade or even the same year. Somehow it doesn't fit Gaussian or Markovian statistical models. Real time series have these "outliers." Such events are termed the Noah Effect.
The Mississippi Flood of 93 was not so much an event as a long series of events with little in between. There is little indication that the 1993 string of thunderstorms was a random drawing from a normal population of summer-type days. The order of events was clearly not expected. It could be viewed as a disturbance in the time series. Such circumstances are called the Joseph Effect after Joseph's 7 years of plenty and 7 years of famine. So, we should consider two kinds of disturbance in climate-ecosystems dynamics. Noah effects, e.g. the 1933 hurricane at Harvard Forest and Hugo at Luquillo and North Inlet LTERs. At the VCR in the 1850s and again in the 1860 there were two Joseph Effect happenings. In both cases there was a run of years with each year having half the long-term average amount of rainfall.
Noah and Joseph: A CED Bibliography
Hurst, H. E. 1965. Methods of using long-term storage in reservoirs, Proc. Soc. Civil Eng., 116:770-.
Mandelbrot, B. B. and J. R. Wallis. 1968. Noah, Joseph, and operational Hydrology. Water Resources Research 4(5):909-918.
Mandelbrot, B. B. and J. R. Wallis. 1969. Computer Experiments with Fractional Gaussian Noises. Part 1, Averages and Variances. Water Resources Research 5(1):228-241.
Mandelbrot, B. B. and J. R. Wallis. 1969. Computer Experiments with Fractional Gaussian Noises. Part 2, Rescaled Ranges and Spectra. Water Resources Research 5(1):242-259.
Mandelbrot, B. B. and J. R. Wallis. 1969. Computer Experiments with Fractional Gaussian Noises. Part 2, Mathematical Appendix. Water Resources Research 5(1):260-267.
Mandelbrot, B. B. and J. R. Wallis. 1969. Some Long-Run Properties of Geophysical Records. Water Resources Research 5(2):321-340.
Mandelbrot, B. B. and J. R. Wallis. 1969. Robustness of the Rescaled Range R/S in the Measurement of Noncyclic Long Run Statistical Dependence. Water Resources Research 4(5):909-918.