In discussing agriculture, Strommen says, global averages are meaningless. Agriculture and its practices take place on a local and regional basis and must be studied on this basis. Strommen offered a number of cases of applied climatology and agricultural system analysis relevant to the Metro-Agro-Plex concept.

Strommen discussed research in Detroit and Lansing, Michigan regarding urban influences on agriculture. A study was undertaken in Detroit to deal with a flooding problem. It was clear that hydrological curves in the area around the city had changed dramatically. Engineers concluded that the problem was from paving of shopping centers and roads. Using a network of 85 recording rain gauges, the researchers extracted onset time of rain fall, duration intensities, etc. They could see that something was producing thunderstorms that were exceeding the expected 100-year return period precipitation amounts 4 to 7 times a year.

On closer examination of temperature data from three city and rural stations, a couple of distinct patterns emerged. One was a clear documentation of urban heat island expansion. It was also seen that a shift in location of rainfall normally based on lake effects had occurred. A seasonal break down showed that all the high precipitation events were concentrated in the summer. When the climatology of the synoptic field was added, it was clear that these events occurred with no significant storm systems in the area, and the intense rain fell only in the urban area and on the nearby surrounding agricultural land.

Wind field data verified that a convergence zone along the lake breeze front was the trigger mechanism for these thunderstorms, taking place in a band only 5-15 miles wide. This convergence zone moved the naturally occurring rainfall events further inland. The urban heat island is one contributing factor in driving these events. Another is the added cloud condensation nuclei (CCN) from urban air pollution. The lake breeze front acts as the trigger mechanism. The lesson from this use of applied climatology was that urban areas can modify rainfall patterns significantly and this can affect agriculture near cities.

What size urban area does it take to produce some alteration in rainfall pattern? Detroit was obviously large enough to create this effect but what about Lansing, a city of 50,000 people? Using a one-mile grid of recording rain gauges, it was found that Lansing was indeed shifting the downwind areas of maximum rainfall. As an industrial area, the pollution did increase CCN and there was also some degree of urban heat island effect. Further research in other urban areas has confirmed that convergence zones tend to develop over cities and then move more frequent and intense showers downwind.

This information can be important to farmers in terms of their fertilizer applications. Downwind of prevailing wind over a city, many trace elements are deposited on agricultural land by these anomalous rains and this alters farmers' fertilizer needs. Farmers need to know more about these events so they can alter their fertilizer applications appropriately.

In another case, Strommen and colleagues measured the pH levels of every rainfall event and analyzed these data, added the synoptic pattern, etc. This allowed the tracing of sources of pollution that were previously unknown, and found some pH levels down to around 2, compared to a norm of 6 or 7. This also provides information that would be important to farmers for obvious reasons. The pattern of pH measurements precisely followed the plume from the steel mills of Gary, Indiana which became more dilute as it moved away from this source. Strommen says it is critically important to help politicians see the practical applications of such research or it will not be supported.

In another case, there was a drive to build cooling towers for a nuclear power plant on the shore of Lake Michigan so as not to introduce warm water into the lake and perturb the salmon environment. Strommen believed that this was a mistake, due to the frequent strong inversion over western Michigan. The cooling tower discharge would put a warm, moist air plume over the interstate highway and create hazardous driving conditions from icing of windshields and could also increase disease problems in the nearby agricultural areas. The towers were built, and the predicted hazardous conditions did indeed result. Eventually data were collected that proved Strommen's hypothesis.

The lesson here is that it is important to understand the total system when considering response strategies, and not make decisions based on a single issue. It is necessary to understand the implications of fallout, downwash, etc., as cities are developed with agricultural areas nearby. As power generation expands, it is necessary to understand the implications for agricultural lands down wind. Comprehensive, interdisciplinary planning is needed.

In another example, Strommen was asked to visit Haiti, which was experiencing a huge decline in food production which was being blamed on climate change. Strommen found that in fact, the climate had not changed, and that the problem seemed to be the result of poor agricultural management practices that had led to the loss of enormous amounts of topsoil. A model demonstrated that changing the water-holding capacity of soil could create up to a four-fold increase in the variability of crop yields. Upon examination, it was found that the A and B horizon of the soil profile was no longer visible and the farmers were trying to grow crops in the C horizon of the soil. Low nutrient uptake and low water retention, resulting from mismanagement, had caused the problem. Sustainable farming techniques do exist, and are being utilized in some places; but when production capabilities are destroyed through bad management, restoring them is a very slow and extremely expensive process.

Research should be followed by a testing and evaluation phase involving users, not just scientific researchers. This is a major missing link in getting research transferred into applications.

In another case, the management technique of applying fungicide every fourteen days to ward off peanut leaf spot was used with good results. When research was conducted and data gathered over a three-year test period, it was found that fungicide applications could be reduced by almost 50%, saving farmers huge amounts of money, reducing energy use, and reducing the amount of fungicides in the environment without loss of crop quality or yield reduction. A real payoff of research is in such applications that protect the environment in many ways while saving money for the farmer, thus creating win-win solutions.