Thomas D. Helman
National Weather Service
Relative humidity usually behaves in a manner closely associated with a particular soil type. For example, a sandy area tends to have lower relative humidity compared to a more fertile soil region due to the more permeable characteristics of sand. This study examined a unique section of southern and central Wisconsin where a contrast in soil types exists between a sandy region across the central and a more fertile soil region to the adjacent south and southeast. This study will also look at a situation where convection formed over the sand region.
DESCRIPTION OF THE SAND REGION
After the departure of the last glacier (Figure 1) a significant contrast in soil type remained across parts of Wisconsin. Fertile soil was left across southern Wisconsin while a less fertile soil was left over the central and north. A rather sandy region was also deposited over some central Wisconsin (Figure 2). It should be noted at this time that reference to this sandy region is of the high permeable type, consisting of exposed sand or areas with very little quality top soil.
Figure 1. Location of the Continental Glacier over Wisconsin.
Figure 2. Solid dark lines locates the sand regions across Wisconsin. Areas with slanted lines indicates locations where the average first freeze is on or before September 24. Also included, a general description of the sand region across central Wisconsin.
Across the eastern half of this sandy region, irrigation systems are needed to support agriculture. Besides the irrigated farm fields, patches of oak and pine trees can also be found. This region has less lowland with most of the marsh land concentrated across southern portions. Meanwhile the western half consists of more marsh land than the eastern half. This area contains a good portion of the cranberry crop grown in Wisconsin. In fact, during the days of the glaciers much of the western region was under ancient Lake Wisconsin (Figure 1). There are also more stands of pine and oak forests in the western region verses the eastern region of this sand area.
Topography in this area of interest is generally flat (Figure 3). A gradual rise is noted to the north and southeast while to the southwest a rapid rise from 850 ft to around 1200 ft Above Sea Level (ASL) exists due to the Baraboo Range and the hills toward the Black River Falls area.
Figure 3. Relief map of Wisconsin with the sand regions overlaid.
Climatology also points out that the sandy soils affect temperatures in the area. Sand has a lower specific heat capacity that allows its temperature to have a shorter response time when subjected to variabilities of radiation as compared to the more moist fertile soil. Due to stronger radiational cooling over sand at night, colder temperatures are usually observed when compared with surrounding areas. In addition, the average first freeze of the season is much sooner than surrounding areas (Figure 2).
From early spring to late fall one of the National Weather Service's responsibilities is to issue forecasts for the fire weather community that includes the Wisconsin Department of Natural Resources and U.S. Forest Service. The typical forecast would include temperature, relative humidity, wind speed and direction for the next day at 1300 LST. This information is used to determine the fire potential and behavior. The fire weather community strategically locates weather stations in areas of the highest fire potential and records a 1300 LST observation each day. The fire weather community is concerned with low relative humidity, since lower relative humidity dry out the fuels faster.
The study period ranged from May 1 to June 22, 1994 since this was a period of peak number of stations reporting observations. Days chosen during the study period were selective and not random with the idea of observing the air mass relative to that particular soil region. For this study, days consisting of high solar radiation and winds less than 10 mph were chosen. High solar radiation days were needed to observe the greatest relative humidity and temperature differences between soil types. Winds less than 10 mph were chosen to avoid contamination from outside sources due to temperature or moisture advection. The rest of the days were a combination of generally cloudy skies, winds too strong and precipitation. There were 15 days during this period meeting the requirements that then were averaged. It was apparent from Figure 4 there is a significant difference between observations in the sandy areas versus adjacent observations. Temperatures were 3 to 5°F higher and relative humidity was 10 to 15 percent lower.
Figure 4. Station plots represent the average temperature in degrees Fahrenheit ( F) and the average relative humidity percent at 1300 LST for the period from May 1, 1994 to June 22, 1994. Solid dark line locates the sand region. Light lines are isopleths of equal relative humidities in 5 percent increments.
Also of interest was whether precipitation occurred within 24 hours of the observation but not near or while of the observation (Figure 5). For this study, precipitation that fell somewhere between 6 to 24 hours before the observation showed the greatest results. Note the rapid drying due to the more permeable sandy soil type verses the fertile soil type to the south and southeast. This relative humidity discontinuity zone may actually be stronger after a period of precipitation.
Figure 5. Station plots represent temperatures in degrees Fahrenheit ( F), relative humidity, and 25 hour precipitation totals in hundredths of an inch fro 1300 LST on May 11, 1994. Solid dark lines locates the sand region. Light lines are isopleths of equal relative humidities in 5 percent increments.
After examining the data from the 1994 study period, the formation of the relative humidity discontinuity zone has been revealed. Next to examine was whether this relative humidity zone over the sand region can generate or effect convection in this region. The following year, on June 2, 1995, a weak frontal boundary passed over the state before exiting southeast Wisconsin early June 3, 1995. As the front's retreated from the state, a ridge of high pressure over southern Canada was building into the state in the boundary's wake. A weak dry northeast surface flow was associated with the building ridge. Note the significantly lower relative humidity over the sand regions across central and even northern Wisconsin (Figure 6). Also note the display of the La Crosse observations for the afternoon of June 2 (Figure 7). The winds shifted from a west direction to a southeast direction then northeast. In concert with the wind shift, temperatures increased and dew points fell dramatically for four hours. After four hours winds shifted to the south with temperatures decreasing and dew points abruptly returning to earlier levels. The easterly surface flow allowed the warmer and drier air mass over central Wisconsin to push into the La Crosse vicinity. What was apparently significant was that this dry intrusion of air helped develop the relative humidity discontinuity zone and may have assisted in the development of moist convection the next day.
Figure 6. Station plots represent temperature in degrees Fahrenheit ( F) and relative humidity for 1300 LST on June 2, 1995. Solid dark lines locates the sand region. Light lines are isopleths of equal relative humidity in 5 percent increments.
Figure 7. La Crosse observations on June 2, 1995.
On the following day, June 3, 1995, high pressure was well settled over the state at the surface as well as at 700mb and 850mb (Figures 8a-8d). Winds were light and mainly from the northeast and solar radiation were in abundance, conditions much like the requirements mentioned above. Around 1900 UTC isolated showers developed about 40 miles east of La Crosse over the sand region. These showers remained nearly stationary but with a light southwest movement for about 5 hours before finally dissipating by around 2330 UTC (Figures 9a and 9b). At 2118 UTC this localized convection reached a maximum of 52 dBZ and may have been aided by weak orographic effects due to the weak easterly flow towards the Baraboo Range. Precipitation was not in the forecast this day. The relative humidity discontinuity zone was well established across central Wisconsin (Figure 10). Examining Figure 10 closer, a case could be made for a weak boundary over the sand region. Note the orientation of winds with respect to the boundary and the separation of relative humidity in the 30's over the sand region from 40's to the south. The convection that occurred this day tended to be near this boundary.
Figure 8a. RGL 850mb height (m) and temperature (C) analysis for 12Z 3 June 1995.
Figure 8b. RGL 6hr surface analysis for 18Z 3 June 1995.
Figure 8c. RGL 850mb height (m), plots, and temperature (C) analysis for 00Z 4 June1995.
Figure 8d. RGL 700mb height (m), and plots analysis for 00Z 4 June 1995.
Figure 9a. Four panel display of refelctivity from WSR-88D KMKX Sullivan, Wisconsin on June 3, 1995 from 1928 UTC to 2027 UTC.
Figure 9b. Four panel display of reflectivity from WSR-88D KMKX Sullivan, Wisconsin on June 3, 1995 from 2046 UTC to 2205 UTC.
Figure 10. Station plots represent temperatures in degrees Fahrenheit ( F) and relative humidity for 1300 LST on June 3, 1995. Solid dark lines locates the sand regions. Light lines are isopleths of equal relative humidity in 5 percent increments. Solid dashed lines indicates weak boundary.
Many questions remain with the sand region across Central Wisconsin. Is the relative humidity discontinuity zone responsible for the formation of the apparent weak boundary over the sand region? Is this weak boundary a consistent feature on days with high solar radiation and light winds, and if so, is it stationary? And finally can the relative humidity discontinuity zone as well as the weak boundary be strong enough to change the local environment, and as a result initiate or enhance convection consistently?
Although it may only be a coincidence, a 20 year period from 1974 to 1994 between the times of 1200 LST and 2100 LST showed a maximum number of tornado occurrences near and within this sand region (Figure 11). Further studies would be needed to support this correlation.
Figure 11. Wisconsin tornadoes from 1974 to 1994 occurring between the times of 1200 LST and 2100 LST. Dots represent tornadoes of little or no paths. Lines indicate paths of tornadoes.
Observations from this study point out that a relative humidity discontinuity zone does exist across central Wisconsin, mainly with days of light winds and strong solar radiation. Significant differences were observed in the temperatures and relative humidity in sandy areas versus adjacent and more fertile soil regions. Due to the permeable differences between the sand region and the more fertile soil, precipitation within 24 hours but not at or near the time of observation actually enhances the relative humidity discontinuity zone. This relative humidity zone could be responsible for the formation of a weak boundary for initiating or enhancing convection. Knowing days when this boundary exists or could develop would be useful for forecasting convection.
Special thanks to John Eise, SOO MKX Sullivan, Eugene Brusky, SOO GRB Green Bay, and Edward Johnston and the rest of the staff of MKX Sullivan for their input in this paper.
Peterson, R., 1965: Geography of Wisconsin. College Printing and Typing, Inc., 178pp.
Schultz, G., 1986: Wisconsin Foundations. Kendall/Hunt Publishing, 211pp.