Abstract
The purpose of this experiment was to ascertain whether texture has a significant role in determining the thermal conductivity of soil. An attempt was made to collect samples which would be good representatives of each of the three soil textural classes. To eliminate moisture as a variable, all samples were oven-dried, and to eliminate color as a variable, all samples were heated from below.
A test chamber was constructed to keep the heating of the soil samples uniform by keeping the samples a constant distance from the heat lamp. Additionally, the test chamber’s sides were enclosed with aluminum sheets to minimize the influence of the air in the room on soil samples being tested.
Each sample was first placed in a freezer to lower its temperature and allow a greater increase in temperature to occur during heating. Exact starting and ending temperatures were not a concern, since it was the slope of the increase that was the focus of the investigation. The warming of each sample over the heat source was measured for twenty minutes, enough time to establish a distinct trend. Soil from each collection site was divided into two testing samples, which were each tested five times.
The data collected showed a direct relationship between thermal conductivity and texture as well as an indirect relationship through the variables influenced by texture, such as pore space and particle density. The direct relationship of texture to thermal conductivity shown was that as particle size decreased, thermal conductivity increased. The indirect relationship shown was that as particle density increases, thermal conductivity increases, and as pore space increases, thermal conductivity decreases. Another conclusion that can be drawn from the data is that in nature, it may ultimately be the water capacity of a soil (which is itself influenced by texture) which determines its thermal conductivity.
INTRODUCTION
It is well-established that soil is, one way or another, of great importance to every living plant and animal on this planet. The purpose of this experiment was to ascertain whether texture has a significant role in determining the thermal conductivity of soil. If texture would significantly affect the thermal conductivity of soil, the fields of both engineering and agriculture would benefit.
The temperature of soil can either increase or decrease its usefulness to us. At lower soil temperatures, biological rates are much slower. At low enough temperatures, biological decomposition is at a near-standstill, thus limiting the rate at which valuable nutrients such as nitrogen, phosphorus, sulfur, and calcium are made available to plants. In Spring, nitrification does not begin until the soil temperature reaches around four and one-half degrees Celsius (40° F), although the process is accelerated to the most beneficial rates when the soil reaches around 27 to 32 degrees Celsius (80°-90° F). Additionally, plant processes, such as root growth or germination, do not occur until the soil reaches certain temperatures depending on the particular plant. Another plant process adversely affected by cold temperatures is the transport of nutrients and water. A better understanding of how different soils warm up would benefit agriculture by allowing for better planning of planting of crops.
In addition to agrarian applications of the textural relationship of thermal conductivity, there are many other applications as well. To reach the highest possible level of efficiency, geothermal heat pumps must be planned in accordance with soil temperature fluctuations. Even with just a general knowledge of thermal conductivity, the insulation of basement walls could be better planned for maximum efficiency.
Many factors influence the thermal conductivity of soil, however this experiment was designed to look solely at the effect of texture on thermal conductivity. The soil was oven-dried to eliminate water content as a variable, something that is known to dramatically influence the thermal conductivity. Pore space was a variable to some extent, however, since pore space is in part dependent on the texture of a soil, accounting for and factoring out pore space would, in essence, be the same as factoring out texture, the focus of this investigation. Because after each test, the soil samples were emptied from the test cylinder into sample bags to be refrigerated, the pore space (and other measurements derived from it, i.e. particle density) varied even with the same sample from test to test. However, pore space was measured in each soil sample so that the measurements could be taken into account (not factored out) when examining the results. Mineralogical composition was also a variable, but again, partly dependent on texture and therefore something that should not to be completely eliminated as a variable. Color was a variable which was also known to influence how much heat would be absorbed, and to eliminate it as a variable, all samples were heated from underneath. An additional test was performed to measure the percentages of organic matter in each sample to take into account its influence on the results of the experiment.
PURPOSE
The purpose of this experiment was to ascertain whether texture has a significant role in determining the thermal conductivity of soil. A better understanding of how different soils warm up would benefit agriculture by allowing for better planning of planting of crops. However, in addition to agrarian applications of knowledge of the relationship of thermal conductivity to texture, there are many other applications as well. To reach the highest possible level of efficiency, geothermal heat pumps must be planned in accordance with soil temperature fluctuations. Also, even with merely a general knowledge of thermal conductivity, the insulation of basement walls could be better planned for maximum efficiency.
PROCEDURE
- Collect materials:
- Soil
- Sample Bags
- Oven
- Balance/Scale
- Enclosed Aluminum Test Chamber
- steel rods
- four 30.5 cm x 61 cm aluminum sheets
- 125 Watt Incandescent Heat Lamp
- Aluminum (355 mL) soda cans
- SensorNet Temperature Probe
- SensorNet Software
- Computer
- Make sure to get soil samples that are a good representation of each texture type. Number each sample according to location.
- Have the texture of each sample analyzed and analyzed and identified by soil scientists.
- Oven-dry soil at 121°C; check weight while drying until at least two successive readings are equal, signifying that all moisture has been evaporated (approx. 2½ hours).
- Prepare test cylinders by cutting off the tops of aluminum soda cans (355 mL) 1.5 cm from the top.
- Construct a test chamber 30.5 cm wide, 30.5 cm deep, and 61 cm high with steel rods; enclose sides with aluminum sheets to help minimize effect of room temperature on the sample being heated.
- Attach the heat-lamp to the bottom of the test chamber with bulb directed upward.
- Construct a ledge for the soil test cylinder 7.5 cm above the heat lamp.
- Place soil samples in freezer to lower their temperature. While they should be as close as possible to 0° C, the exact starting point for the experiment is not critical. Due to various factors, different samples, even though exposed to the same air temperatures for equal amounts of time, will not themselves chill to the same temperatures as one another. This is inconsequential since it is the rates of warming, not specific temperatures attained, that are the focus of this investigation.
- Remove a sample from the freezer.
- Place soil in test cylinder within 1 cm of the top and place the test cylinder in the test chamber.
- Place temperature probe 4.5 cm into the soil.
- Turn on heat lamp and begin recording temperature, setting the computer to take a reading once every second for twenty minutes, the time period necessary to establish the slope on a graph of time versus temperature.
- Freeze each individual sample again and test for a total of five times, each time placing the soil in the freezer before being tested again.
- After all runs are completed, test each sample for bulk density and particle density; To calculate bulk density, take a volume of soil and weigh, and divide the weight by the volume. Next, to calculate particle density, take that same volume of soil and add about twice as much water as the volume of the soil. Stir the soil around to allow the water to completely fill the pores, and then note the volume of the water and soil combined. To calculate particle density, then divide the weight of the soil by the volume of the solids (total volume of the soil minus the pore space).
- To calculate the organic matter, first weigh a small crucible and then add about 5 grams soil to it. Place crucible in a ring stand over a Bunsen burner and heat to a constant mass. Calculate the weight loss, and divide by the initial weight of the soil (not including the crucible) to determine the percentage of organic matter.
DISCUSSION
Many factors influence the thermal conductivity of soil, however this experiment was designed to look solely at the effect of texture on thermal conductivity. The soil was oven-dried to eliminate water content as a variable, something known to drastically influence the thermal conductivity of soil. Pore space was a variable to some extent, however since pore space is in part dependent on the texture of a soil, accounting for and factoring out pore space would in essence be the same as factoring out texture, defeating the whole purpose of this investigation. Additionally, because after each test the soil samples were emptied from the test cylinder into sample bags, pore space and other measurements derived from it varied from test to test with each of the samples. Mineralogical composition was a variable that could not be accounted for, but again, is partly dependent on texture and therefore something that should not be eliminated as a variable. One major variable was color, which was eliminated for testing by heating from below. To isolate the samples being tested from the influence of the room’s air temperature and currents, the sides of the test chamber were enclosed with aluminum sheets.
Although the samples were all refrigerated for the same amount of time, the temperatures they reached were often different, with samples one and eight consistently being coldest at the beginning of the tests. This difference can be attributed to differing thermal conductivities which allowed the cold air to migrate through the samples at different rates, although the conclusion that can be inferred by observing the starting temperatures of the various samples does not coincide with the one reached through testing.
Throughout testing, the slope of the data of each sample varied greatly from the lowest slope to the highest, and the slopes of different samples from the same site also differed from test to test. When a graph was made on with the texture of soil on the x axis and average slope on the y axis, a general trend of increasing thermal conductivity moving from left to right (from larger particle sizes to smaller particle sizes) was noted. A linear trendline was fitted to this graph, further demonstrating the trend.
Other graphs were also made in an attempt to determine if there was a stronger correlation between thermal conductivity and one of the variables that was accounted for but not eliminated. First, a graph was made with percent organic matter on the x axis and thermal conductivity on the y axis. On the basis of this graph, no correlation better than or equal to that between soil texture and thermal conductivity was discerned. Next, a graph was created which attempted to show if there was a strong correlation between percent pore space and thermal conductivity. A linear trendline was fitted to the data, showing a decrease in thermal conductivity as pore space increased. This seems contrapositive to the relationship already established by the data of decreasing particle size corresponding to increasing thermal conductivity. However, this actually strengthens the relationship by showing that the effect of texture itself on thermal conductivity to be stronger than the effect of pore space. Finally, a graph was created with particle density on the x axis and thermal conductivity on the y. This graph also showed some correlation between particle density and thermal conductivity, with an increase in thermal conductivity corresponding to an increase in particle density.
It was not expected that this investigation would produce completely conclusive or clear-cut results. Looking at the possible applications of the knowledge of a relationship of texture to the thermal conductivity of soil, an exact, precise formula representing this relationship would be just as useful as a general, comparative analysis, since in the field, there are many variations of soil even in the same field and in many cases on even a smaller scale. The data from this experiment establish a relationship of texture to thermal conductivity that is quite contrary to the rather well known but poorly documented fact that, in nature, coarser-grained textures (e.g., sand) have a higher thermal conductivity than finer-grained textures (e.g., clay). The relationship indicated by the data is that thermal conductivity increases as the texture size decreases. There may be several reasons for these contradictions. First, one major variable that was eliminated is water, something that drastically affects the thermal conductivity of soil in nature. Since it is extremely rare that there would be no moisture in a soil, it may ultimately be the water-holding capacity (which is itself influenced by texture) that outweighs all other variables and determines the true thermal conductivity of soil. The pore space of a soil is what determines how much liquid and gas a soil can hold; in the absence of liquid, the pore space cannot simply disappear. Because of this, we can conclude that it must then contain air, which would to some extent act to insulate the soil. Drawing from this, the results of the investigation still seem to be in contradiction with “common” knowledge, in fact, they seem even more contradictory. In nature, however, it may be the evaporative cooling due to the water that fills the pores that has a greater influence on the thermal conductivity of the soil than any other factors. The graph of thermal conductivity versus pore space supports this conclusion in that, as pore space increases, thermal conductivity decreases. While there is some crossing over of pore space between textural categories, texture does definitely influence pore space. We can from this conclude that it is ultimately the different factors such as pore space that directly determine thermal conductivity. Texture itself influences thermal conductivity only by how much it allows or facilitates those factors.
When looking at possible applications of differently textured soils on a basis of thermal conductivity, it should also be considered that other external variables, such as geography of the land and vegetative cover, might have a greater effect on the thermal conductivity of the soil than the texture itself or even the of the variables influenced by texture.
CONCLUSION
In conclusion, I have found that texture both directly and indirectly influences the thermal conductivity of soil. The direct relationship of texture to thermal conductivity is that as particle size decreases, thermal conductivity increases. Indirectly, texture affects thermal conductivity by influencing pore space and particle density, which both influence thermal conductivity. As pore space increases, thermal conductivity decreases, and as particle density increases, thermal conductivity increases. It should be noted that one variable that was completely eliminated, water, plays an important role in determining thermal conductivity of soil, and for this reason, the results of this investigation do not reflect “common knowledge” of how differently textured soils act in nature. Additionally, it is significant that this experiment showed that the thermal conductivity of any soil is a constant, and, with other factors such as color, slope, and moisture entered in, can be predicted with a great deal of accuracy.
Awards won by this project
March 13, 1997
First place in the Senior Earth & Environmental Science division of the Lancaster County Science & Engineering Fair
Environmental Science Award – from the Sigma Xi, The Scientific Research Society, which awarded one student in biology, chemistry, environmental science, psychology, and physics.
Certificate of Achievment for an outstanding project – from the United States Army, which awarded a total of 16 students in various categories.
Student Awards for Geoscience Excellence – from the Association for Women Geoscientists, which gave away one award.
Naval Science Award – from the United States Navy and Marine Core.
The competition…
At the fair, there were a total of 429 science projects, with ~40 being in the Earth and Environmental Science division