An Experimental Analysis of Environmental Factors: Surface Spread of Oil on Water

By Jackson Weir


This paper investigated the effects of several environmental factors, specifically temperature and salt concentration, on the surface spreading of crude oil on water in order to better understand the behaviour of oil after an ocean spill. Current literature does not appear to encompass adequate environmental factors in oil modeling predictions that describe the dispersion of oil after a spill. Therefore, this research aimed to explore experimentally whether water temperature and salinity significantly affect the surface spread of oil on water, and if they should be included in future oil-spill prediction softwares.

IBISWorld estimates the oil and gas industry makes up approximately 5% of the entire global economy (IBISWorld, 2017). However, with such a large global usage, error is inevitable. Unfortunately, even small errors in storage or transport can result in devastating consequences to the environment. Scientists are improving how spills are dealt with, but there is much work to be done. In one of the larger oil spills in recent memory, the Deepwater Horizon spill, Joye and her research team estimate 24-55% of the oil is still unaccounted for (Joye, 2015). Therefore, more accurate prediction models are needed to better inform decision makers that devise clean-up methods, to ultimately reduce environmental damage.

Oil Spills and Oil Modeling Systems

An oil spill is defined as the unintentional release of oil into a body of water. This paper focusing primarily on oil spills in seawater. According to the National Oceanic and Atmospheric Administration, oil spills are a common occurrence, with small-scale spills happening nearly every day in lakes, rivers and along ocean coasts in the United States alone (NOAA, 2017). While most can be dealt with easily and environmental harm can be mostly avoided, larger spills are devastating to mammals, birds, fish and plants (EPA, 1999).

Although responders are improving how they deal with spills, there is a long way to go. In an attempt to better understand how oil is distributed when coming in contact with water during a spill, scientists have identified several general behaviours of oil. These behaviours occur at the oil-water interface, defined as the surface of contact between the liquids. The International Tanker Owners Pollution Federation (ITOPF) lists the following as the most significant physical and chemical behaviours of oil: spreading, emulsification, evaporation, photo-oxidation, dissolution, dispersion, biodegradation and sedimentation (ITOPF). This paper will only directly consider spreading in its experimental analysis but gained valuable information from reviewing research done on emulsifications because the oil-water interface contributes to both spreading and emulsification, so factors affecting one are likely to affect the other.

Different oil-spill modeling softwares have been created to aid decision makers take accurate decisions after a spill to reduce environmental damage. Some of the commonly used oil-modeling systems include the Marine Environmental Modelling Workbench of the Dose-Related Exposure Assessment Model, SIMPAR oil module, General NOAA Oil Modelling Environment, Oil Weathering Model, and Automated Data Inquiry for Oil Spills ADIOS (Vos, 2005). These computer softwares are typically referred to as oil-modeling systems or oil- weathering systems and predict to what extent different oil processes will occur and their impact on the overall distribution of oil in the surrounding area. Some models are more extensive and detailed than others. as they have larger databases, both from laboratory data and actual field data.

A comparison of five current oil-spill modeling softwares, written by R.J. Vos, was the primary reference source for information on oil-modeling systems because it is the most in depth, with extensive testing and comparing. The paper breaks down how each of the five models incorporate the processes that impact the overall oil distribution after a spill in their predictions. It also tests the accuracy of four of the five modules in the laboratory (Vos, 2005). Several recommendations were made for future research with one of these suggestions being more laboratory testing.

Other research and testing done on different oil-spill models include that of Lehr, Lehr et al., Faggetter and Hall, Restrepo et al., Belore, and Mishra and Kumar (2010; 2002; 2015; 2015; 2002; 2015). They all mathematically describe the different oil processes and how they are predicted in oil modeling systems. They also all make suggestions on how to improve the models and admit that predictions could be more accurate. Restrepo et al., for example, claims accuracy can be improved by looking at the mathematical descriptions through several different approaches rather than only using one (Restrepo, 2015). Belore concluded his paper by stating that more experimental data is necessary to improve the current systems and their algorithms (Belore, 2002).

Additionally, in Kumar and Mishra’s research, it was found that “initial spill conditions and initial oil properties critically affect the evolution of oil slicks” and claim many other researchers agree (Mishra and Kumar, 2015, p. 7). This is applicable to this investigation, as it justifies the analysis of different environmental factors and how they impact oil processes. Although, as presented in research done by Vos, different oil modeling systems mathematically describe oil processes in slightly different ways, most researchers agree on the primary factors that affect each process.

Current Knowledge and Research on Emulsions

When oil comes in contact with water, it behaves in different ways. One behaviour at the oil-water interface is the formation of emulsions. Most of the recent research on oil-water systems focus on oil-water and water-oil emulsions in agitated system (Beattie and Gray-Weale, 2012; Schafer and Horbach, 2014). An emulsion occurs when droplets of one liquid appear in another while the system remains immiscible. Most emulsion studies involve elaborate equipment to stir the mixture and take measurements on the number of emulsions, frequency of emulsions, the mean interfacial surface area, etc.

Emulsions themselves are well studied but information on this oil process could be valuable to research on other processes at the oil-water interface. Equation 1, from Kraume’s emulsion study, demonstrates an important concept of the oil-water interface considered in emulsions, drop breakage. It is essentially a measure of stability in an emulsion. Kinetic energy is used to determine this value, and temperature is measured as the average kinetic energy of a system. Therefore, temperature is an influencing factor of emulsion behaviour of oil-water mixtures (Kraume, 2010).

Most emulsion papers include temperature in the experiments or mathematical descriptions. An example comes from research by Schafer and Horbach, as it was found in collected experimental results that temperature affects the dispersion of oil in water mixtures (Schafer and Horbach, 2014).

Screen Shot 2018-09-21 at 12.04.52 PM.png

Research has also been done investigating whether ion concentration in a mixture affects emulsion stability. As stated by Beattie and Grey-Weale, whether or not ion concentration affects the occurrence and stability of emulsions has been a long debate in the field (Beattie and Gray-Weale, 2012). However, the most recent research seems to show ion concentration does indeed impact the interface. Hanczyc’s research team also found that pH, and therefore ion concentration, affects the symmetry of the oil-water interface in an emulsion (Hanczyc et al. 2007). Liu and associates found that the “stability of the oil droplets depended on the balance of the Van der Waals attraction and electrical repulsion between the oil droplets in low ion concentration” (Liu et al. 2001). Consequently, ion concentration of the water in a oil-water mixture impacts the behaviour of the oil in water.

This paper does not collect data on emulsions. However, the most recent research on the oil-water interface is being done on emulsions and it is important to examine emulsion results and test whether they apply to other oil processes.

Current Knowledge and Research on Surface Spreading

Surface spreading was the oil process investigated in this paper. When oil is spilled, the bulk of the substance initially spreads along the surface of the water. Work has been done in the past to determine factors that affect the surface spreading of oil. Research started by considering the spread of liquids on solid surfaces, rather than liquid, to establish a fundamental knowledge. Barry wrote an article in 2005 mathematically describing the distribution of liquids on solid surfaces. Although not directly applicable to the spread of liquids on other liquid surfaces, the same factors would be assumed to affect both spreads. Barry summarizes the solid surface spread in three regimes: gravity-inertia regime, gravity-viscous regime and viscous surface tension regime (Barry, 2005). Temperature was considered in the surface tension regime.

On the other hand, an ITOPF paper on liquid-liquid interfaces, considering specifically oil-on-water surface spreading, attributes viscosity and volume as the primary factors contributing to the spread and does not refer to all of the regimes Barry postulated (ITOPF). This includes not mentioning temperature as a factor that affects the spread. Similarly, Kumar and Mishra mention three similar stages to those presented by Barry but there is no mention of temperature impacting surface spread (Mishra and Kumar, 2015).

Furthermore, all oil-modeling systems mentioned above describe surface spreading slightly differently but all receive comparable results when tested (Lehr, 2010; Lehr, 2002; Faggetter and Hall, 2015; Restrepo et al., 2015; Belore; Vos, 2005; Mishra and Kumar, 2015). However, not one of the modeling systems reviewed in this investigation mentions temperature or ion concentration as influential variables to surface spreading. Temperature is included when predicting other processes, but not surface spreading.

Other investigations describing the surface spreading of immiscible liquids include that done by the International Petroleum Industry Environmental Conservation Association (IPIECA). IPIECA released an article in 2015 discussing the effect of oil spills on the environment, but they did touch on different oil processes after a spill. In the description of surface spreading, colder water temperatures were claimed to impact the rate of oil spread (IPIECA, 2015). This was the only article from a review of the literature that mentioned temperature affecting surface spreading of oil on water and the claim was not supported by any evidence and no research can be found to support this claim. Therefore, temperature has not been found to be included in any oil-on-water surface spreading calculations, despite its consideration in liquid-on-solid equations.


After reviewing literature about the oil-water interface from both a surface spreading and emulsion point of view, not only is there a request for more data from researchers involved in spill modeling, there appears to be a discrepancy in the factors claimed to impact oil emulsions compared to those that impact surface spreading. Experimental evidence supports the effect of temperature and ion concentration on the interfacial behaviour of oil and water in emulsions, so theoretically, these factors also affect the surface spreading of oil on water. The IPIECA article did indeed mention that colder temperatures likely impact the spread of oil but the claim is not supported by a mathematical description or any data. Therefore, the exclusion of temperature and ion concentration in surface spreading equations warrants additional research as temperature and Na+ and Cl- concentration are fluctuating environmental variables, different in different bodies of water.

Since temperature significantly influences the behaviour of oil in water from an emulsion perspective, to the point where it is incorporated into mathematical descriptions, and temperature is included in liquid spread on solid surfaces, it is hypothesized that temperature should also affect the surface spread of oil on a water’s surface. Furthermore, ion concentration has received recent attention because of its impact on emulsions. Therefore, it is hypothesized that the ion concentration in water should affect the oil surface spread.


The impact of lab-simulated environmental factors, specifically temperature and ion concentration, on the surface spread of oil was analyzed using an experimental approach. Different quantities of crude oil were introduced to the surface of water and the spread of the oil was observed and measured using video-recording technology and geometry-measuring tools. The temperature of the liquids in the system and the salt-ion concentration of the water was varied. It was then determined to what extent temperature and ion concentration affect the surface spread of oil by analyzing the spread data collected.

The phenomenon of investigation was a chemical process. To describe this process as accurately and objectively as possible, quantitative data was collected and analyzed experimentally to explore and explain the effect of the variables of interest on spread. This approach aligns with the purpose of this project to improve future oil-modeling systems.


1. Light grade crude oil

2. Petri dish

3. Distilled water

4. NaCl

5. Thermometer

6. Ice Bath

7. Hot Plate

8. Pipette

9. Slow motion video camera (720p at 240fps)

10. Stopwatch (to the hundreth of a second)

11. White paper with 2 circles of known radii (1 cm and 2 cm)

12. 1000 mL volumetric flask

13. 100 mL volumetric flask

14. 500 mL Erlenmeyer flask

15. 500 mL beaker

16. Weighing boat

Variables and Data

The independent variables of this study were temperature and salt concentration. The dependent variable was the surface spread of oil on water. Temperature refers to the temperature of the water. Different amounts of Na+ and Cl- ions were used to vary the salt concentration of the water. Although there are different types of salt in the ocean, NaCl is the most abundant. The range of NaCl concentrations were based upon the average ocean salinity of 0.6 M (Office of Marine Programs), and temperature variation aligned with ocean temperatures in both the Atlantic Ocean and the Pacific Ocean (US Department of Commerce). The independent variables were all varied to determine if they impact the spread of oil and ultimately if they should be included in future oil-modeling systems. To measure whether or not the independent variables affect the spread of oil, the dependent variable, spread rates were calculated in change in area over time. The experimental method used involved placing crude oil on the surface of water and observing its spread over time by analyzing video footage. The variables used in this study align with the question of investigation as they will help determine whether environmental factors, the independent variables, affect oil surface spread. For this research to be valuable to mathematical models in modeling systems, discrete quantitative data is necessary.


All necessary equipment and materials were obtained before the experiments. All necessary safety procedures were taken and materials properly disposed of.

The data collection system was set up as follows. A petri dish was positioned on white paper, centered visibly above two circles of 1 cm and 2 cm radii. Immediately to the left of the dish laid a stopwatch, capable of counting to the hundreth of a second. A slow motion camera was secured approximately 30 cm above the petri dish and the timer as to capture both objects in the frame of the video.

For all variations of water temperature and salinity, the same data collection procedure was followed with the water and oil. Firstly, 30 mL of the solution of interest was added to the petri dish. Once the solution became still, the timer adjacent the petri dish and the slow motion video camera was started. After 10 seconds, the first drop of light grade crude oil was carefully released in the center of the petri dish and in correspondence with the center of the two underlying circles. At the 20 second mark on the timer, another drop was placed into the center of the solution. Drops were also released after 30 and 40 seconds for a total of 4 individual drops separated by 10 second intervals. Each drop was released above the water to avoid varying amounts of oil seeping out of the pipette. After the 4th drop was let to spread for 10 seconds, the camera was turned off and the video footage was later analyzed. A fresh petri dish was used for each trial.

The first independent variable of interest was temperature. Using an ice bath and hot plate, along with a thermometer, six different temperatures of water were analyzed. When cooling the water, a flask of distilled water was placed into the ice bath with a thermometer. When it reached the intended temperature, the water was quickly poured into a petri dish and a trial was performed. A similar process was used when heating the water. A hot plate allowed for a flask of distilled water to be heated. When the intended temperature was reached, it was quickly placed into a dish, ready for data collection. The time from which the water read the desired temperature to the first drop of oil was kept as constant as possible to minimize cooling and heating effects of the atmosphere. For temperatures of 5°C, 10°C, 15°C, 20°C, 25°C, and 30°C, three replicate trials were conducted for a total of eighteen spread rate data points per drop of oil.

The other independent variable was the ion concentration of the water. Using 58.4 g of NaCl, a 1000 mL volumetric flask, a 1.0 M stock salt solution was made. This solution was diluted in 100 mL volumetric flasks to prepare the following solutions of varying salinity for data collection: 0.0 M, 0.1 M, 0.2 M, 0.3 M, 0.4 M, 0.5 M, 0.6 M, 0.7 M, 0.8 M, 0.9 M, 1.0 M. For each solution, 3 replicate trials of video footage were collected using the aforementioned general data collection method. Tap water was used for these trials, but, assuming the contaminants of the water remained constant, salt concentrations were all relative and still make for reliable comparisons.

After slow motion video footage was collected for all temperature and salt solution trials, surface spread data was calculated and recorded in a spreadsheet. Looking at the timer in the slow motion video, an initial time was recorded when each drop came in contact with the water and a final time was recorded when the oil spread to fill the desired circular area marked underneath each petri dish. Spread rates were calculated using the time for the oil to reach a certain area and was recorded in cm2/s. For drops one, three, and four, rates were calculated relative to the underlying circle of 1 cm. For drop two, rates correspond to time to fill the 2 cm circle.

After spread rates were calculated for each trial, scatter plots were constructed for each drop comparing the spread rates of the oil and the temperature or the water salinity. Then, Pearson Product Moment Correlation tests were conducted on each of the eight sets of data represented by scatter plots. Scatter plots were shown to visually represent the relationship amongst the two variables while the Pearson tests aim to quantify the relationship relative to a linear trendline. The test gave an r value; the measure of linear correlation between the variables. The test also gave a two-sided P value to show the statistical significance of the relationship and test the null hypothesis. The two null hypotheses explored were that temperature and salinity have no significant effect of the surface spreading of oil on water. By convention, a significance interval of P = 0.05 was selected. A two-sided P value was used to consider both directions of correlation, opposed to the one-sided P value which only considers one direction.


Surface spread rates for the variation of temperature and salt concentration trials can be found at the back of this paper. Scatterplots and Pearson Correlations of the spread data for each drop of both factor variances are presented below. These analyses describe the significance varying temperature and solution salinity have on the surface spread of crude oil on water. Images taken from the slow motion camera are also shown below with the intent to ease the conceptualization of the experiment and to qualitatively compare behaviour of the crude oil not represented in spread rates.

Salt Concentration Data

Figure 1 graphs A-D show that spread rates decrease linearly with increasing salinity. The strength of these correlations are shown in table 1.

Jackson G1.png
Jackson G2.png
Jackson G3.png
Figure 1. Scatter plot of salt concentrations from drop 1 (A), drop 2 (B), drop 3 (C), and drop 4 (D).

Figure 1. Scatter plot of salt concentrations from drop 1 (A), drop 2 (B), drop 3 (C), and drop 4 (D).

Table 1. Pearson product motion correlation test for Figure 1 graphs A-D.

Table 1. Pearson product motion correlation test for Figure 1 graphs A-D.

As shown by the Pearson Product Moment Correlation, drop 2 and 3 data show a strong correlation between salinity and spread rates because |r| > 0.50. Both correlations are significant because P < 0.05 . Drop 1 data shows a moderate correlation because 0.50 > |r| > 0.30 that is also significant because P < 0.05. Drop 4 data shows a weak correlation because |r| < 0.30 and is not significant because P > 0.05. Also, corresponding with graphs in Figure 1 A-D, the negative r values show a negative correlation.

In Figure 3, the left image shows oil spreading uniformly throughout the petri dish while the right image shows small inconsistencies in the spread by breaks in the oil.

Figure 3. Images taken from slow motion footage.  Left: 0.0 M solution 10 seconds after first drop was released. Right: 1.0 M solution 10 seconds after first drop was released.

Figure 3. Images taken from slow motion footage. Left: 0.0 M solution 10 seconds after first drop was released. Right: 1.0 M solution 10 seconds after first drop was released.

Temperature Data

Figure 2 graphs A-D all show positive correlations between water temperature and spread rates. The strength of these correlations are shown in table 2.

Jackson G5.png
Jackson G6.png
Jackson G7.png
Figure 2. Scatter plots of temperature and spread rate data from drop 1 (A), drop 2 (B), drop 3 (C), and drop 4 (D).

Figure 2. Scatter plots of temperature and spread rate data from drop 1 (A), drop 2 (B), drop 3 (C), and drop 4 (D).

Table 2. Pearson product motion correlation test for Figure 2 graphs A-D comparing temperature and drop spread rate.

Table 2. Pearson product motion correlation test for Figure 2 graphs A-D comparing temperature and drop spread rate.

As shown by the Pearson Product Moment Correlation, drop 1, 3 and 4 data show a strong correlation between temperature and spread rates because |r| > 0.50. For each of these drops P < 0.05 therefore the correlations are significant. Drop 2 data shows a moderate correlation because 0.50 > |r| > 0.30. However, the correlation is not significant because P > 0.05. Also, corresponding with graphs 6-9, the positive r values show a positive correlation.

In Figure 4, the oil in the right image is much less uniform in distribution than the oil in the left image. In the left image, the oil reached a maximum spread radius and ceased all visible spread for the remainder of the trial.

Figure 4. Images taken from slow motion footage.  Left: 5°C water 10 seconds after second drop was released. Right: 30°C water 10 seconds after second drop was released.

Figure 4. Images taken from slow motion footage. Left: 5°C water 10 seconds after second drop was released. Right: 30°C water 10 seconds after second drop was released.

Discussion and Conclusion

As presented in the introduction, temperature and salt concentration are not currently included in models predicting the surface spread of crude oil during a spill. However, it has been shown experimentally by numerous researchers that these factors largely impact the behaviour of oil in other process, such as emulsions (Beattie and Gray-Weale, 2012; Hanczyc et al. 2007; Kraume, 2010; Liu et al. 2001; Schafer and Horbach, 2014). This investigation aimed to answer whether or not these factors significantly affect the surface spread of crude oil and ultimately should temperature and salt concentration be included in future prediction algorithms. Surface spread rates are presented to quantify the spread of the oil in the presence of varying factors while qualitative pictures act to support the spread rates and explain anomalies. The results of the experiment showed that temperature should be included in future oil modeling systems but more research is needed to justify the addition of salt concentration.

Salt Concentration Discussion

As shown in table 1, drops 1 to 3 of varying salinity demonstrate a statistically significant relationship between salt concentration and crude oil spread rate. The experiment showed oil spreads slower in more concentrated water. The null hypothesis was rejected and the conclusion can be made that salinity affects surface spread. However, as more oil is added to the system, the spread rates decreased. For example, the mean spread rate of drop 1 in 0.0 M water was 30.2 cm2/s while drop 4 had an average rate of 0.84 cm2/s. Additionally, the effect of salt concentration on spread was lessened as more oil was introduced. Drop 1 spread rates differed from up to 18 cm2/s while drop 4 spread rates differed by only 0.3 cm2/s. This is a likely explanation for the correlation of drop 4 lacking significance.

It has been shown experimentally by Liu, Hanczyc and other researchers that emulsion behaviour is affected by the ion concentration of the water (Hanczyc et al. 2007; Liu et al. 2001). As hypothesized, this research attests that the salinity (ion concentration) of the water also influences the surface spread of oil. Seemingly contradictory to the data presented in this investigation, Vos demonstrated in his critical comparison of five oil modeling systems, water salinity is not included in surface spread predictions (Vos, 2005).

Seeing that three of the four sets of data proved that salinity and spread rate are significantly negatively correlated, it would be reasonable to recommend the inclusion of the variable as a parameter of future prediction programs. However, the extent to which one factor influences another is more important to how far the data points deviate from a trend line. Oil spill cleanups can take weeks to months to even years. Using a spread difference of 18 cm2/s on 0.0 M versus 1.0 M water, the calculation results in a difference of approximately 2000 m2 after the oil is let to spread for two weeks. (This calculation is merely an estimate and assumes the spread rates of the outer edges of the slick do not decrease with time). Based on this estimate, there is indeed a difference in area the oil would cover over different concentrated salt water over time. However, for a difference of 1.0 M, the impact salt concentration would have on spread of the oil is not great enough to warrant the inclusion of the factor in future predictions.

Temperature Discussion

Regarding the temperature results of the investigation, drops 1, 3 and 4 all demonstrated significant correlations between water temperature and spread rates and the null hypothesis was rejected. Drop 2 lacked significance because of the drop off in 30°C rates. As seen in figure 10, the oil on higher temperature water lacks uniformity in spread. The drop off at 30°C is likely a result of the unpredictability of spread at higher temperatures. This unpredictability is also shown in the scatter plots because higher temperature spread rates deviated farther from the trendline than lower temperature rates. For example, the standard deviation of drop 1 spread rates at 5°C was calculated to be 9.44, while the standard deviation of drop 1 spread rates at 30°C is higher, at 16.35 (see Figure 1B).

As identified in the introduction, temperature was not included in the models reviewed by Vos (Vos, 2005). Temperature, however, has indeed been proven to impact emulsion occurrence on the same oil-water interface (Kraume, 2010; Schafer and Horbach, 2014). It is reasonable to assume the same factors would likely affect all processes of the same system.

Using the same calculation comparing both extremes of the environmental factors as seen above, at difference of up to 50 cm2/s results in difference of oil area coverage of approximately 6000 m2. This value is more significant than the salinity calculation. Over the 2 week span, the oil could cover over a football field greater in area in higher temperature water. Also, at 5°C, drop 2 stopped spreading all together while. At 30°C, the oil spread rapidly and more randomly. Therefore, for the aforementioned reasons, temperature should be included in future oil spill surface spread prediction algorithms.


Since temperature was previously not included in oil modeling system surface spread predictions, the finding that temperature greatly impacts spread creates a new understanding in the field of knowledge surrounding the oil-water interface and oil spill models. Additionally, although the data collected in this paper does not warrant the incorporation of ion concentration into future oil modeling systems, it does present a new factor for further considerations.

The initially proposed hypothesis was correct. Both null hypotheses were rejected, therefore, temperature and salinity have statistically significant effects on surface spreading of oil on water.

Scientists have indeed improved these oil modeling systems over time, but, predictions are far from perfect as oil is still being lost to the environment. Temperature is a varying environmental factor. This paper has shown that oil will behave differently in colder water opposed to warmer water. Therefore, if oil is spilled nearer the Equator or in the Northern Hemisphere, it will spread faster and more broken up in the warmer water near the Equator. With temperature’s addition into prediction algorithms, decision makers could be made aware of quantitative spread differences over time and observable behavioural differences of the oil. In turn, more informed decisions will be made when creating clean-up methods. Ultimately, adding temperature to future models could reduce environmental damage.


The presented conclusions are limited by the alignment of experimental conditions to realistic oil spill situations. Firstly, massive quantities of oil are released into the ocean in an actual spill compared with the fractional amounts of oil used in this experiment. The effect of temperature and salinity may have appeared lessened with larger amounts of oil. Secondly, only one type of crude oil was used for data collection. A different grade of crude oil common in oil spills may have responded differently to the variation of environmental factors. Finally, the range of salinity variation used is only representative of Atlantic Ocean averages. A wider range would be more appropriate in depicting global ocean salinity variation.

In terms of general replicability of the given results, the fact that both temperature and salinity affect the surface spread of oil on water is highly likely to be repeated because of the large spread differences between extremes of temperature and salinity trials. However, the replicability of specific spread rates is less likely because of the large deviation seen amongst each set of three trials for a given variable (see Temperature Discussion).

Recommendations for future study lie in the analysis of water salinity’s effect on spread. Although this project did not find a reason to include the environmental factor for oil spill predictions, a small range of salt concentration was used. Studying larger range would entail a more effective analysis of its impact on spread rates in all bodies of water. Also, although this paper suggests the addition of temperature into future oil modeling systems, it does not present how this factor should be included. Future research should focus on the development of a mathematical model of surface spread that includes temperature.


Barry, J. (2005, January). Estimating Rates of Spreading and Evaporation of Volatile Liquids. CEP Magazine, pp. 32-39.

Beattie, J. K., & Gray-Weale, A. (2012). Oil/Water Interface Charged by Hydroxide Ions and Deprotonated Fatty Acids: A Comment. Angewandte Chemie, 13115-13116.

Belore, R. (2002). The SL Ross Oil Spill Fate and Behaviour Model. SL Ross. EPA. (1999). Understanding Oils Spills and Oil Spill Response. Office of Marine Programs. (n.d.). Estuarine Science- Salinity. Narragansett Bay Commission.

Faggetter, B., & Hall, K. (2015, September 7). Oil Spill Modeling. Ocean Ecology.

Hanczyc, M. M., Toyota, T., Ikegami, T.,Packard, N., & Sugawara, T. (2007). Fatty Acid Chemistry at the Oil-Water Interface: Self-Propelled Oil Droplets. American Chemical Society, 9386-9391.

IBIS World. (2017, January). Global Oil & Gas Exploration & Production: Market Research Report.

IPIECA. (2015). Impacts of oil spills on marine ecology. London: IOGP.

ITOPF. (n.d.). Fate of Marine Oil Spills.

Joye, S. B. (2015). Deepwater Horizon, 5 years on. Marine Science, 349(6248), 592-593.

Kraume, M. (2010). Drop size distributions in stirred liquid/liquid systems. Technische Universität Berlin, 1-70.

Lehr, W. J. (2010). Review of modeling procedures for oil spill weathering behaviour. NOAA.

Lehr, W., Jones, R., Evans, M., Simecek-Beatty, D., & Overstreet, R. (2002). Revisions of the ADIOS oil spill model. Environmental Modelling & Software, 189-197.

Liu, X., Nakajima, M., Nabetani, H., Xu, Q., Ichikawa, S., & Sano, Y. (2001). Stability Characteristics of Dispersed Oil Droplets Prepared by the Microchannel Emulsification Method. Journal of Colloid and Interface Science, 23-30.

Mishra, A. K., & Kumar, G. S. (2015). Weathering of Oil Spill: Modeling and Analysis. Aquatic Procedia, pp. 435-442.

NOAA. (2017, October 13). Oil Spills. Office of Response and Restoration.

Paladino, E., & Maliska, C. (2006). An Hydrodynamic Model For The Calculation of Oil Spill Trajectories. SINMEC.

Restrepo, J. M., Ramírez, J. M., & Venkataramani, S. (2015). An Oil Fate Model for Shallow-Waters. Journal of Marine Science and Engineering, pp. 1504-1543.

Schafer, E., & Franziska Horbach, S. E. (2014, May 20). Modeling of Liquid−Liquid Interfacial Properties of Binary and Ternary Mixtures. Journal of Chemical & Engineering Data , 3003-3014.

Takamura, K., Loahardjo, N., Winoto, W., Buckley, J., Morrow, N. R., Kunieda, M., . . . Matsuoka, T. (2012). Spreading and Retraction of Spilled Crude Oil on Sea Water. Crude Oil Exploration on the World, 107-124.

US Department of Commerce, & NOAA National Centers for Environmental Information. (2018, April 25). Coastal Water Temperature Guide.

Vos, R. (2005). Comparison of 5 oil-weathering models. Rijkswaterstaat.


About the Author

First and foremost, I love learning. Seeking to understand how the universe works drives most of what I do. I hope that someday my pursuit of knowledge will be able to improve the lives of others. Academically, I strive in science and math courses but I also have an interest for literature. Other than that, I enjoy playing sports, salmon fishing and exploring the outdoors.