**2. Methodology**

measures to limit the nutrients from entering free-flowing water in the hydrological system

Two important seminal developments occurred in search to predict non-toxic water quality. Much of this work was conducted in Minnesota and Wisconsin, the USA, where there are thousands of lakes to study and manage. The first development was the recognition that nitrogen ions and suspended phosphorus in the water were key limiting nutrients influencing water quality. Investigators could approximate the amount of these nutrients in the water by simply placing a Secchi disk in the water and determining the depth at which the disk disappeared. The quicker it disappeared under the water, the more nutrient rich was the water and the poorer was the water quality [17]. The second development was the ability to assign phosphorus contributions to free-moving water by the land area and land cover type. By combining these contributions from the land to the hydrological calculations of water flowing through a watershed, the concentration of phosphorus in the water on a yearly basis could be estimated [18]. It was a long series of equations to estimate water quality, with each variable having a fair amount of variance. The reliability of such a series of equations to predict water quality was at first suspect because it was believed that the accuracy of each section of the equations could lead to a highly inaccurate prediction. However, as we introduce in the methodology and results sections, the set of equations produced a relatively accurate estima-

Our intent in this investigation is to illustrate how this environmental prediction methodology is applied to management watershed in the states of Michigan (Rainbow Lake and Minnesota (Sauk River Watershed), the USA. In addition, the chapter discusses how the quest to manage water in rural and urban environments continues in parts of the world, employing two examples from the People's Republic of China (PRC). The chapter also includes statements and comments from an interview (May 2017) of Dr. Jon Bryan Burley, FASLA, a landscape architect who has been engaged in teaching and writing academic papers concerning landscape for over 40 years. He has witnessed the evolving changes in this technology over this time period.

"When I was a young professor at age 27 in 1982, modeling watershed quality in Minnesota was just becoming possible," recalls Dr. Burley. "I used to have students in my landscape planning classes model land-use development within watersheds for selected lakes and prepare land-use plans to prevent a change in the perceived water quality of that lake. The Minnesota Department of Natural Resources provided publically available information concerning land cover and water quality in the state. Combined with information about rainfall and evapotranspiration, it was possible to estimate potential water quality changes due to changes in the landscape," describes Dr. Burley. "In 1990, when I went to the University of Michigan to work on my PhD this type of watershed modeling was considered as a possible topic for my dissertation. But in discussions with faculty at the School of Natural Resources, such a topic seemed far too ecologically complex to produce a meaningful dissertation. They seemed to be weary of such a long series of equations to predict water quality. I certainly understand their caution. Their concern was that if a series of 10 equations each could only explain 80% of the variance, then by the 10th equation (80% of 80% repeated ten times), the results maybe explaining less than 10% of the variance," expressed Dr. Burley. "So I pursued another topic, addressing surface mine reclamation and in the meantime, I attempted to communicate the

are also very effective and reduce the need for water treatment basins.

tion of water quality for a water body.

266 Land Use - Assessing the Past, Envisioning the Future

#### **2.1. Sauk River Watershed, Minnesota**

An overview of fundamental methodology is illustrated by a study on the Sauk River, Minnesota, conducted in the 1980s. The Sauk River is in central Minnesota on the border between the Western prairie lands and the Eastern woodlands. Before emptying into the Mississippi River, the water passes through Cedar Island Lake. This lake has experienced algae blooms and diminished water quality. Recreational housing along the lake was blamed for poor water quality. Employing the GIS technology of the time, maps of the watershed's land cover types and soils were generated (**Figures 2** and **3**) [20]. Predicted phosphorus concentrations for each sub-watershed in the study area were statistically compared to measure levels, employing Kendall's Coefficient of Concordance, a nonparametric statistical method to search for significant agreement (note that most statistical tests examine significant difference) [21]. It was discovered that the scores significantly agreed. Therefore, the modeling approach had some degree of validation. This can be surprising, as one might expect the error and increasing amount of variance to be passed from one equation to the next. Yet, at the end of the computations, the process approximates reality. Being able to estimate real conditions, the next step in the process was to examine pre-settlement conditions of the lake with current conditions. The results of the study indicated that the predicted water quality from pre-settlement times should be no different than existing land uses with agriculture and housing in the watershed. Yet, the measured phosphorus levels were higher in the lake than the model predicted. Upstream along the main course of the river, a point source of phosphorus was discovered. The water was not toxic and the effluent met discharge requirements, but the level of suspended phosphorus in the discharged water was enough to influence the water quality downstream. Removing the point-source discharge of phosphorus improved the water quality of the lake and matched predictions. "I believe the interesting part of this study was that farmers (many dairy farms) and recreational home owners were blamed for the problems. But they were not at fault and were unfairly blamed," states Dr. Burley. "A very simple fix at the food processing plant upstream would solve much of the problem. I thought this study was a very practical example concerning how ecological modeling and landscape planning could work together. But sometimes investigators believe their discoveries are much more important than the rest of the academic and professional communities believe. I am afraid I too was susceptible to that disease of over estimating the importance of my study," recalled Dr. Burley.

Back in 1983, BASIC programming was a mathematical tool to make calculations, especially repeated calculations. "In 1974, I had taken a Fortran IV programming class and found Basic programming quite easy [22]. In an afternoon, I could put together the foundation of a computer program to calculate phosphorus concentration for a lake. In a 90 minute lecture in 1985, I gave a detailed explanation of the program to a graduate level landscape architecture class. I thought I was going to be brilliant; instead I was preposterously boring. Unless one wanted to use the model, what I had to say was of little interest," reflected Dr. Burley. "My lecture did not match the interests of the audience."

Today, all this calculation work can be accomplished on a spreadsheet. "Spreadsheets were just being invented and applied back then in the early 1980s," recalled Dr. Burley. "In addition, land cover data, graphic presentation, and area tabulation are much more convenient in this era. Back then it took hundreds of hours just to code and typing to make one map," observed Dr. Burley. The handbooks and machines that could engage the old BASIC computer programming have been discarded for 25 years. "Towards the end of these studies with my students, I was using EPPL 7, a GIS program that could be facilitated with map digitization and color map, making the process must faster to obtain results," notes Dr. Burley [23].

"At Michigan State University, in the early 1990s I imagined that I might pursue such water quality studies for nearby lakes or even for lakes such as Lake Erie. ARC-GIS and other more-friendly GIS software was available to be incorporated into the study. But when I inquired about conducting such studies, the university's hydrology research institute was interested in other kinds of investigations and I met the same kind of skepticism at my school

**Figure 2.** A map of the Sauk River, Minnesota pre-settlement vegetation. Back in the mid-1980s, black and white alphanumeric GIS maps were often the standard. The technology had been transferred down from mainframe computers to the recently developed micro-computers (copyright 1983 Jon Bryan Burley, all rights reserved, used by Permission).

Comparative Approaches in Managing Wetland Environments and Land Uses: Rainbow Lake…

http://dx.doi.org/10.5772/intechopen.79323

269

Comparative Approaches in Managing Wetland Environments and Land Uses: Rainbow Lake… http://dx.doi.org/10.5772/intechopen.79323 269

for poor water quality. Employing the GIS technology of the time, maps of the watershed's land cover types and soils were generated (**Figures 2** and **3**) [20]. Predicted phosphorus concentrations for each sub-watershed in the study area were statistically compared to measure levels, employing Kendall's Coefficient of Concordance, a nonparametric statistical method to search for significant agreement (note that most statistical tests examine significant difference) [21]. It was discovered that the scores significantly agreed. Therefore, the modeling approach had some degree of validation. This can be surprising, as one might expect the error and increasing amount of variance to be passed from one equation to the next. Yet, at the end of the computations, the process approximates reality. Being able to estimate real conditions, the next step in the process was to examine pre-settlement conditions of the lake with current conditions. The results of the study indicated that the predicted water quality from pre-settlement times should be no different than existing land uses with agriculture and housing in the watershed. Yet, the measured phosphorus levels were higher in the lake than the model predicted. Upstream along the main course of the river, a point source of phosphorus was discovered. The water was not toxic and the effluent met discharge requirements, but the level of suspended phosphorus in the discharged water was enough to influence the water quality downstream. Removing the point-source discharge of phosphorus improved the water quality of the lake and matched predictions. "I believe the interesting part of this study was that farmers (many dairy farms) and recreational home owners were blamed for the problems. But they were not at fault and were unfairly blamed," states Dr. Burley. "A very simple fix at the food processing plant upstream would solve much of the problem. I thought this study was a very practical example concerning how ecological modeling and landscape planning could work together. But sometimes investigators believe their discoveries are much more important than the rest of the academic and professional communities believe. I am afraid I too was susceptible to that disease of over estimating the

Back in 1983, BASIC programming was a mathematical tool to make calculations, especially repeated calculations. "In 1974, I had taken a Fortran IV programming class and found Basic programming quite easy [22]. In an afternoon, I could put together the foundation of a computer program to calculate phosphorus concentration for a lake. In a 90 minute lecture in 1985, I gave a detailed explanation of the program to a graduate level landscape architecture class. I thought I was going to be brilliant; instead I was preposterously boring. Unless one wanted to use the model, what I had to say was of little interest," reflected Dr. Burley. "My

Today, all this calculation work can be accomplished on a spreadsheet. "Spreadsheets were just being invented and applied back then in the early 1980s," recalled Dr. Burley. "In addition, land cover data, graphic presentation, and area tabulation are much more convenient in this era. Back then it took hundreds of hours just to code and typing to make one map," observed Dr. Burley. The handbooks and machines that could engage the old BASIC computer programming have been discarded for 25 years. "Towards the end of these studies with my students, I was using EPPL 7, a GIS program that could be facilitated with map digitization and color map, making the process must faster to obtain results," notes Dr. Burley [23]. "At Michigan State University, in the early 1990s I imagined that I might pursue such water quality studies for nearby lakes or even for lakes such as Lake Erie. ARC-GIS and other

importance of my study," recalled Dr. Burley.

268 Land Use - Assessing the Past, Envisioning the Future

lecture did not match the interests of the audience."

**Figure 2.** A map of the Sauk River, Minnesota pre-settlement vegetation. Back in the mid-1980s, black and white alphanumeric GIS maps were often the standard. The technology had been transferred down from mainframe computers to the recently developed micro-computers (copyright 1983 Jon Bryan Burley, all rights reserved, used by Permission).

more-friendly GIS software was available to be incorporated into the study. But when I inquired about conducting such studies, the university's hydrology research institute was interested in other kinds of investigations and I met the same kind of skepticism at my school as I had encountered at the University of Michigan. So I did not pursue the topic. Yet I am pleased that somebody persevered, and have applied such activities to pursue what I had also envisioned," exclaimed Dr. Burley.

"It takes a special kind of landscape student to be interested in landscape planning studies. At Michigan State University, I found most landscape students were interested in site design," observed Dr. Burley. "Most students expect to make their professional careers from site development. In the landscape planning classes, students who are interested in landscape planning are interested in the shapes and patterns of land cover types (traditional landscape ecology applications) for greenways, plant preservation, wildlife habitat, and to find land that is optimal for development. No one seemed interested in lake and hydrological modeling," proposed Dr. Burley. "But in the past decade something changed. Nations such as P.R. of China, have a renewed interest in managing water, greenways, and water quality. This is a nation with a long history concerning the management of water. International landscape graduate students envision opportunities to model the environment and prepare regional land use plans. They come to study with me at Michigan State University and to learn about how to model site hydrology and water quality. However, I will retire soon. This article is an opportunity to 'pass-along' the present state of knowledge concerning predicting non-toxic water quality to an interested

Comparative Approaches in Managing Wetland Environments and Land Uses: Rainbow Lake…

http://dx.doi.org/10.5772/intechopen.79323

271

Rainbow Lake is a small private artificial lake (a lake created by placing a dam across a creek) for housing development in Gratiot County, Michigan (**Figure 4**). The valley that was formed by Pine Creek was produced when the nearby Maple River was the outlet for Glacial Lake Saginaw, sending massive amounts of glacial melt-water down the Maple River, into the Grand River, into post-glacial Lake Michigan (Lake Chicago), eventually reaching the Mississippi River. The valley that was cut by glacial melt-waters also meant that the nearby Pine Creek would also cut a small valley into the landscape to meet the Maple River. This valley was suitable for the creation of a dam and a lake. The surrounding landscape is mostly agricultural as Gratiot County has the largest percentage of agricultural land for any of Michigan's county

**Figure 4.** An image of Rainbow Lake in Gratiot Country, Michigan looking from east to west. The lake is a long winding water body through a post-glacial valley (copyright 2017 Jon Bryan Burley, all rights reserved, used by Permission).

international readership from around the world," replied Dr. Burley.

**2.2. Rainbow Lake, Michigan**

**Figure 3.** A map of the Sauk River, Minnesota study area comprised of land-uses with soil type. Based upon this information the amount of water runoff and phosphorus contribution can be estimated (copyright 1983 Jon Bryan Burley, all rights reserved, used by Permission).

"It takes a special kind of landscape student to be interested in landscape planning studies. At Michigan State University, I found most landscape students were interested in site design," observed Dr. Burley. "Most students expect to make their professional careers from site development. In the landscape planning classes, students who are interested in landscape planning are interested in the shapes and patterns of land cover types (traditional landscape ecology applications) for greenways, plant preservation, wildlife habitat, and to find land that is optimal for development. No one seemed interested in lake and hydrological modeling," proposed Dr. Burley. "But in the past decade something changed. Nations such as P.R. of China, have a renewed interest in managing water, greenways, and water quality. This is a nation with a long history concerning the management of water. International landscape graduate students envision opportunities to model the environment and prepare regional land use plans. They come to study with me at Michigan State University and to learn about how to model site hydrology and water quality. However, I will retire soon. This article is an opportunity to 'pass-along' the present state of knowledge concerning predicting non-toxic water quality to an interested international readership from around the world," replied Dr. Burley.

#### **2.2. Rainbow Lake, Michigan**

as I had encountered at the University of Michigan. So I did not pursue the topic. Yet I am pleased that somebody persevered, and have applied such activities to pursue what I had also

**Figure 3.** A map of the Sauk River, Minnesota study area comprised of land-uses with soil type. Based upon this information the amount of water runoff and phosphorus contribution can be estimated (copyright 1983 Jon Bryan

envisioned," exclaimed Dr. Burley.

270 Land Use - Assessing the Past, Envisioning the Future

Burley, all rights reserved, used by Permission).

Rainbow Lake is a small private artificial lake (a lake created by placing a dam across a creek) for housing development in Gratiot County, Michigan (**Figure 4**). The valley that was formed by Pine Creek was produced when the nearby Maple River was the outlet for Glacial Lake Saginaw, sending massive amounts of glacial melt-water down the Maple River, into the Grand River, into post-glacial Lake Michigan (Lake Chicago), eventually reaching the Mississippi River. The valley that was cut by glacial melt-waters also meant that the nearby Pine Creek would also cut a small valley into the landscape to meet the Maple River. This valley was suitable for the creation of a dam and a lake. The surrounding landscape is mostly agricultural as Gratiot County has the largest percentage of agricultural land for any of Michigan's county

**Figure 4.** An image of Rainbow Lake in Gratiot Country, Michigan looking from east to west. The lake is a long winding water body through a post-glacial valley (copyright 2017 Jon Bryan Burley, all rights reserved, used by Permission).

growing corn, soybeans, winter wheat, dairy herds, beef cattle, horse farms, apple orchards, and hardwood lumber. Surrounding the lake is a residential development, enclosed by the rural environmental agricultural matrix. The lake is used for recreational boating, summer fishing, and winter ice fishing.

To predict the water quality of the lake, one must know the soil types for the watershed, the land cover types, the amount of yearly rainfall, the evapotranspiration rate, the volume of water in the lake, and the existing phosphorus concentration in the lake. The yearly expected phosphorus concentration can be calculated by the series of equations presented in the results section of this chapter. **Table 1** illustrates the relationship of the final calculated Carlson Index Score with lake quality. The Carlson Index Score can be calculated from either phosphorus concentrations, nitrogen concentrations, or Secchi disk readings. Phosphorus estimations can be used for landscape modeling applications. **Table 2** presents the series of variables

Comparative Approaches in Managing Wetland Environments and Land Uses: Rainbow Lake…

http://dx.doi.org/10.5772/intechopen.79323

273

The trophic state of the lake is associated with the life forms and levels of nutrients found in the water body. Ultra-oligotrophic lakes are very nutrient poor and contain relatively little aquatic life in them. Oligo-mesotrophic lakes are suitable for cold-water fish species such as trout and salmon (*Salmonidae G. Cuvier*). Meso-eutrophic lake contains fish species like walleye (*Sander vitreus*; Michill, 1818). Eu-polytrophic lakes often are suitable for large-mouth bass (*Micropterus salmoides* Lacepede) and bluegill (*Lepomis macrochirus* Rafinesque). Polytrophic

**Figure 5** illustrates the land cover types within the study area; while **Figure 6** presents the soil types for the study area. A series of 11 equations (Eq. (1)–(11)) resulted in the final calculated Carlson Index Score, which was 92.52, placing the lake at the high end of a eu-polytrophic

V1 = ((1.047) ∗ S1/3.14) ∗ (L2) (1)

(derived from CAD and public maps); L2 = 4 m.

M2 = V1/S1 (2)

.

R1 = ((U6<sup>∗</sup> 0.8) + (U7<sup>∗</sup> 0.6) + (U8<sup>∗</sup> 0.7) + (B6<sup>∗</sup> 0.5) + (B7<sup>∗</sup> 0.1) + (B8<sup>∗</sup> 0.3) + (P 6<sup>∗</sup> 0.4) +(P7<sup>∗</sup> 0.2) + (P8<sup>∗</sup> 0.225) + (W6<sup>∗</sup> 0.2) + (W7<sup>∗</sup> 0.3) + (W8<sup>∗</sup> 0.125)

+ (I6<sup>∗</sup> 0.9) + (I7<sup>∗</sup> 0.85) + (I8<sup>∗</sup> 0.8) + O1)) ∗ P1 (3)

lake. The predicted results match the classification results for the lake.

V1 = 1760827.261 m3

M2 = 1.33 m (calculate).

; S1 = 1320200.00 m2

employed in calculating the Carlson Index Number.

lakes often contain no fish.

Lake volume in meter cube

where S1 = 1,320,200 m2

Mean lake depth in meters

where V1 = 1760827.261 m<sup>3</sup>

Runoff calculations

**3. Results**


**Table 1.** Trophic state lake classification and the associated Carlson Index Score.

C1 = terrestrial and other water body phosphorus supply in kg/ yr.


6 = clayey soils, 7 = loamy soils, 8 = sandy soils.

**Table 2.** The list of variables employed to calculate non-toxic water quality.

To predict the water quality of the lake, one must know the soil types for the watershed, the land cover types, the amount of yearly rainfall, the evapotranspiration rate, the volume of water in the lake, and the existing phosphorus concentration in the lake. The yearly expected phosphorus concentration can be calculated by the series of equations presented in the results section of this chapter. **Table 1** illustrates the relationship of the final calculated Carlson Index Score with lake quality. The Carlson Index Score can be calculated from either phosphorus concentrations, nitrogen concentrations, or Secchi disk readings. Phosphorus estimations can be used for landscape modeling applications. **Table 2** presents the series of variables employed in calculating the Carlson Index Number.

The trophic state of the lake is associated with the life forms and levels of nutrients found in the water body. Ultra-oligotrophic lakes are very nutrient poor and contain relatively little aquatic life in them. Oligo-mesotrophic lakes are suitable for cold-water fish species such as trout and salmon (*Salmonidae G. Cuvier*). Meso-eutrophic lake contains fish species like walleye (*Sander vitreus*; Michill, 1818). Eu-polytrophic lakes often are suitable for large-mouth bass (*Micropterus salmoides* Lacepede) and bluegill (*Lepomis macrochirus* Rafinesque). Polytrophic lakes often contain no fish.
