previous

Table of Contents | Landscape Clinic | Environmental Studies

next

5. Environmental Analysis

Our environmental analyses of the Harvey Mudd College landscape focused primarily on the lawn areas of campus. The motivation for this focus is two fold: grass is a very resource intensive plant (Perry [2000]) and it comprises more that 70% of the HMC landscape. Hence it is an area in which landscape change will have a profound effect. This year we examined water use, fertilizer use, pesticide use, and sustainability of the lawns and some other areas on campus and our experimental garden so we can compare the environmental benefits of resource conserving landscape as opposed to our current landscape. Because Linde Field is planted with Bermuda grass and the other turf on campus is Kikuyu and they are maintained differently, water and fertilizer use data have been divided into two categories: 'Linde Field' and the 'other turf'.

5.1. Water Use

The Harvey Mudd College landscape is watered on two separate schedules, one for the summer months (approximately March to October) and one for the winter months. The scheduling is limited to two settings because of the difficulty of changing the clocks that control the sprinklers. The system has 18 different clocks, each with up to 30 different stations. To change the watering schedule takes the grounds crew about one week. The landscape as a whole is watered with 1,479,000 gallons of water per week while on the summer schedule, with 1,083,000 gallons/week being sprayed on the 11 acres of grass (2 acres of Linde Field, 9 acres of other turf). This amounts to 7 inches/week on Linde Field and 2.9 inches/week on the rest of the turf. During the winter watering schedule, 593,000 gallons/week are sprayed on the whole landscape with 419,000 gallons/week going to the lawns. This translates to 3.4 inches/week on Linde Field and 0.8 inches/week on the rest of the grass (Table 5.1.2). A description of the calculations performed to obtain these numbers can be seen in Appendix B.

To estimate the theoretical water needs of the turf, evapotranspiration (ET) data were used. Evapotranspiration is the total water evaporated from soil and transpired through the leaves of a plant. A landscape (in this case grass) cannot use more water than that which is evapotranspired. The reference evapotranspiration (ET_o) is multiplied by a crop coefficient for a particular plant to obtain the water needs of that plant. Any extra water applied is lost into the groundwater; essentially it is wasted. Reference evapotranspiration data were obtained online from the California Irrigation Management Information System (CIMIS) located about a mile north of the college at the Three Valleys Municipal Water District. These data are provided on a daily basis; monthly totals can also be found. The crop coefficient for Kikuyu and Bermuda grasses is 0.6, meaning these warm-season turf species need 60% of the reference evapotranspiration, and some of this need will be taken care of by precipitation (Costello et al., 137). Evapotranspiration and precipitation data were used to determine that the yearly water needs of the lawns on the Harvey Mudd campus are about 6.5 million gallons (Table 5.1.1, Appendix C). If HMC installed a new sprinkler system that was more efficient and controlled by one central computer, the watering of the lawns at Harvey Mudd could be tailored to daily ET_o data and water savings per year could be as high as 28,298,700 gal/yr (Table 5.1.1). To achieve this savings watering must be even to all areas of the grass. In practice, however, some areas may need slightly more water to ensure that the water level in the soil is raised to the desired level in all areas. The benefit of a sprinkler system with a central computer, however, is that it can easily be adjusted until the water needs of all areas of the lawn are met. It should be noted that the figure in Table 5.1.1 is merely an estimate based on one year of data (12/99-11/00). If southern California were to have a drought or a very wet year, the water needs, and thus the savings, would change.

Unfortunately, there are constraints to the watering schedule imposed by the current sprinkler system (only two watering schedules are programmed per year). To design a new watering plan for the turf under the constraints of the current sprinkler system, the month with the highest ET_o rate from the summer schedule months was used to calculate the theoretical maximum water needs for the summer months. A similar calculation was used for the winter schedule months (Table 5.1.2, Appendix C and D).

Since the highest ET_o data from each season was used to make the calculations, the amount of water that would be applied under this new plan is 14,905,000 gallons per year more than is actually needed by the turf (Table 5.1.2). These numbers also do not account for precipitation, as this value cannot be known from year to year. Still, it can be seen that watering based on a worst-case scenario for the summer and winter months saves 485,700 gal/week on the summer program and 195,000 gal/week on the winter program. Averaged over the entire year these weekly savings are enough to provide water to almost 120 families of four for a year ("AWWA Research Foundation").

It is important to note that the planting beds are being watered more than the Kikuyu ("other turf") on campus (Table 5.1.2). In addition, this number was found by dividing the water not put on the grass by the area of planting and bark beds (4.5 acres). As the bark beds are not watered, the planting beds are most likely getting more water than the numbers suggest. 3.9 inches in the summer and 1.7 inches in the winter are probably much more than the plants need. For example, ivy (which is a major planting in the beds) has a landscape coefficient of 0.25 - 0.7, depending on microclimate (although 0.4 - 0.6 is probably a more reasonable range), meaning it only needs about 40-60% of ET_o (Costello et al.). The plantings are currently getting more than reference evapotranspiration.

To estimate the water needs of our experimental garden, the landscape coefficient method was used (Costello et al.). The landscape coefficient is similar to a crop coefficient in that it is a number multiplied by the reference ET to find the water needs of a particular garden or landscape. The landscape coefficient is estimated by considering the species that occur in the planting and what their approximate water needs are in the inland region we live in, the density of the plants in the planting, and the microclimate the planting occurs in (shaded, windy, near reflective surfaces, etc.). By using this method, we estimate that once our garden is established and mature, it will need 18% of ET_o to survive and look healthy (Appendix E). About half of the water needs of the garden will be covered by precipitation every year. If we account for precipitation, our garden will need only 0.425 ft/yr of water, which translates to 9,900 gallons/yr (Appendix E). A comparison of the water needs of different plans we could implement (including landscaping using resource-conserving plants) can be seen in Figure 5.1.1.

Figure 5.1.1.Current water use of lawns compared to the lawn water use under the new plan suggested
for the old system, lawn water use with a new sprinkler system, and water use of our experimental garden.
All numbers are in percent of yearly reference evapotranspiration that would be used.

5.2. Fertilizer Use

During this project, the grounds crew decided to take over the application of fertilizer and pesticides on campus. TruGreen ChemLawn (the company which serviced the HMC lawns) applied a fertilizer with a 64-0-0 (percent by weight) blend of nitrogen-phosphorous-potassium until the end of 2000. The grounds crew began applying fertilizer to the lawn in January 2001.

The grounds crew is now applying a 16-6-8 N-P-K fertilizer to the lawns. The first application occurred in early January, and the most recent application occurred March 16. The application rate for these two occasions was 4 lb/1000 ft², as suggested by the manufacturer. Thus less nitrogen is spread than previously (Table 5.2.1), but it remained to be seen whether or not this was too much.

Groundwater samples were taken from the lawn east of Thomas-Garrett hall from March 9 to April 10, about 2 times per week, and later tested for ammonium and nitrate to determine whether or not nitrogen was leaching into the groundwater (see Appendix F for methods). We expected to see the nitrogen in the form of nitrate because the negatively charged ion is repelled by the negatively charged soil particles; thus nitrate easily leaches out. In Figures 5.2.1 and 5.2.2, it can be seen that nitrogen (mostly in the form of nitrate) is leaching out of the grass and into the groundwater.

 

Figure 5.2.1. Nitrate concentrations in groundwater samples taken east of Thomas-Garrett hall.

 

Figure 5.2.2. NH₄⁺-N concentrations in groundwater samples taken east of Thomas-Garrett hall.

The Maximum Contaminant Level (MCL) of nitrate established by the Environmental Protection Agency, the State Department of Health and Services and the California Public Utilities Commission is 10 ppm nitrate as nitrogen (which equals 714 mM nitrate). The MCL is the highest level of a contaminant allowed in drinking water. Nitrate levels above the MCL pose a health risk to infants under 6 months, as high nitrate levels in drinking water can cause blue baby syndrome (a condition in which the oxygen-carrying capabilities of red blood cells are reduced) (Skipton and Hay). Water provided to Claremont from the Southern California Water Company (SCWC) is both purchased water from the Three Valleys Municipal Water District and groundwater pumped from Chino, Pomona, Canyon, and Claremont Heights groundwater basins. The groundwater in the Claremont system has nitrate levels that, while below the MCL (5.8 ppm nitrate-N), are more than half the MCL (Southern California Water Company b). The average levels of nitrate in the groundwater tested on campus were 9.8 ppm nitrate-N. Our campus may very well be contributing to the nitrate levels of the groundwater in the Claremont system.

From the ammonium and nitrate data, we estimate that the lawns are overfertilized by 1 lb N/1000 ft² per year (Appendix G). This translates to 479 lb nitrogen put on the grass per year that it does not use (and is put into the groundwater), and 2,994 lb of fertilizer that can be saved per year. This assumes that nitrogen fertilization should also be reduced to the Bermuda grass of Linde Field. However, it cannot be known if this different grass species is being overfertilized unless tests are done. The grounds crew plans to fertilize 4 times each year, so they could theoretically cut their fertilizer use from 7,664 lb/yr to 4670 lb/yr while still providing plenty of nutrients to the grass. The lawn is not using the nitrogen that is leaching below the root depth, so why put it on in the first place? The lower fertilization rate may provide benefits to the grass as well decreasing nitrogen leaching to the groundwater. Too much nitrogen makes turf more sensitive to drought stress, and increases thatch and disease susceptibility ("Appropriate Technology Transfer for Rural Areas"). Overfertilization with nitrogen also increases greatly the amount of water that turf uses (as the growth rate increases with nitrogen addition) and decreases the root number and root depth. When roots are deeper, more water is available to the grass, and thus it would not need to be watered as much (Duble).

In comparison to the fertilizer needs of the lawns on the Harvey Mudd campus, our experimental garden needs to be mulched, but does not need fertilizer, as the plants are adapted to the native soil.

5.3. Pesticide Use

The only pesticides used on the landscape at HMC are herbicides to prevent or kill weeds on the lawns and bark beds, a growth retardant to prevent fruit production on the olive trees, and a fungicide to stave off fungal infection in the sycamores. No insecticides or rodenticides are used in the landscape.

Until the fall of 2000, the majority of chemicals used on the Harvey Mudd Campus landscape were applied by TruGreen ChemLawn. TruGreen fertilized and administered weed control to the lawns approximately four times per year and also treated the olive and sycamore trees. In addition to the pesticides applied by TruGreen, the HMC grounds crew also used RoundUp Pro to kill weeds in areas that were not practical for TruGreen to treat.

Early in 2001 the grounds crew decided to take over the chemical treatment of the lawns (TruGreen is still employed to treat the trees). This move has meant substantial financial savings for the Facilities and Maintenance Department and some changes in the chemicals applied. The main product currently used for general broadleaf weed control on the lawns is Trimec, which will be applied once or twice a year as Turf Supreme 16-6-8 fertilizer + Trimec. Although it has not been applied to date, the grounds crew plans to use Pendulum, a pre-emergent herbicide, along with Trimec, to help control weeds. RoundUp Pro continues to be used for spot control of weeds, especially on the bark beds. TruGreen still applies Cleary 3336 to the sycamores and Embark 2-S to the olives.

Despite a change in brands of herbicides applied to the lawns, the active chemicals applied by the grounds crew are the same or similar to those that were used by TruGreen (Table 5.3.1). TruGreen used a mixture of Banvel, which contains Dicamba, and MCPA Amine-4, a chlorophenoxy herbicide, for broad-leaf weed control on the lawns. Trimec, the product currently being used, has three active ingredients: dicamba and two chlorophenoxy herbicides--2,4-D and mecoprop. Pendulum 2G, a pre-emergent herbicide, contains the same active ingredient, Pendimethalin, as Lesco Pre-M 60, the pre-emergent used by TruGreen.

The World Health Organization (WHO) classifies commonly used pesticides on the basis of acute toxic risk. In general these classifications are made based on LD₅₀ values. LD₅₀ is the mg of chemical per kg of test animal it takes to kill 50% of a population of test animals. The classification assumes that the chemicals are applied with proper equipment and at proper concentrations. Rankings go from Extremely Hazardous (LD₅₀ <20) to Negligible (LD₅₀ >3000). Most chemicals applied to the HMC campus are classified by the WHO as having a slight (LD₅₀ >500) or negligible risk of acute illness except for 2,4-D, which is classified as moderate (LD₅₀ 50-500) (Table 5.3.1).

A chemical with a WHO classification of slight or negligible acute toxicity does not mean that it is completely without risk. Between 1994 and 1999, 145 confirmed or probable illnesses or injuries from the pesticides used at HMC have been reported by physicians to the California Pesticide Illness Surveillance Program (CPISP 1997-2001). Although pendimethalin, the active ingredient in Pendulum and Lesco Pre-M 60, is classified as only slightly toxic, 7 out of 69 people who ingested pendimethalin (average consumption of 106 mL), experienced effects more severe than nausea, including severe retching, hematemesis and seizures (Chuang et al.). Furthermore, chemicals with structures similar to pendimethalin cause methemoglobinemia in mammals (Chuang et al.).

However, most of the concern about pesticides today is their possible implication in chronic and subtle effects that are not reflected in acute toxicity. In particular, there is growing evidence that some pesticides are carcinogenic or cause endocrine disruption, immunological impairment, or behavioral disorders. The literature on non-acute effects of pesticides is vast and contradictory. Differences in methodology and subjects or test organisms make a comprehensive critical evaluation beyond the scope of this project. We will, however, present a sample of some of the reported effects of the pesticides used recently or currently on the HMC landscape.

Some of the pesticides used at HMC have been reported to be associated with genotoxic effects and cell damage, which could lead to cancer. Chlorophenoxy herbicides have been shown to cause cell membrane damage, uncoupling of oxidative phosphorylation, and the disruption of acetyl-coenzyme A metabolism (Bradberry et al.), and 2,4-D has been implicated in the onset of apoptosis and in the disruption of cytoskeleton formation (Kaioumova et al.). Dicamba has also been found to damage DNA by decreasing unwinding rates and inducing unscheduled DNA synthesis (Perocco et al. [1990]).

Several studies suggest that the fungicide thiophanate-methyl may be a co-carcinogen. Alone thiophanate-methyl is cytotoxic, but when combined with metabolic activation, it transformed cells in culture, implicating thiophanate-methyl in multistep carcinogenesis as a possible genotoxic and/or co-carcinogenic agent (Perocco et al. [1997]). Furthermore, prolonged exposure to thiophanate-methyl induces cytochrome-P450, which could result in accelerated metabolism of co-administered drugs with important implications for cocarcinogenicity (Paolini et al.). Thiophanate-methyl also produces an acute allergic reaction in some people (Assini et al.).

Although the pesticides used at HMC are not among the prime suspects for endocrine disruption, there are a few reports linking chlorophenoxy herbicides with hormonal changes or reproductive defects. Farmers who used chlorophenoxy herbicides had significantly elevated free testosterone levels and depressed levels of follicle-stimulating hormone FSH (Garry et al.). In addition, glyphosate, the active ingredient in Roundup Pro, has been found to inhibit progesterone production and steroidogenesis (Walsh et al.).

The herbicides at HMC have not been extensively implicated in behavioral deficits; however there is one report of 2,4-D causing behavioral changes such as serotonin syndrome behaviors, catalepsy, and right-turning preference in rats (Bortolozzi et al.).

In addition, there are increasing concerns about the effects of herbicides on animals and non-target plants in the environment. Although Roundup (glyphosate) is widely used without noticeable ill effects for weed control in environmentally sensitive areas, a number of studies report that it is potentially toxic to fish and invertebrates. Roundup was found to be acutely toxic to channel catfish and bluegill sunfish at concentrations below the use formulation (Abdelghani et al.). Roundup also caused DNA damage (genotoxicity) in tadpoles (Clements et al.) and, after several generations of chronic exposure, abnormalities in snails (Pseudosuccinea columella) (Tate et al.). Pendemethalin is also toxic to fish and invertebrates. It is highly toxic to channel catfish, bluegill sunfish, and brown trout, and has a high potential to bioaccumulate in fish (US EPA [1997]). It is highly toxic to aquatic invertebrates on an acute basis, and chronic exposure may also cause reproductive impairment (US EPA [1997]).

A few effects on plants and animals have been reported for the other HMC pesticides. Triclopyr butoxyethyl ester (BEE) is highly toxic to channel catfish, bluegill sunfish, and brown trout (other forms are only slightly toxic), and it is slightly toxic to aquatic invertebrates (US EPA [1998]). It is also reported that 2,4-D "reduces successful hatching of birds... and is acutely toxic to earthworms... and fish" (Cox). Dicamba is acutely toxic to fish at concentrations as low as 28mg/L (Caux et al.).

Given these serious effects, it is important to ask the question: What are the chances that humans, other animals, or non-target plants will come into contact with the chemicals applied to the HMC landscape? Although it is difficult to accurately predict the chance or level of exposure, the risk of toxic effects is largely dependent on a chemical's persistence in the environment and its tendency to leave the area in which it is administered.

Most studies suggest that the chemicals used at HMC are, with a few exceptions, not highly likely to contaminate groundwater or persist in the environment. It is difficult to generalize, however, because persistence and degradation are highly dependent on the exact environmental conditions, such as soil type, salinity, pH, moisture, temperature, and microbial activity. One study found that under normal conditions there was little chance that runoff from Trimec would contain maximum contamination levels of 2,4-D (Moss et al.). In contrast, Dicamba is quite soluble in water (6.5 g/L at 25°C) and has a half-life as high as 555 days in soil (Caux et al.). A study preformed on a golf course with a 5% grade found that 14% of applied Dicamba and Mecoprop left the area of treatment (Shuman et al.). Pendimethalin, when applied at standard concentrations, showed little potential for lateral or vertical runoff from turfland (Lee et al.), and Shuman et al. found that only 0.01% of applied pendimethalin left a test golf course. Glyphosate is touted for its quick degradation rate in the environment; the manufacturer states that glyphosate-treated areas may be replanted as soon as one day after application. Under some conditions, however, it persists much longer. Its half-life is more than a month in groundwater in darkness (Mallat and Barcelo), and EPA documents state that glyphosate could persist as long as 3 years under some conditions (US EPA [1993]).

From these reports it is difficult to predict whether we can expect to find the pesticides applied at HMC in the environment. Several studies have, however, looked at the actual concentrations of these chemicals in groundwater. Tests done on more than 400 drinking wells across more than 20 counties in California found no verified contamination with 2,4-D, dicamba, or glyphosate ("Well Inventory Report" [1998, 1999]).

5.4. Energy Budget

Professor Robert Perry of Cal Poly Pomona has developed a method of quantifying the sustainability of landscapes by assessing the energy flow into a system (such as a lawn) and comparing it to the energy stored (in standing biomass) of that system. Sustainability, in this case, is gauged by how long it takes before the energy put into a system to maintain it exceeds the amount of energy that is stored in the biomass.

The three most visible actions that use energy to maintain a lawn are energy to pump water through the sprinklers, energy to make fertilizer that is applied to the lawn, and energy used to run maintenance equipment like lawnmowers and related service vehicles. Using standard values given to us by Professor Perry for these three energy inputs, we calculated an energy budget for the lawn at Harvey Mudd (Table 5.4.1, Appendix H).

By dividing the energy of the standing biomass of the lawn by the energy used to maintain that lawn, we see that the sustainability of the lawn is 1.7 years. This result means after 1.7 years, we are putting more energy into the grass to maintain it than we have in biomass of lawn; we are running an energy deficit. Another way to look at this is after 1.7 years, the maintenance of the lawn is releasing more carbon into the atmosphere than is stored in the grass and is using more oxygen than the grass creates. The sustainability of the lawns could increase somewhat if watering, fertilizer application, and mowing were reduced. This sustainability of 1.7 years is the sustainability of the lawns alone and not the entire campus. As stated in section 3.4, the college has nearly 690 trees growing on campus, which means that the sustainability of the entire landscape would be greater than 1.7 years, due to the fact that so much energy is stored in trees, and relatively little energy is used to maintain them.

In comparison to the lawns on campus, the sustainability of our experimental garden was found to be 240 years (Table 5.4.2, Appendix H). While this number is greater than the lifetime of these plants, it shows that a planting like our experimental garden is much more sustainable than a lawn.

previous

Table of Contents | Landscape Clinic | Environmental Studies

next