Printer Friendly

Non-point source pollution and land use pattern linkage: a watershed approach.

ABSTRACT

Nutrient run-off and siltation are the main cause of non-point source pollution (NPSP) in US waterways. EPA has set a total maximum daily load (TMDL) of nutrient concentration in a given body of water and watershed approach to reduce NPSP. This study develops a model that integrates land use pattern and nutrient concentration in the creeks that drain into the J.B. Converse Lake watershed in Alabama, and to identify the sources of NPSP. Water quality data for three nutrient concentrations, mainly dissolved and total nitrogen and phosphorous, and the major land use, forest, pasture/hay, nurseries and urban development, in around these creeks, were used in the analysis. The results indicated that land use had impact on the concentration of dissolved and total nitrogen. The coefficient for forest had a negative sign indicating that creeks with large forest production had lower level of nitrogen concentration. Pasture/hay and urban development had positive signs, which showed that the increase in pasture/hay production and urban development increases the nutrient concentration. With refined and detailed data on water quality, land use, rainfall and soil type, the model could be used to complement watershed approach to measure the direct contribution of each land use to NPSP and assist in the management of pollution reduction.

Key Words: concentration, non-point source pollution, nutrient, land use, watershed, and water quality.

INTRODUCTION

In contrast to point source pollution, where the kind and volume of contaminants are from an identified source, non-point pollution (NPSP) is diffused pollution in an aggregate of small contaminant distributed throughout a watershed. The major source of pollutants of waterways is runoff, which includes: agriculture, construction (urban development), forest, urban storm water, and manufacturing activities. Agriculture and forestry have been identified as the largest contributors to NPSP (USEPA, 1992). Ground and surface water quality are impacted by NPSP. Surface water such as rivers and lakes are essential, not only for the supply of drinking water but also important for recreational use, and estuaries of wild habitat. Water quality is measured by the concentrations of constituent pollutants in the water (Chesters and Schierow, 1985).

The regulation to reduce point source pollution as designed by The National Pollutant Discharge Elimination System (NPDES), under the 1977 Clean Water Act (USEPA, 1980), used fines and output restrictions that resulted in pollution reduction. However, state and local agencies have struggled with the problem of NPSP of surface waters, and NPSP has been a topic of research for the last two decades. It was estimated that NPSP was responsible for more than half the water quality problems in the US (USEPA, 1992). Control programs were not easy to establish, or maintain due to the complex nature of NPSP (Hariston, 1995). Worobec and Hogue (1992) stated that the NPSP programs implemented have generally been ineffective in controlling the release of most toxic substances and chemical run-off, especially from farmland. According to Patrick et al. (1992) the solution to NPSP, siltation and nutrient loading emanating from farmlands, lies within land use management. The USEPA 1990 data showed that siltation was the major cause of river and stream impairment, affecting 36 percent of the nation's impaired river miles. Nutrients were the second most common cause of impairment, affecting 28 percent of the nation's impaired river miles. The 1992 data for Alabama were higher than the national average siltation affecting 38 percent of the damaged river miles, and nutrients affecting 42 percent (USEPA, 1992). This finding indicated that NPSP was a problem that had to be addressed in order to successfully manage Alabama's surface water quality.

Under the Clean Water Act, States were to establish State standards for water quality under the Total Maximum Daily Load (TMDL) mandated in 1979. TMDL is the total amount of a pollutant that a body of water can handle from all sources, and put limits on the amount of pollutant that can be discharged into the river, lake or stream from all sources. The cumulative effect of pollutants from different sources is not regulated directly. The potential causes of impairment of water bodies are as varied as human activities, discharges from industrial and municipal sources, urban, agricultural and other forms of landscape modifications, and changes in flow. USEPA developed Watershed Approach Framework taking into consideration both ground and surface water flow. The approach uses a hydrologically defined geographic area and involves stakeholders in the watershed in the key management decision that allows voluntary change in activities to address the specific concern (USEPA, 1996). The proposed methodology will bring together the watershed approach and TMDL process to identify the major source of pollutants among the different land use in a given location. We used this method to evaluate the major contributor to NPSP of the J.B. Converse Lake in Alabama.

J. B. Converse Lake encompasses approximately 64,000 acres and is fed by five main creeks: Big, Crooked, Collins, Hamilton, and Juniper. The location of the lake is seen in Figure 1. These creeks drain approximately 72 percent of the watershed; Table 2 provides the land use pattern. The terrain can be described as gentle to moderately rolling and the typical land uses are for forest, pasture/hay production, nurseries, urban development, and woody wetland.

[FIGURE 1 OMITTED]

The watershed was selected primarily because the data were already collected by USGS, and they were ready to share the data set with us. The watershed was a relatively large area with sufficient land use activities that would allow us to evaluate the influence of each activity on water quality. The lake is a source of drinking water for the city of Mobile and it supports fish, wildlife, and recreational activities. It was not considered as polluted by USEPA, however, the water quality data from the pumping stations on the lake from 1994 - 1997 indicate a slight increase in levels of nutrient, especially nitrate (The Board of Water and Sewer Commission, 1998, unpublished). This problem, while not an immediate threat now, could be a problem in the future. An understanding of the relationship between land use and water quality will assist in the development of land use management with the intervention of improving and maintaining water quality.

This paper applied the watershed approach to identify nutrient concentration in the different creeks, and develops nutrient concentration/land use integrated model to identify the major source of NPSP in the J.B. Converse Lake watershed. It was hypothesized that the water quality in and around the watershed will be useful to assess the sources of NPSP on ground and surface water. The integrated model will be an input for the public service agencies and communities in the watershed in instituting land management practices to reduce NPSP, maintain and improve surface water quality.

THEORETICAL FRAMEWORK

The main characteristic of NPSP, according to Bailey and Swank (1983), and Chesters and Schierow (1985), was that the sources of discharge are diffused, and it was difficult to trace the origin of contaminants (Hariston, 1995). The main components of NPSP were sediments, nutrients, toxic material, and organic waste (Hariston, 1995), with sediments contributing the largest volume of transported materials. Nitrogen and phosphorous were identified as important water pollutants (Duttweiler and Nicholson, 1983). These nutrients were responsible for rapid plant growth and noxious algal blooms (Duttweiler and Nicholson, 1983; Hariston, 1995). The vegetative growth in water bodies could restrict navigation, reduce recreational value, cause fish kills, affect the taste in drinking water supplies, and incur cleaning costs. Phosphorous tended to be a sediment bound pollutant, and therefore, has a tendency to be lost with surface runoff and increase with stream discharge, while nitrogen was lost through both surface and sub-surface flow and decrease with stream discharge (Alberts and Spoomer, 1985).

Land use and other human activities can affect water quality through their impact on sedimentation, chemical loads, and watershed hydrology. Agriculture has been identified as the most pervasive of all NPSP sources in every region of the United States (Myers et al., 1985). USEPA reported that agricultural runoff affected 60 percent of impaired river miles, and was by far the most extensive source of pollution (USEPA, 1992). Agricultural production includes both cropland and livestock operations that have been sources of nutrient pollutants (Myers et al., 1985). Keeney (1989) stated that pasture/grassland acted as a source of nutrient pollutant when intensively managed and especially, if it was used to graze animals.

Osborne and Wiley (1988) studied the Salt Fork watershed in East Central Illinois and suggested that agricultural practices had a minimal effect on instream soluble reactive phosphorous concentrations. The results were similar to those obtained by Smith (1977) who studied the nutrient loading of major river catchments in Northern Ireland, and observed no relationship between phosphorous fertilizer usage and mean annual river concentrations of ortho-phosphate and total phosphate. Agricultural practices appeared to largely affect nitrate and total nitrogen concentration when fertilizer is applied within the watershed. Osborne and Wiley (1988), Tufford et al. (1998), and Hirose and Kuramoto (1981) showed that ammonium was positively correlated with croplands, however, nitrates and nitrites were not used in the area, and there was a predominance of reductive paddy fields within the basin studied.

Areas that have undergone urban development have also become sources of water pollution. Rainwater runoff from roofs, lawns, streets, industrial and other pervious and impervious surfaces of urban areas can become non-point sources of water pollution. Smith (1977) and Hirose and Kuramoto (1981) showed that the in-stream concentration of phosphate and nitrogen (nitrate and nitrite) within the streams increased with the increase in urbanization. The use of fertilizers around homes, parks, and golf courses does contribute to the nutrient content of urban runoff. Thomas et al. (1992) suggested that the part of the watershed where there is more urban development had more nitrate and phosphate concentration in streams.

Forestry generates a smaller volume of NPSP than agricultural production. The major NPSP from forestry activities was sedimentation, and this was associated with road building and harvesting. Forest density and forest type affected nitrogen fixation and uptake, and the presence of riparian forests directly regulated the amount of nitrogen reaching streams from upstream areas (Sollins et al. 1980). Forested areas act as a sink, or an active transformation zone for nitrate. As the forested area increased, the nitrate level downstream decreased (Basnyat et al., 1999). Studies by Tufford et al. (1998) and Jordan et al. (1993) also demonstrated that forests were characterized by lower in stream concentration of total nitrogen and total phosphorous.

Although land use is a major factor influencing NPSP, other factors such as vegetation topography, and soil type also affect the transportation of pollutants to surface water. Johnson et al. (1997) reiterated the importance of these factors, and stated that the adsorption properties of different soil types differ in sediment transport. Sandy/gravel soils had a low nutrient adsorptive capacity and a high infiltration rate. The high infiltration rate meant that it allowed for the water to move downward toward ground water; thus, reducing surface runoff. Clay soils (clay loam), on the other hand, had high nutrient adsorption capacity and a low infiltration rate. These soils tended to erode easily resulting in a high nutrient export to surface run-off (Sonzogni et al., 1980; Miller, 1979).

According to the soil survey done by USDA/Soil Conservation Service, J. B. Converse Lake watershed is dominated by Troup soils, the upper northern region consisting of Troup-Heidel-Bama soils. This soil unit is used mainly for the cultivation of crops and pastures. The limitation of this soil unit for farming is its low water holding capacity and erosivity; however, it is considered good to fair for woodland/open land and wildlife habitat. The remainder of the watershed consists of Troup-Benndale-Smithton soils. The landscape of this area is described as nearly level to hilly, with both well and poorly drained soils, and used mainly as woodlands and pasture (Hickmen and Owens, 1980).

Rainfall influences nutrient export through both its pollutant load and its influence on runoffs. Beaulac and Reckhow (1982) suggest that run-off variability is strongly influenced by the intensity, quantity, duration and seasonal distribution of rainfall. Muir (1973) studied the factors influencing the concentration of nutrients in surface water and concluded that high concentration of nitrogen in streams during peak flows could partially be attributed to the high nitrogen concentration of rainfall. Lucey and Goolsby (1993) demonstrated the effect of rainfall on concentration of nitrates in rivers; with increase discharge, there was a dilution of nitrates in surface waters and as discharge declined, nitrate concentration increased. Alberts and Spoomer (1985) concluded that rainfall, along with plant residue and soil type, were sources of nitrogen pollution in an agricultural watershed of southwestern Iowa. This study uses stream water discharge as a proxy to rainfall. Stream discharge is high when there is heavy rainfall and low when the rainfall is low.

Empirical studies used regression analysis to assess the relationship between land use and NPSP. Studies by (Basnyat et al., 1999; Johnson et al., 1997; Tufford et al. 1998; Osborne et al. 1988; and Lynstrom et al., 1978) utilized multiple regression modeling to analyze the effect of joint land uses on different nutrient content of streams. Hill (1981) used a regression model to assess the relationship between phosphorous losses and land use, soil, and topography in a study conducted in the Duffin Creek Watershed, Canada. Johnson et al., (1997) used multiple regression analysis utilizing landscape factors such as geology and average slope, to determine the magnitude and direction of the interaction between land use and landscape factors, and water chemistry. More than 50 percent of the variance in total nitrogen and nitrate concentration and more than 40 percent of the variance in the phosphate concentration were accounted for by all land use and landscape factors. Basnyat et al., (1999) developed a semilog regression model to link land use/land cover and NPSP in Alabama. Their study was conducted in Baldwin County that boarders Mobile County in the east, which is the location for the watershed covered by this study. The common feature of the two locations was that the watersheds were large enough to observe the potential effect of land use on the NPSP. The land use pattern showed that forest was the dominant activity, followed by grassland and urban development.

Most literature in the past used log linear and semi log regression models to show the relationship between land use and water quality as an indicator of the source of NPSP. Multiple land use and nutrient concentration data were used in the models; however, this paper adds water discharge in the model to see if rainfall has effect on the level of NPSP, and is able to improve in the identification of the source of NPSP.

Integrated Model

The transport of NPSP downstream from particular contributing areas can be modeled using an equation that incorporates the flow of nutrients from the various land use in and around the nearest creek. The general assumption is that the concentration of soluble nutrients in the surface water is the result of the amount of residues from agricultural chemical application. Ordinary least square regression models (Greene, 2000) were used in this study to integrate land use and the concentration of nutrients to identify the source of NPSP. The model can be implicitly defined as follows:

(1) NPSP = f(Dis,F,N,P,U)

Where the dependent variable, NPSP, is the concentration of nutrient or total suspended solids from a sample collected from a creek: total nitrogen (TN), dissolved nitrogen (DN), and dissolved phosphorous (DP). The independent variables are: Dis water discharge, measured in cubic feet per second, F the percentage of land under forest, N the percentages of land under nurseries, P the percentage of land under pasture/hay, and U the percentage of land under urban development. The model can be explicitly written in two functional forms as:

(2) log NPSP = [alpha] + [[beta].sub.1] log Dis + [[beta].sub.2] log F + [[beta].sub.3] log N + [[beta].sub.4] log P + [[beta].sub.5] log U + [epsilon]

(3) NPSP = [alpha] + [[beta].sub.1] log Dis + [[beta].sub.2] log F + [[beta].sub.3] log N + [[beta].sub.4] log P + [[beta].sub.5] log U + [epsilon]

Equation 2 and 3 respectively represents the log-linear function, and the semi-log functional forms. Where [epsilon] is the error term, [alpha] is the intercept, and [[beta].sub.1], [[beta].sub.2], [[beta].sub.3], [[beta].sub.4], and [[beta].sub.5], are coefficients to be estimated.

Based on past studies it was hypothesized that water discharge will have a positive relationship with dissolved phosphorous and negative relationship with nitrogen. Phosphorous tends to be a sediment bound pollutant, and therefore, has a tendency to be lost with surface runoff and increase with increase in water discharge to streams (Baker, 1985; Alberts and Spoomer, 1985). When water discharge increases, nitrogen concentration declines in the stream (Baker, 1985; Lucey and Goolsby, 1993).

The coefficient for forest ([[beta].sub.2]), was expected to have a negative relationship with the nutrient concentration, for both nitrogen and phosphorous. The negative relationship for forest was based on the minimal and infrequent use of fertilizers in forestry (Jordan et al., 1993).

A positive relationship was expected to hold for the coefficients for nurseries ([[beta].sub.3]), pasture/hay ([[beta].sub.4]), and urban development ([B.sub.5]). This was expected because the use of fertilizer was generally associated with these activities (Smith, 1977; Shamblen and Binder, 1996; Park et al., 1994,).

Log linear and semi-log functions were the two widely used models to relate NPSP, nutrient concentration, and land use. This study used both models and applied the selection criterion (PC-Prediction Criterion) developed by Amemiya (1980) to select the model that can best fit the given data and relationship. The selection method uses the Unconditional Mean Square Prediction Error (UMSPE) that takes into consideration the losses associated with choosing an incorrect model.

(4) PC = [[sigma].sup.2] (1 + [K.sub.1]/T)

(5) [[sigma].sup.2] = (T - K[).sup.-1][f([x.sub.t][beta][).sub.t] - f([x.sub.t] [^.[beta]])[].sup.2]

When expressed in terms of [R.sup.2] is

PC = [[sigma].sup.2](T + [K.sub.1]/T - [K.sub.1])(1 - [R.sup.2.sub.1])(TSS/T)

Where [[sigma].sup.2] is standard error of the estimate, T number of observations, K number of independent variables, X independent variables, [beta] is the coefficient to estimated, [^.[beta]] is the non-linear least squares estimator of [beta], and TSS is total sum of squares. Adjusted [R.sup.2] incorporates a penalty for reducing the degree of freedom while revealing an improvement in fit; however, PC has higher penalty for adding variables than adjusted [R.sup.2] (Judge et al., 1988). The value for the PC ranges between 0 and 1 and a smaller value indicates the goodness of fit.

Data

The study used land use data obtained from United States Geological Survey (USGS) collected primarily from results of aerial photos and surveys (USGS, 1992). Table 2 presents the land use distribution of the five major creeks in the watershed. Forests, nurseries, and pasture/hay are the dominant land use in the watershed, but with different proportions in each creek.

Water quality data collected by USGS in cooperation with other State, Municipal and Federal agencies from 1992 through 1994 were used for the analysis (USGS, AL1992-AL1996). Water quality was sampled on a regular basis, usually biweekly and monthly, in recording stations located in each creek (Big, Hamilton, Crooked, Juniper and Collins). The parameters examined were the nutrients commonly associated with NPSP, namely dissolved nitrogen (nitrates and nitrites), total nitrogen, and dissolved phosphorous. A seasonal aggregation of the water quality data was done as follows: Winter--January to March; spring--April to June; summer--July to September; and fall--October to December. There were a total of 60 observation, four seasons for the five creeks for the period covered (1992-1994). The land used for forest, pasture/hay, nurseries, and urban development was reported as percentages of the total area covered by each creek (Table 2). The sub-watersheds covered by these creeks accounted for 66 percent of the total watershed. Since the soil was dominated by Troup soil, it was assumed to be the same for all creeks and water discharge was used as a proxy for rainfall.

RESULTS AND DISCUSSION

Summary statistics of nutrient concentration in the creeks are provided in Table 1. The table shows the annual average mean concentration of dissolved phosphorous (DP), dissolved nitrogen (DN) and total nitrogen (TN)) for the different creeks. The results showed that there was variation in chemistry in the creeks. Phosphorous had the least mean concentration of 0.02 mg/L for all creeks while total nitrogen had the highest mean concentration, 0.78mg/L, 0.70 mg/L, 0.68 mg/L, 0.64 mg/L, 0.44 mg/L for Crooked, Juniper, Big, Hamilton, and Collins creek respectively.

The land distribution in Table 2 showed that Crooked creek had the lowest forest cover, 46% compared to the rest of the creeks, and had the highest concentration of total nitrogen. About 73%, 61%, 59%, and 52% of the land in Big, Juniper, Collins, and Hamilton creeks were, respectively under forest production. The production of nurseries and pasture/hay requires some level of fertilizer application. Crooked creek had the highest concentration of nitrogen among the creeks. The land distribution showed that Crooked creek had the largest area of land, 47% under nurseries and pasture/hay cultivation, compared to the rest of the creeks 24%, 37% in Big creek and Hamilton creek respectively, and 34% each in Collins creek and Juniper creek. This may be a result of the fertilizer application by the producers in the area, and the difference in land use in and around the different creeks. The water quality and the land use data were used in the regression analysis. Description of the dependent and independent variables is provided in Table 3. The prediction criterion (PC) was used to compare the prediction accuracy of the log-linear and semi-log functional forms for the given data. The result of the Prediction Criterion (PC) presented in Table 4 shows that the log-linear estimation has smaller prediction error than the semi-log estimation.

The result of the log-linear regression model for dissolved nitrogen and total nitrogen is presented in Table 5. The F statistic for dissolved nitrogen (LNDN) and total nitrogen (LNTN) models were statistically significant at 5% level and showed that there was a relationship between land use and nutrient concentration. The model for dissolved phosphorous did not show a statistically significant relationship between land use and nutrient concentration and is not reported. This may be because of the low level of phosphorous observed in the creeks. Past studies showed that phosphorous could be affected by geological factors (Thomas et al., 1974; 1992).

The regression model results for nitrogen (LNDN and (LNTN) showed that the coefficient for water discharge (LNdis) was significant and had the expected negative relationship with concentration. The nutrient concentration will be diluted with higher water discharge and hence a negative relation is noted. The coefficient for forest (LNF) had the negative sign, which indicates the inverse relationship between area of land under forest and nutrient concentration. Past studies by (Tufford et al., 1998; and Basnyat et al., 1999, 2000) had obtained similar result that is with the increase in forest area there is a decline in nitrogen concentration in the surrounding water.

Pasture/hay (LNP) and urban development (LNU) carried the expected positive sign, which shows that an increase in the production of pasture/hay and urban development would increase dissolved and total nitrogen in the lake. This implies that there was a significant use of nitrogen fertilizers in the production of hay. Past literature showed that the type of management of grasslands impacts the amount of nitrogen run-off to streams (Keeney, 1989). Intensive management practice in pasture/grasslands is expected to act as major source of nutrient concentration. Pasture and hay are the second major activity in the watershed and the positive relationship could be attributed to this land use.

The positive sign for urban development is consistent with past literature on urban development (Basnyat et al., 1999, 2000), which indicated that with an increase in urban development there would be more nutrient runoffs to the surrounding creeks. The main sources of pollutants are run off from streets, lawns, roofs, and industrial activities. The coefficients for nursery production (LNN) depicted a negative relationship with nutrient concentration. The inverse relationship in this watershed could be due a limited use of fertilizers in the nursery production.

Basnyat et al. (2000) used a whole watershed and contributing zone models, with nitrate concentration in the creeks as the dependent variable, and the land use: forest, residential area, agriculture and nurseries as independent variables. The contributing zone model had a higher [R.sup.2] and the expected negative sign for forest and positive for residential areas, agriculture and nurseries. This study used a whole watershed model and the same dependent and independent variables but included water discharge as one of the dependent variable and obtained the expected signs.

CONCLUSION

The results of this study indicated that land management practices had impact on the concentration of dissolved and total nitrogen in the creeks. Land under forest production, pasture/hay, and urban development had effect on the nutrient concentration in the creeks. Forest production acted as a sink for nitrates, as the proportion of land under forest product increased the nutrient concentration in the creek decreased. On the other hand pasture/hay and urban development had positive relationship with nutrient concentration. The expansion of pasture/hay production and urban development would have a potential adverse effect on water quality in the watershed. Urban development is only possible by turning land under forest or other use to residential and industrial activities that create pervious and impervious surface. Consequently, the reduction in forest, that acts as a sink for nutrients, and increase the impervious surface leads to more runoffs. Future expansion of urban development might have to be accompanied by better land management that will reduce runoffs and hence non-point source pollution.

The model developed in this study examined the relationship between water quality and land use to assist in identifying major source of pollutants. The results of the study is limited by the data and assumptions, however, with refined and detailed data on water quality, land use, rainfall and soil type, the model could be used to complement watershed approach in addressing pollution concern and making land management decisions in a given geographic area.
Table 1. Mean Nutrient Concentration in Big, Crooked, Collins, Hamilton,
and Juniper Creeks (1992 - 1994) in Mg/L

 Dissolved Dissolved Total
Creek Phosphorous Nitrogen Nitrogen
 (DP) (DN) (TN)

Big Creek 0.02 0.38 0.68

Collins Creek 0.03 0.20 0.44

Crooked Creek 0.02 0.51 0.78

Hamilton Creek 0.02 0.36 0.64

Juniper Creek 0.02 0.42 0.74

Source: US Geological Survey, Water Resources Data Alabama 1992 - 1994

Table 2. Percentage Land Use Distribution in the Seven Creeks of J.B.
Converse Lake Watershed, Alabama, 1992.

Land use Creeks

 Big Collins Crooked Hamilton Juniper

Forest 72.92 59.34 46.21 52.25 61.55

Nursery 11.24 13.93 17.94 15.51 13.60

Pasture/Hay/Grass 12.8 20.87 29.44 21.78 20.5

Woody Wetlands 0.81 0 1.09 5.27 0.78

Urban 2.21 5.86 5.32 5.19 3.75

Rock/Sand Clay 0.01 0 0 0 0

Source: USGS Alabama Office, Montgomery, Alabama (unpublished), 1992

Table 3. Description of variables in the regression analysis

Variables Description

Dependent Variables

LNDN log transformation dissolved nitrogen

LNTN log transformation total nitrogen

Independent variables

LNdis log transformation of water discharge

LNF log transformation of land under forest

LNP log transformation of land under pasture/hay

LNU log transformation of land under urban
 development

Table 4. Prediction Criterion for Log Linear and Semi-log Estimations

 Prediction Criterion (PC)

Model LNDN LNTN

Log-linear 0.141 0.060

Semi-log 0.150 0.061

Table 5. Regression model result for dissolved nitrogen, and total
nitrogen using the log-linear functional form

Independent Dissolved Nitrogen Total Nitrogen
Variables (LNDN) (LNTN)

Intercept -35.21 ** -18.92 **
 (19.72) (7.29)

LNdis -0.27 ** -0.01
 (0.09) (0.03)

LNF -13.60 ** -7.16 **
 (7.05) (2.61)

LNN -15.71 * -9.26 **
 (9.31) (3.44)

LNP 2.68 * 1.33 **
 (1.61) (0.59)

LNU 0.18 * -0.05 **
 (0.08) (0.01)

Adj. [R.sup.2] 0.49 0.30

F Value 15.16 8.14

Numbers in the parenthesis are standard deviations, ** significant at
0.01,


ACKNOWLEDGMENT

The study was funded in part by a USDA/CSREES Capacity Building Program. The authors would like to thank the editor and the anonymous reviewers for the invaluable suggestions and comments. However, any errors are the authors'.

REFERENCES

Alberts, E. E. and, R. G. Spoomer. 1985. Dissolved nitrogen and phosphorus in runoff from watersheds in conservation and conventional tillage. Journal of Soil and Water Conservation 40: 153-157. Soil and Water Conservation Society, Iowa.

Amemiya, T. 1980. Selection of regressors. International Economic Review. 21: 331-353.

Bailey, G.W. and R.R. Swank, 1983. Modeling agricultural nonpoint source pollution: A research perspective. In Agricultural management and water quality. F. W. Schiller and G. W. Bailey eds. Iowa State University Press, Ames.

Baker, D. B., 1985. Regional water quality impacts of intensive row crop agriculture: A Lake Erie Basin case study. Journal of Soil and Water Conservation 40: 125-132. Soil and Water Conservation Society, Iowa.

Basnyat, P., L. D. Teeter, K. M. Flyn and B. G. Lockaby, 1999. Relationships between landscape characteristics and nonpoint source pollution inputs to coastal estuaries. Environmental Management Vol.23, No.4: 539-549.

Basnyat, P., L. D. Teeter, B.G. Lockaby, and K. M. Flyann, 2000. Land Use Characteristics and Water Quality: A Methodology for Valuing of Forested Buffers. Environmental Management Vol. 26, No2:153-161.

Beaulac, M. N. and K. H. Reckhow, 1982. An examination of land use- nutrient export relationships. Water Resource Bulletin 18: 1013-1024.

Chesters, G. and L. Schierow, 1985. A Primer on Nonpoint Source Pollution. Journal of Soil and Water Conservation 40: 9 - 13. Soil and Water Conservation Society, Iowa.

Duttweiler, D. W. and, H. P. Nicholson. 1983. Environmental problems and issues of agricultural nonpoint source pollution. In Agricultural management and water quality. F. W. Schiller and G. W. Bailey eds. Iowa State University Press, Ames.

Greene, W. H., 2000. Econometric Analysis (4th edition). Prentice Hall Inc., New Jersey

Hariston, James E., 1995. Nonpoint Source (NPS) Pollution of Alabama Waters. The Alabama Extension Cooperative Extension Service, Auburn University, Alabama, UPS, 6: 95 ANR-790.

Hickmen, Glenn and Charles Owens. 1980. Soil Survey of Mobile County Alabama, Washington, D.C. USDA and Soil Conservation Service.

Hill, A. R. 1981. Stream phosphorous export from watersheds with contrasting land use in southern Ontario. Water Resources Bulletin 17:627-634.

Hirose, T. and, N. Kuramoto. 1981. Stream and water quality as influenced by land use pattern in the Kakioka Basin, Japan. Journal of Soil and Water Conservation 10: 184-188. Soil and Water Conservation Society, Iowa.

Johnson, L. B., C. Richards, G. E. Host and, J. W. Arthur. 1997. Landscape influences on water chemistry in Midwestern stream ecosystems. Freshwater Biology 37: 193-208.

Jordan, T. E., D. L. Correll and, D. E. Weller. 1993. Nutrient interception by a riparian forest receiving inputs from an adjacent cropland. Journal of Environmental Quality 22: 467 - 473.

Judge, George, R.C. Hill, W.E. Griffiths, H.Lutkepohl, and Tsoung-Chao Lee. 1988. Introduction to the Theory of Econometrics, New York: John Wiley & Sons: 845-848.

Keeney, D. R., 1989. Sources of nitrate to ground water. In Developments in agricultural and managed-forest ecology 21; Nitrogen Management and ground water protection. R. F.Follett ed. Elsevier Science Publishing Company Inc., New York.

Lucey, K. J. and, D. A. Goolsby. 1993. Effects of climatic variation over 11 years on nitrate-nitrogen concentrations in the Raccoon River, Iowa. Journal of Environmental Quality 22: 38 - 46.

Lynstrom, D. J., F. A. Rinella, D. A. Rickert, and L. Zimmerman. 1978. Multiple regression modeling approach for regional water quality management, US Environmental Protection Aagency. EPA-600/7-78-198.

Miller M. H. 1979. Contribution of Nitrogen and Phosphorous to subsurface drainage water from intensively cropped mineral and organic soils in Ontario. Journal of Environmental Quality 8: 42 - 48.

Muir, J., E.C. Seim and, R. A. Olsen. 1973. A study of the factors influencing the nitrogen and phosphorous contents of Nebraska waters. Journal of Environmental Quality 2: 466-470.

Myers, C. F., J. Meek, S. Tuller and, A. Weinberg. 1985. Nonpoint Sources of Water Pollution. Journal of Soil and Water Conservation 40: 14 - 18. Soil and Water Conservation Society, Iowa.

Osborne, Lewis L. and Michael J. Wiley, 1988. Empirical Relationships Between Land Use/Cover and Stream Water Quality in an Agricultural Watershed. Journal of Environmental Management 26: 9 - 27.

Park, S.W., S. Mostaghimi, R.A. Cooke, and P.W. McClellan. 1994. BMP impacts on watershed runoff, sediment and nutrient yield. Water Resource Bulletin 30(6), 1011-1023.

Patrick, R., F. Douglass, D. M. Palavage and, P. M. Stewart. 1992. Surface water quality: have the laws been successful. Princeton University Presss, Princeton, N.J.

Shamblen, R. G. and, D. M. Binder. 1996. The effect of watershed reservoir volume, and rainfall on nitrate levels in surface drinking water supplies. Journal of Soil and Water Conservation 51: 457-461. Soil and Water Conservation Society, Iowa.

Sonzogni, W. C., G. Chesters, D. R. Coote, D. N. Jeffs, J. C. Konrad, R. C. Ostry, and J. Robinson. 1980. Pollution from land runoff. Environmental Science and Technology 14: 148 - 153.

Smith, R. V.. 1977. Domestic and agricultural contributions to the inputs of phosphorous and nitrogen to Lough Neagh. Water Research 11: 453-459.

The Board of Water and Sewer Commission. 1998. Mobile area water and sewer service water treatment data, 1994-1998, City of Mobile, Alabama (un published document)

Thomas, D. W. and J. D. Crutchfield, 1974. Nitrate-nitrogen and phosphate-phosphorous contents of streams draining small agricultural watersheds in Kentuky. Journal of Environmental Quality 3: 46-49.

Thomas, D. W., G. R. Haszler and J. D. Crutchfield. 1992. Nitrate-nitrogen and phosphate-phosphorous in seven Kentuky streams draining small agricultural watersheds: eighteen years later. Journal of Environmental Quality 21: 147-150.

Tufford, D. L., H. N. McKellar Jr. and, J. R. Hussey. 1998. In-Stream Nonpoint source Nutrient Prediction with Land-Use Proximity and Seasonality. Journal of Soil and Water Conservation 27: 100-111. Soil and Water Conservation Society, Iowa.

USEPA. 1992. National Water Quality Inventory: 1990 Report to Congress. USEPA, Dep. of Water, Washington, DC. EPA 503/9-92/006.

USEPA. 1980. 1977 Clean Air Act Amendment Washington: U.S. Government Print.

USEPA. 1996. Watershed Protection Approach Framework, U.S. Environmental Protection Agency, http://www.epa.gov/owow/watershed/framework/ch1.html

USGS. 1992. Land Use in the J.B Converse, Lake Basin (unpublished).

USGS. AL 92-AL 96. Streamflow in and water quality and bottom material analyses of the J. B. Converse Lake basin. U.S. Geological Survey Water-Resource Investigations Report AL 92-AL 96, USGS, Montgomery, Alabama.

Worobec M. D. and C. Hogue. 1992. Toxic substances control Guide, Bureau of National Affairs Inc. Washington D.C.

Ellene Kebede

Department of Agricultural & Environmental Sciences

210 Campbell Hall

Tuskegee University

Tuskegee, AL 36088

Phone: (334) 724-4522

Email: kebede@tusk.edu

Jianbang Gan

Department of Forest Science

2135 TAMU

Texas A&M University

College Station, TX 77843

Phone: (979) 845-5059

Email: j-gan@tamu.edu

Zewdu Gebeyehu (Corresponding Author)

Department of Chemistry and Geology

Columbus State University

4225 University Avenue

Columbus, GA 31907-5645

Phone: (706) 569-3025

Email: Gebeyehu_Zewdu@colstate.edu
COPYRIGHT 2003 Alabama Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Author:Kebede, Ellene; Gan, Jianbang; Gebeyehu, Zewdu
Publication:Journal of the Alabama Academy of Science
Date:Jul 1, 2003
Words:5960
Previous Article:The causes of mid-year budget shortfalls and their impact on education funding in Alabama, 1930-2001.
Next Article:Intra-administration conflict: the CIA v. the state department in South Vietnam during the Kennedy years.
Topics:

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters