HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (26) Citing Articles via Google Scholar Google Scholar Articles by Richards, R. P. Articles by Baker, D. B. Search for Related Content PubMed PubMed Citation Articles by Richards, R. P. Articles by Baker, D. B. Agricola Articles by Richards, R. P. Articles by Baker, D. B. Related Collections Surface Water Quality Best Management Practices Nutrient Management Statistics Tillage

SPECIAL SUBMISSIONS
Findings from the USDA-sponsored Lake Erie Agricultural Systems for Environmental Quality Project

Trends in Water Quality in LEASEQ Rivers and Streams (Northwestern Ohio), 1975–1995

Water Quality Laboratory, Heidelberg College, 310 E. Market Street, Tiffin, OH 44883

* Corresponding author (prichard{at}heidelberg.edu)

Received for publication August 12, 2000.

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

 
Trends in water quality in four northwest Ohio rivers over the period 1975–1995 were identified using datasets of daily concentrations containing 4500 to 6800 observations per river during the study period. Concentrations were log-transformed prior to analysis, and adjusted for flow using a locally weighted scatterplot smoother (LOWESS) fit between log(concentration) and log(flow). Seasonality was modeled using one- and two-cycle sinusoidal oscillations and monthly additive constants. Substantial decreases in total and soluble reactive phosphorus were documented at all stations. Smaller but highly significant decreases in total Kjeldahl nitrogen were documented at all stations, and significant decreases in total suspended solids were documented at three of the four stations. Nitrate did not show significant trends at the two stations draining major watersheds, and showed significant trends in opposite directions at the two stations on smaller watersheds. Comparisons using nonparametric, nonlinear trend fits (LOWESS) suggest that changes in fertilizer and manure application rates are the most important cause of trends in phosphorus and total Kjeldahl nitrogen; point sources are insufficient to account for the phosphorus trends. The conflicting trends for nitrate are enigmatic, but may reflect diverging land use in the two smaller watersheds.

Abbreviations: ANCOVA, analysis of covariance • LEASEQ, Lake Erie Agricultural Systems for Environmental Quality • LOWESS, locally weighted scatterplot smoother • SRP, soluble reactive phosphorus • TKN, total Kjeldahl nitrogen • TP, total phosphorus • TSS, total suspended solids • USGS, United States Geological Survey • WQL, (Heidelberg College) Water Quality Laboratory

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

 
THE BASIC FRAMEWORK of the Lake Erie Agricultural Systems for Environmental Quality (LEASEQ) program is presented in the introductory paper in this series (Richards et al., 2002b); readers unfamiliar with the program are referred to that paper for general information. A major goal of the LEASEQ program is to evaluate possible trends in water quality in the study area and to determine to what extent these trends may be caused by changes in agricultural practices during the study period (1975–1995). In this paper we present the results of trend analyses for flow, total suspended solids, and four nutrient parameters, each at four sampling stations, during the period 1975–1995. We also evaluate possible linkages between the water quality trends and trends in agricultural management practices.

The Water Quality Laboratory (WQL) has intensively sampled the major rivers in the LEASEQ study area, and several of their tributaries, for up to 25 years (Table 1). The resulting databases form the raw material for the water quality component of this trend study. Data on agricultural practices were gleaned from annual Ohio Agricultural Statistics reports; data on conservation tillage were from the Conservation Tillage Information Center. These data were used for comparison with the water quality trends, in order to determine what aspects of changes in agriculture, if any, might help explain the trends in water quality. An analysis of trends in these agricultural variables is presented elsewhere (Richards et al., 2002a) in this series of papers.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

Samples were collected using refrigerated ISCO (Lincoln, NE) autosamplers located in U.S. Geological Survey (USGS) gauging stations. The samplers collected three samples per day (four per day before May 1988, when unrefrigerated samplers were in use); sample sets were collected once per week and returned to the WQL, where they were analyzed within three days of arrival in the lab. All samples taken during storm runoff events were analyzed, as was one sample per day during low flow periods.

Flows were calculated by the WQL from the USGS provisional stage reading at the time of sampling, using the current USGS rating curve for each station. Stage measurements are subject to revision and correction by USGS at the end of each water year, prior to calculation of the published mean daily flow values. The corrected stages themselves are not published. For this reason, the flows we calculated were compared with the published mean daily flows, and when clear discrepancies were identified, corrected flows were estimated from the mean daily values by referencing them to 12:00 noon and interpolating between them to estimate the flow at the time of sampling.

All analyses followed standard WQL methods, which have been developed from and are closely related to standard methods approved by the USEPA (USEPA, 1979; Greenberg et al., 1992). Total suspended solids (TSS) concentrations were determined by filtering a sample aliquot of known volume through a pre-dried and pre-weighed glass fiber filter, then redrying and reweighing the filter; results were reported as mg/L. Soluble nutrients were measured on filtrates passing through 0.45-µm membrane filters. Phosphorus was determined using the ascorbic acid colorimetric procedure. Soluble reactive phosphorus (SRP) was measured on a filtered sample without digestion, using a Technicon Autoanalyzer II (pre-1986) or a TRAACS 800 autoanalyzer (post-1986) (Bran+Luebbe, Buffalo Grove, IL). Nitrate-plus-nitrite nitrogen was determined by cadmium reduction of a filtered sample using the same equipment as SRP. This technique measured the sum of nitrate and nitrite, but nitrite concentrations in these samples (measured independently on an unreduced aliquot) were uniformly very low. Hence, the nitrate + nitrite analyses essentially reflected the nitrate concentrations in the sample. In the text and tables that follow, nitrate + nitrite will simply be referred to as nitrate (NO₃). Total phosphorus (TP) was measured on whole-water samples digested using persulfate reagent and Technicon Autoanalyzer II equipment. Total Kjeldahl nitrogen (TKN) was measured on whole-water samples digested using a sulfuric acid–mercuric sulfate–potassium sulfate reagent. The digestion converted TKN to ammonium ion, which was quantified colorimetrically. Total Kjeldahl nitrogen is the sum of ammonia and organic nitrogen. Ammonia concentrations were generally quite low in these waters, so TKN mostly reflected organic nitrogen. Reporting units were mg/L as N or P.

Data Integration and Trend Analysis
Conversion of Agricultural Data to a Watershed Basis
Agricultural statistics are presented at the county level. For this study, we assumed that the agricultural parameters were uniformly distributed in each county, estimated the percentage of each county that falls within the Maumee and Sandusky River watershed, and allocated the agricultural data accordingly. Trends and levels of these parameters are presented elsewhere in this issue (Richards et al., 2002a), as is a discussion of the large number of government programs that have influenced farming in these watersheds during the study period (Forster and Rausch, 2002).

Trend Analysis
The WQL sampling program is intentionally focused on sampling storm runoff, with three samples per day during runoff events and daily samples at other times. In order to avoid biasing the trend analysis toward the high-flow regime, daily average concentrations (flow-weighted) were calculated from the raw data, and the trend analyses were done on these daily averages.

Because flows and concentrations are usually approximately log-normally distributed, it is common practice to log transform them prior to trend analysis. This produces a better match between the data and the assumptions of regression analysis. Seasonal fluctuations in concentration and short-term fluctuations related to fluctuations in flow are two factors that greatly increase variance and hinder trend detection. Concentration data are often adjusted for these effects (Hirsch et al., 1991; Smith et al., 1987) by incorporating flow and sinusoidal seasonal terms as covariates with time in the regression analysis used to identify the trend.

In this study, concentrations and flows were log-transformed prior to analysis. However, some relationships between log-concentration and log-flow were decidedly nonlinear, so including log-flow as a (linear) covariate was inappropriate. Furthermore, there were significant temporal trends in log-flow at some stations, and incorporating flow into the general analysis might therefore introduce an apparent trend into otherwise trend-free concentration data. For this reason, the log-transformed concentration data were adjusted for flow before the trend analysis. Log-flow was detrended by taking the rough (analogous to the residuals in a regression analysis) from a LOWESS fit (Helsel and Hirsch, 1992; Cleveland, 1979) of log-flow versus time, and the flow correction was accomplished by taking the rough of a LOWESS fit of log-concentration versus detrended log-flow. The trend analysis was then performed on these flow-adjusted log-concentrations.

Trends were evaluated using analysis of covariance (ANCOVA), with time (trend) and seasonality as independent variables. Seasonal fluctuations were represented as pairs of sine and cosine functions of 2 x x time and 4 x x time. The first pair produces one maximum and one minimum per year, and the second pair produces two maxima and two minima per year. Inclusion of the two cycle terms allows a seasonal pattern more complex than a simple annual sinusoidal oscillation to be modeled. In addition, the month of the sample was included in the ANCOVA as a categorical variable, to capture other more complex seasonal patterns not adequately modeled by the sine–cosine functions. Explanatory variables that were not significant (p 0.05) were removed in a stepwise fashion, beginning with the least significant, until only significant variables remained in the model, except that sine and/or cosine functions were removed only if neither of a pair was significant. Interaction terms were not incorporated into the analysis.

Autocorrelation is a confounding issue in trend analysis, especially with closely spaced data (Loftis and Ward, 1980). In the typical situation where autocorrelation is positive, a trend result is less significant than is indicated by the regression model, which assumes no autocorrelation. To adjust for this, autocorrelation was assumed to be adequately modeled by an autoregressive process of order one (AR1), in which case a simple formula (see Richards and Baker, 1993) can be used to adjust the t value corresponding to a chosen probability level (e.g., p = 0.05). All trend results reported in this paper account for autocorrelation in this way.

Linear trends in log-concentration correspond to exponential changes in the untransformed concentration. Because of the complex, partially nonlinear modifications made to the data in the course of the analysis, the most meaningful way to present the results is to calculate the percent change in concentration over the study period, based on this exponential relationship; this is the approach adopted in this paper.

All trend analyses were done using concentrations, not loads. However, given the approach used, the results are identical whether the initial data are loads or concentrations. In particular, the rough from the flow-adjustment step is identical whether loads or concentrations are used. Also, back-transformation leads to the same estimate of percent change over the period of record whether loads or concentrations are used.

To facilitate interpretation of trends and their possible causes, LOWESS smooths of the raw data for the time trace of each parameter were prepared. The nonlinear trends shown in these plots have no associated statistical significance. However, they depict changes in the rate of increase or decrease of concentrations, and in some cases reversals in the direction of change. Information of this sort cannot be obtained from the regression approach, in which a linear change in log-concentration with time is assumed.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

 
The results of the trend analyses are presented in Table 2. Flow increased at the Maumee and Sandusky stations and decreased at the Honey and Rock Creek stations, but only the Rock Creek change was statistically significant, even though the apparent changes were as large as 30%. This reflects the shorter period of record of the Rock Creek station, and the fact that it began operation during a period of higher average flow than occurred at the beginning of the period of record for the other stations. Highly significant decreases in SRP and TP occurred at all stations. The percent decreases were particularly large for SRP, ranging from more than 70% to nearly 90%. Significant decreases also occurred at all stations for TKN, and at all but the Honey Creek Station for TSS. The two major river stations showed minor shifts in NO₃ concentration that were not statistically significant, while Honey Creek had a large increase and Rock Creek a large decrease in NO₃ concentrations.

For this reason, percent changes were also calculated directly from the daily average calculations using simple regression analysis and LOWESS fits, for selected examples. The trends indicated by these calculations are usually similar in magnitude to those derived from the more complex ANCOVA approach (Table 3). In cases where the agreement is not good, results from the regression analysis of untransformed data tend to disagree with those from the other two approaches. The results cannot be compared statistically: the LOWESS procedure does not include a measure of statistical uncertainty, and the regression analysis on untransformed data violates the assumptions of normality and independence of the residuals, and therefore statistical comparisons of the results cannot be trusted. Nonetheless, the fact that the results usually agree in direction and magnitude suggests that the trends detected in the modified data are generally applicable to the untransformed data as well.

View larger version (38K): [in this window] [in a new window]   Fig. 1. Scatterplots of flow-adjusted log-concentration of total suspended solids and soluble reactive phosphorus, showing the relationship between the Sandusky River data, the predicted values from the analysis of covariance (ANCOVA) model, and the long-term linear trend.

The LOWESS smooths of concentration and flow over time are shown in Fig. 2 through 8 . When the smoothed flows for each river are expressed as a percent of the mean of the smoothed flows for the river, the flow histories nearly coincide, with maxima about 1983 and 1992, minima about 1988 and at both extremes, and little net change over the period of record. Suspended solids concentrations show different patterns at different stations, except for the final five years, when a downward trend is seen at all stations. Total phosphorus and soluble phosphorus trend downward throughout the period of record, except for upward trends early in the records for Honey and Rock Creek. Points of inflection in some trends, and a trend reversal for SRP at the Maumee station, correspond to maxima and minima in the flow trends. Nitrate shows an upward trend until about 1990 at all stations except Rock Creek, and all stations show a downward trend since 1990. Total Kjeldahl nitrogen has minima and maxima that correspond generally to maxima and minima, respectively, in the flow trend. In addition, a general downward trend is evident. Total nitrogen is dominated by nitrate, although trends in TKN dominate at certain times. The general trend is upward until 1990 and downward since then.

View larger version (22K): [in this window] [in a new window]   Fig. 2. LOWESS smooth of flow for the period of record. For each river, flow is expressed as a percentage of the mean flow for the period of record, in order to fit all of the graphs on the same axes.

View larger version (21K): [in this window] [in a new window]   Fig. 8. LOWESS smooth of total nitrogen for the period of record. Total nitrogen was calculated as the sum of nitrate and total Kjeldahl nitrogen.

View larger version (20K): [in this window] [in a new window]   Fig. 3. LOWESS smooth of total suspended solids for the period of record.

View larger version (21K): [in this window] [in a new window]   Fig. 4. LOWESS smooth of total phosphorus for the period of record.

View larger version (22K): [in this window] [in a new window]   Fig. 5. LOWESS smooth of soluble reactive phosphorus for the period of record.

View larger version (21K): [in this window] [in a new window]   Fig. 6. LOWESS smooth of nitrate for the period of record.

View larger version (22K): [in this window] [in a new window]   Fig. 7. LOWESS smooth of total Kjeldahl nitrogen for the period of record.

Phosphorus fertilizer sales increased slightly during the early part of the study period, peaked about 1980, and have decreased since then, by nearly 40% in the Maumee basin and by 25% in the Sandusky (Richards et al., 2002a). Honey Creek and Rock Creek basins are too small to permit this analysis, but are assumed to parallel the Sandusky, of which they are a part. Phosphorus from manure application has declined fairly steadily throughout the study period, by 34% over 20 years in the Sandusky basin and by 17% in the Maumee basin.

Nitrate fertilizer sales also peaked about 1980 in the Maumee basin, and have decreased by about 28% since then. In the Sandusky basin, nitrate fertilizer sales continued to increase throughout the study period. The overall increase amounted to about 46% of the initial level. Nitrogen from manure application has decreased by 22% in the Maumee basin and 37% in the Sandusky. This is primarily ammonia and organic nitrogen (measured by total Kjeldahl nitrogen) when applied, but mineralizes with time to nitrate. Thus, decreases in manure application should lead to decreases in export of both TKN and NO₃.

Annual loads of phosphorus in these rivers are on the order of 5 to 10% of annual applications of fertilizer and manure. Thus, the changes in fertilizer and manure use are more than adequate to account for the changes in phosphorus concentrations and loads observed in the rivers. Calhoun et al. (2002) have demonstrated close temporal linkages between fertilizer sales, Bray phosphorus concentrations in soils, and SRP concentrations in river water. This appears to indicate that some of the applied phosphorus does not enter the soil phosphorus pool by adsorption onto soil particles, but is short-circuited directly into rivers and streams. Baker and Richards (2002) hypothesize that changes in management, specifically decreases in fall application of fertilizers, may play a role in the large reduction in SRP as well.

Moog and Whiting (2002) document increases in winter precipitation, runoff, and river flow; these changes alone would tend to raise concentrations of nutrients leaving the land, not lower them. They also documented an increase in temperature and decrease in snow cover in the winter months. These may be important factors in the observed increases in nitrate concentrations, because these increases are most pronounced in the winter season. Increases in precipitation and temperature and decreases in snow cover imply increased tile flow during the winter, and tile flow is the major pathway of movement of nitrate into streams. As detailed in Baker and Richards (2002), these climatic factors may also have played a role in decreased SRP concentrations.

Changes in point-source inputs are not a major contributor to these trends. As can be seen from Baker and Richards (2002), the phosphorus loads from upstream point sources in these rivers are small compared with the total loads. In fact, they are too small to account for the magnitude of the observed trends, even if they were totally eliminated (Richards and Baker, 1993). In the Maumee basin, for example, point sources of total phosphorus do not constitute more than 15% of the annual load in any year of the study period. While point-source phosphorus loads have decreased, they are a minor contributor to the overall trend. Point-source data for other parameters are not available, but the small volume of point-source effluents in these systems implies that point sources are not important contributors to trends in the other components as well.

	SUMMARY AND CONCLUSIONS

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

 
Analyses of trends in flow, sediment, and two measures of phosphorus and nitrogen were performed for the Maumee and Sandusky Rivers and for two tributaries to the Sandusky River, Rock and Honey Creeks, using daily average concentrations covering the period 1975–1995. A regression approach that treated flow-adjusted log-transformed concentration as a function of time and seasons was supplemented by LOWESS smooths of plots of concentration over time. The trend analyses document major changes in some parameters during the study period.

Flow increased at three of the four stations, but the trends were not statistically significant.

Total suspended solids decreased significantly at all stations but Honey Creek. Decreases range from 18 to 53% over the 21-year period.

Highly significant decreases in phosphorus occurred at all stations. Soluble reactive phosphorus decreased by 72 to 88%, whereas total phosphorus decreased by 29 to 58% over the 21-year period.

Nitrate showed no significant trend in the larger basins, and significant trends in opposite directions in the two smaller watersheds. Total Kjeldahl nitrogen decreased significantly at all stations; decreases ranged from 14 to 57% over the 21-year period.

The observed trends are believed to be primarily a result of changes in agricultural management. Both increases in conservation tillage and decreases in fertilizer and manure application appear to be important. Changes in climatic factors, especially during winter months, may also have had a significant influence on the observed water quality trends. Point sources are minor contributors in these watersheds. While inputs from point sources have decreased, they are relatively unimportant in explaining the water quality trends.

	ACKNOWLEDGMENTS

 
This research was funded by a grant to The Ohio State University, Case Western Reserve University, and Heidelberg College from USDA-CSREES. We thank our LEASEQ colleagues for stimulating and insightful discussions about techniques for trend analysis and about the causal relationships between these water quality trends and trends in agricultural practices.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS AND DISCUSSION SUMMARY AND CONCLUSIONS REFERENCES

 

• Baker, D.B., and R.P. Richards. 2002. Phosphorus budgets and riverine phosphorus export in northwestern Ohio watersheds. J. Environ. Qual. 31:96–108 (this issue).[Abstract/Free Full Text]
• Calhoun, F.G., J.M. Bigham, and B.K. Slater. 2002. Relationships among plant available phosphorus, fertilizer sales, and water quality in northwestern Ohio. J. Environ. Qual. 31:38–46 (this issue).[Abstract/Free Full Text]
• Cleveland, W.S. 1979. Robust locally weighted regression and smoothing scatterplots. J. Am. Stat. Assoc. 74:829–836.[Web of Science]
• Eckert, D.J., and P.W. Levison. 1999. The changing face of agriculture in Ohio's Lake Erie basin. p. A28. In R.S. Schneider (ed.) Program and Abstracts, 42nd Conf. Int. Assoc. Great Lakes Res., Case Western Reserve Univ., Cleveland, OH. 24–28 May 1999. Int. Assoc. Great Lakes Res., Ann Arbor, MI.
• Forster, D.L., and J.N. Rausch. 2002. Evaluating agricultural nonpoint-source pollution programs in two Lake Erie tributaries. J. Environ. Qual. 31:24–31 (this issue).[Abstract/Free Full Text]
• Greenberg, A.E., L.S. Clesceri, and A.D. Eaton. 1992. Standard methods for the examination of water and wastewater. 18th ed. Am. Public Health Assoc., Washington, DC.
• Helsel, D.R., and R.M. Hirsch. 1992. Statistical methods in water resources. Studies in Environ. Sci. 49. Elsevier, Amsterdam.
• Hirsch, R.M., R.B. Alexander, and R.A. Smith. 1991. Selection of methods for the detection and estimation of trends in water quality. Water Resour. Res. 27:803–813.
• Levison, P.W., and D.J. Eckert. 1998. Evolution of agriculture in the Lake Erie basin. p. 276. In 1998 Agronomy abstracts. ASA, Madison, WI.
• Loftis, J.C., and R.C. Ward. 1980. Cost-effective selection of sampling frequencies for regulatory water quality monitoring. Environ. Int. 3:297–302.
• Moog, D.B., and P.J. Whiting. 2002. Climatic and agricultural contributions to changing loads in two watersheds in Ohio. J. Environ. Qual. 31:83–89 (this issue).[Abstract/Free Full Text]
• Resource Management Associates. 1979. Land resource information for the Lake Erie drainage basin. Co-occurrence of land resource features. Volume II. Major river basins. U.S. Army Corps of Engineers, Buffalo, NY.
• Richards, R.P., and D.B. Baker. 1993. Trends in nutrient and sediment concentrations in Lake Erie tributaries, 1975–1990. J. Great Lakes Res. 19:200–211.
• Richards, R.P., and D.B. Baker. 1998. Twenty years of change: The Lake Erie Agricultural Systems for Environmental Quality (LEASEQ) project. p. 223–229. In Proc. Natl. Watershed Water Quality Symp., Washington, DC. 22–26 Sept. 1997. EPA/625/R-97/008. USEPA Office of Water, Washington, DC.
• Richards, R.P., D.B. Baker, and D.J. Eckert. 2002a. Trends in agriculture in the LEASEQ watersheds, 1975–1995. J. Environ. Qual. 31: 17–24 (this issue).[Abstract/Free Full Text]
• Richards, R.P., F.G. Calhoun, and G. Matisoff. 2002b. The Lake Erie Agricultural Systems for Environmental Quality project: An introduction. J. Environ. Qual. 31:6–16 (this issue).
• Smith, R.A., R.B. Alexander, and M.G. Wolman. 1987. Water-quality trends in the nation's rivers. Science 235:1607–1615.[Abstract/Free Full Text]
• USEPA. 1979. Methods for chemical analysis of water and wastes. EPA-600/4-79-020. Environ. Monitoring and Support Lab., Cincinnati, OH.

This article has been cited by other articles:

				R.P. Richards, D.B. Baker, and J.P. Crumrine Improved water quality in Ohio tributaries to Lake Erie: A consequence of conservation practices Journal of Soil and Water Conservation, May 1, 2009; 64(3): 200 - 211. [Abstract] [PDF]

				M.D. Tomer, T.B. Moorman, D.E. James, G. Hadish, and C.G. Rossi Assessment of the Iowa River's South Fork watershed: Part 2. Conservation practices Journal of Soil and Water Conservation, November 1, 2008; 63(6): 371 - 379. [Abstract] [PDF]

				W. H. Renwick, M. J. Vanni, Q. Zhang, and J. Patton Water Quality Trends and Changing Agricultural Practices in a Midwest U.S. Watershed, 1994-2006 J. Environ. Qual., August 8, 2008; 37(5): 1862 - 1874. [Abstract] [Full Text] [PDF]

				R. P. Richards, D. B. Baker, J. P. Crumrine, J. W. Kramer, D. E. Ewing, and B. J. Merryfield Thirty-Year Trends in Suspended Sediment in Seven Lake Erie Tributaries J. Environ. Qual., August 8, 2008; 37(5): 1894 - 1908. [Abstract] [Full Text] [PDF]

				Y. Qian, K. W. Migliaccio, Y. Wan, Y. C. Li, and D. Chin Seasonality of Selected Surface Water Constituents in the Indian River Lagoon, Florida J. Environ. Qual., January 25, 2007; 36(2): 416 - 425. [Abstract] [Full Text] [PDF]

				G. F. McIsaac and R. D. Libra Revisiting Nitrate Concentrations in the Des Moines River: 1945 and 1976-2001 J. Environ. Qual., November 1, 2003; 32(6): 2280 - 2289. [Abstract] [Full Text] [PDF]

				R. P. Richards, F. G. Calhoun, and G. Matisoff The Lake Erie Agricultural Systems for Environmental Quality Project: An Introduction J. Environ. Qual., January 1, 2002; 31(1): 6 - 16. [Abstract] [Full Text] [PDF]

				R. P. Richards, D. B. Baker, and D. J. Eckert Trends in Agriculture in the LEASEQ Watersheds, 1975-1995 J. Environ. Qual., January 1, 2002; 31(1): 17 - 24. [Abstract] [Full Text] [PDF]

				F. G. Calhoun, J. M. Bigham, and B. K. Slater Relationships among Plant Available Phosphorus, Fertilizer Sales, and Water Quality in Northwestern Ohio J. Environ. Qual., January 1, 2002; 31(1): 38 - 46. [Abstract] [Full Text] [PDF]

				F. G. Calhoun, D. B. Baker, and B. K. Slater Soils, Water Quality, and Watershed Size: Interactions in the Maumee and Sandusky River Basins of Northwestern Ohio J. Environ. Qual., January 1, 2002; 31(1): 47 - 53. [Abstract] [Full Text] [PDF]

				D. B. Moog and P. J. Whiting Climatic and Agricultural Factors in Nutrient Exports from Two Watersheds in Ohio J. Environ. Qual., January 1, 2002; 31(1): 72 - 83. [Abstract] [Full Text] [PDF]

				D. B. Moog and P. J. Whiting Climatic and Agricultural Contributions to Changing Loads in Two Watersheds in Ohio J. Environ. Qual., January 1, 2002; 31(1): 83 - 89. [Abstract] [Full Text] [PDF]

				D. B. Baker and R. P. Richards Phosphorus Budgets and Riverine Phosphorus Export in Northwestern Ohio Watersheds J. Environ. Qual., January 1, 2002; 31(1): 96 - 108. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (26) Citing Articles via Google Scholar Google Scholar Articles by Richards, R. P. Articles by Baker, D. B. Search for Related Content PubMed PubMed Citation Articles by Richards, R. P. Articles by Baker, D. B. Agricola Articles by Richards, R. P. Articles by Baker, D. B. Related Collections Surface Water Quality Best Management Practices Nutrient Management Statistics Tillage