Agricultural Phosphorus Assessment in the Great Lakes Basin: 
A Case Study 

Da Ouyang, Yung-Tsung Kang, Jon Bartholic

Institute of Water Research, Michigan State University, East Lansing, MI 48823

Submitted to

The Great Lakes Commission
April 1996

Presented at The Great Lakes Agricultural Summit, April 23-24, 1996, East Lansing, MI

Abstract

Phosphorus (P) has been identified as the critical element responsible for eutrophication problems in the Great Lakes and has become an increasing concern over the last two decades. Since the 1960s, large quantities of commercial fertilizer have been applied in cropland each year to obtain higher yields, particularly in late 1970s and early 1980s, resulting in a high phosphorus build-up in agricultural soils. In addition, animal manure is another important input of phosphorus in the soil. With surface runoff and soil erosion, soil phosphorus can be delivered into surface waters either in soluble phosphorus form or in sediment-bound phosphorus. A high concentration of phosphorus leads to eutrophication and causes both environmental and economic problems. Agricultural activities are now playing a more important role in water quality. To reduce the phosphorus loading requires an understanding of the potential phosphorus sources and determining their inputs and outputs. This research is conducted to initiate the process for obtaining a better understanding of agricultural impacts in the Great Lakes Basin from a nonpoint source pollution perspective. In this report, phosphorus inputs, outputs and accumulation in the agricultural system will be examined and estimated. Phosphorus source areas with high potential risk in the Great Lakes Basin, major watersheds and the subwatershed levels will be evaluated and targeted to help with the development of prioritizing strategies in watershed management.

Background

Overenrichment of nutrients in freshwater stimulates algal and rooted aquatic plant growth, and results in oxygen depletion, fish kills, odor problem and consequently eutrophication. Eutrophication impairs the use of surface waters for recreation, fisheries, industry and drinking, and thus raises serious environmental and economic problems. Eutrophication is a natural process which can be accelerated by human activities, namely accelerated or cultural eutrophication (Thomann and Mueller 1987). Previous studies have shown that phosphorus (P) is a crucial element in the eutrophication of the Great Lakes (PLUARG 1978; IJC 1980; Gregor and Johnson 1980; and Berg 1980). Reduction of phosphorus loadings was a major focus of the 1972 Great Lake Water Quality Agreement (GLWQA) signed by both the United States and Canada, under which substantial programs were undertaken to reduce phosphorus loads from municipal and industrial sources (IJC 1980). In 1978, GLWQA was renewed by both countries committed to control phosphorus and update phosphorus loading targets for each lake. 

Phosphorus enters surface water from a variety of sources that can be categorized into two kinds of sources: point source and nonpoint source. Point sources mainly include municipal and industrial wastewater discharges. Nonpoint sources include runoff from croplands, pasturelands, forests, urban areas and atmospheric load, etc. In agricultural activities, animal wastes may be a point source from storage in animal feedlots or become a nonpoint source from transportation and application to fields as manure. 

Since the 1970's GLWQA, great progress has been made in controlling point source phosphorus loadings because of phosphorus-detergent restrictions and improved wastewater treatment facilities. Approximately 85% of the 40 largest municipal discharges in the Great Lakes basin were reduced to the 1 mg per liter phosphorus limit by 1985 and current phosphorus loadings have achieved the target loads of 1978 Agreement in all five Great Lakes (Neilson et al. 1994). It is expected that further control of the remaining point source will become increasingly less cost-effective, and thus more attention is being given to the control of nonpoint sources, particularly in agricultural runoff. In recognition that further efforts needed to be directed at nonpoint sources of phosphorus in the Lower Lakes, a Phosphorus Load Reduction Supplement was added to the Agreement in 1983. To target agricultural nonpoint sources for Lake Erie, Canada initiated a Soil and Water Environment Program during 1985-1993, which was successful in achieving its agricultural nonpoint source phosphorus loading reduction targets for Lake Erie (Neilson et al. 1994). In 1990, the U.S. Congress enacted Section 6217 of the Coastal Zone Act Reauthorization Amendments of 1990 (CZARA) requiring coastal states including the Great Lakes states, farmers and other developers to control nonpoint source pollution such as nutrients and sediments from farmlands. 

The International Reference Group on Great Lakes Pollution from Land Use Activities (PLUARG) estimated in 1976 that about 24% of all phosphorus (or 10,850 metric tons of phosphorus) entering the Great Lakes came from the basin's agricultural land (PLUARG 1978). In Ontario agricultural watersheds studied by PLUARG, phosphorus loading from cropland counted up to 70% of total phosphorus loads. Also, it has been estimated that about 60% of the total phosphorus loads currently entering Lake Erie are from rural nonpoint sources (Baker et al.1988). Previous studies suggest that high phosphorus losses from croplands are associated with intensive row crop agriculture, fertilizer additions and tillage, etc. (Nelson, and Logan 1983). Because of the diffuse nature of nonpoint sources, many challenges are still remaining to identify and control nonpoint sources. One of purposes in this study is to examine major potential agricultural phosphorus nonpoint sources in the Great Lakes Basin, which are primarily from commercial fertilizers and animal manures.

Phosphorus exists in various forms in nature. In general, however, particulate phosphorus is a major form of phosphorus transport from agricultural land to water bodies. In most watersheds where the land use is grain crops, the portion of particulate phosphorus transport is more than 90% of the total phosphorus lost from the area (Nelson et al., 1980). If watersheds have large areas of pasture and intensive animal agriculture, soluble phosphorus will account for a significant portion in the phosphorus losses (Armstrong et al. 1974). Soluble phosphorus is a form in which phosphorus is available for algal growth while particulate phosphorus tends to be a long-term source for pollution problem. Phosphorus buildup in soils is considered as a potential nonpoint source for phosphorus loading to surface water. Thus, it is important to determine phosphorus budgets in agricultural systems and identify the areas in high phosphorus buildup in order to prioritize the critical source areas and improve watershed management by reducing phosphorus loadings. In this case study, a simplified mass balance model of phosphorus (NRC 1993) is used to estimate the phosphorus balance taking into account major phosphorus inputs and outputs in the agricultural soils. 

Objectives

The objectives of this study are (1) to examine and estimate phosphorus inputs, outputs and accumulation in the agricultural system; (2) to evaluate and target agricultural phosphorus source areas and prioritize watersheds for management action; and (3) to demonstrate the use and usefulness of the Agricultural Profile Database from a nonpoint source perspective.

Data Sources

The major data source for this case study is the Agricultural Profile Database which contains a variety of agriculture related data in the Great Lakes Basin for both the U.S. side (1982, 1987, 1992) and Canadian side (1981, 1986, 1991). It includes (1) farm number, area, ownership and economics; (2) crop production; (3) livestock, poultry and other animal production; (4) agricultural chemicals and manure usage; (5) agricultural management practices, etc. The database covers the basinwide on county basis, and may be used for various purposes. The initial sources of the Agricultural Profile Database are the Censuses of Agriculture which are from both the U.S. Census Bureau and the Statistics Canada. Commercial fertilizer usage data are not included in the database. The fertilizer data was obtained from USGS for the U.S. counties (1945-1991) and from the Fertilizer Institute of Ontario for Ontario province (1966-1995). For the U.S. counties, fertilizer data in 1991 was used for the phosphorus balance analysis in 1992. Phosphorus balance analysis for Canadian counties is incomplete because fertilizer data on county level is not available. Census data by zipcode was extracted from the CD-ROM provided by the U.S. Department of Commerce. Soil testing data (Bray-Kurtz P1) in Michigan median soil was provided by the Soil Testing Lab, Michigan State University.

Phosphorus Use in Agriculture of the Great Lakes Basin

Chemical fertilizer. Phosphorus is an essential nutrient for crop growth. It is usually a limiting factor for crop yield in many areas of the world. Application of phosphorus fertilizer has contributed to a large increase in crop yield and maintains adequate soil fertility for crop planting in later periods. As a consequence, large quantities of phosphorus fertilizers were applied to obtain higher yield in the Great Lakes Basin like anywhere else in the world since the 1960s. The general trends of fertilizer phosphorus (FP) usage is shown in Figure 1. The historical data shows that phosphorus from fertilizer use increased by 49% from 143,000 metric tons in 1966 to 213,000 metric tons in 1979 on the U.S. side while it increased by 109% from 32,000 metric tons to 67,000 metric tons on the Canadian side in the Great Lakes Basin. Accordingly, soil testing phosphorus levels were increasingly high. In Michigan, soil test levels of phosphorus increased from 23 pound per acre in 1962 to 107 pound per acre in 1980 and has remained high. Phosphorus levels increased nearly 5 times during the twenty year period. For most field crops grown on mineral soil, however, there is little chance that banded phosphorus will increase crop yields when soil test phosphorus level is above 60 pounds per acre (Christenson et al. 1992). Soil test phosphorus has reached a level at which less phosphorus fertilizer application is needed in many areas of the Great Lakes Basin. Due to the awareness of environmental concerns and fertilizer recommendations, since the 1980s, total chemical fertilizer use has declined by 35% in the Great Lakes Basin, from 255,000 metric tons (1981/82) to 166,000 metric tons (1991). The application rates decreased from 20.6 kg/ha to 14.4 kg/ha in the U.S. and from 19.4 kg/ha to 13.7 kg/ha in Canada (Table 1).

Table 1 Phosphorus Fertilizer Application in Cropland of the Great Lakes Basin

Animal manure. As an alternative in fertilization, animal manure plays an important role in improving soil fertility and soil physical conditions by supplying nutrients and organic matter. Larger proportions of nutrients in manure are in complex organic forms which release available nutrients to soils after they are mineralized by micro-organisms. Compared with chemical fertilizers, manure contains low concentrations of nutrients which are also variable depending upon the type of species, ration fed and bedding material, etc. However, manure can provide a number of valuable nutrients including phosphorus to the soil. While manure has beneficial effects on crop production, it also causes environmental problems including odors and flies. Concentrated animal production provides manures for excessive addition in relatively small area of cropland. Continual use of excessive quantities of manure on the soil may result in high phosphorus accumulation, which is considered as a potential nutrient loading to the surface water. Manure phosphorus distributions by county and by zipcode in Michigan are shown in Figure 2. Manure application may, to a great extent, be responsible for high soil test phosphorus levels (Figure 3). The scenarios can be considered conservative estimation because in some cases, a few acres of cropland may receive large quantities of manures. Among the animal species, the number of cattle decreased by 17% from 1981/82 to 1991/92 while the numbers of poultry broilers and turkey increased by 17% and 29%, respectively in the Great Lakes Basin. Our calculations show that the general trends of animal manure phosphorus has declined from 116,000 metric tons in 1981/82 to 103,000 metric tons in 1991/92 due mainly to a higher portion of manure phosphorus from cattle than that from poultry in the total manure phosphorus. Table 2 shows animal numbers and concentrations in the Great Lakes Basin for 1991/92, which are calculated from the data in the Agricultural Profile Database.

Table 2 Livestock and the Concentrations in the Great Lakes Basin (1991/92)

From an agricultural viewpoint, maintaining a relatively high phosphorus level in the soil has a long-term profitability for crop production because phosphorus can remain in the soil for a long period of time without significant loss through leaching. However, a high concentration of phosphorus in the soil increases the risk of phosphorus loading to surface waters through runoff and soil erosion. Thus, there are trade-offs and optimization in fertilizer application making it profitable and environmentally sound. Crop phosphorus needs depend on crop types, yield goal and soil test P level. Fertilizer recommendations are based on these three factors and soil type as well. Phosphorus buildup in the soil is affected by phosphorus addition, soil type and climate. Regional fertilizer recommendations have been developed by different states. In Michigan, it is recommended that manure application is restricted for soils with the soil test P ranging from 150-300 pounds per acre. If the soil test P is higher than 300 pounds per acre, no manure application is recommended. 

Mass Balance of Phosphorus

In the agro-ecosystem, the flux of nutrients is an important process in the biocycling processes. As a useful tool, mass balance has been used for assessing nutrient budgets with different scales and details (NRC, 1993). The mass balance of phosphorus can be expressed as phosphorus inputs minus phosphorus outputs. Calculated phosphorus balance can be viewed as the potential accumulation of phosphorus in the soil or the potential phosphorus loading. In this study, the mass balance of phosphorus was simplified to assess only major phosphorus inputs and outputs as well as the excessive phosphorus in agricultural soils. It is not intended to take into account all possible phosphorus sources in agricultural systems such as sewage sludge. Also, different forms of phosphorus are not differentiated in mass balance although they affect bioavailability of phosphorus. Instead, total phosphorus was considered which can be used for a proxy of the bioavailable phosphorus (NRC, 1993). Because crop residues remain in the soil after harvest, phosphorus from crop residues is considered as both inputs and outputs in cropland ( i.e. it is an output for the previous year and an input for the next year). Therefore, this item in mass balance is offset. 

Phosphorus inputs. In the farming system, the primary inputs of phosphorus are from commercial fertilizers and animal manures. Fertilizers account for a larger portion in the total additions of phosphorus in the soil. In the Great Lakes Basin, fertilizer phosphorus addition is 51- 60% of the total phosphorus input while manure phosphorus accounts for about one-third of the input. The remainder is from the input of crop residues. The amount and rate of phosphorus added in the cropland vary among different counties depending upon crop type, planting areas and soil type, etc. The rates of phosphorus application in the Great Lakes region range from 4.2 kg/ha (Minnesota) to 24.1kg/ha (Illinois) (see Table 1). 

Manure phosphorus was estimated based on the phosphorus excreted by 500 kg weight animal during one year. Phosphorus content in manures is variable depending upon the animal species, ages, ration fed, bedding material, collecting facilities and measuring methods. Therefore, it is difficult to obtain exact and consistent phosphorus contents in manures. Table 3 lists the phosphorus factors from the different sources which were used to estimate manure phosphorus in this study. 

The equation used for estimating manure phosphorus is as follo