Field-scale variation of soil phosphorus within small Alberta watersheds.
The small proportion of added P that is available for uptake by crops is typically characterized by measuring soil-test phosphorus (STP). It is well established that as concentrations of STP increase, there is a corresponding increase in P concentrations in runoff water at plot (Vadas et al. 2005; Wright et al. 2006), field (Sharpley et al. 2002), and watershed scales (Quinton et al. 2003). Thus, STP is a good indicator of P release from soil.
However, concentrations of STP can vary significantly within farm fields (Morton et al. 2000; Schepers et al. 2000), due to factors such as differential management (Cambardella and Karlen 1999; Mallarino 1996) and redistribution by erosion (Cornish et al. 2002). Soil properties that influence the capacity of a soil to adsorb P, such as clay and organic matter content (Tisdale et al. 1985), also vary at field scales (Franzen et al. 2006). The variation in STP concentration by depth of soil layer (Crozier et al. 1999) adds to the difficulty of quantifying STP levels that relate to P concentrations in runoff, since the zone of soil interaction with runoff is often limited to a few surface centimeters (Sharpley 1985). These factors should be considered when STP is used as an indicator for potential P runoff from agricultural fields.
Further complicating the relationship between concentrations of P in soil and in runoff is the observation that different areas of a landscape contribute unequally to runoff (Gburek and Sharpley 1998; Kleinman et al. 2006). Heathwaite and Dils (2000) and Cornish et al. (2002) observed that P concentrations in runoff from small plots increased with distance downslope and attributed this to the movement of P-rich material with time. The net downslope translocation of soil by tillage may be as much as 5.5 kg [m.sup.-2] [yr.sup.-1] (1 lb [ft.sup.-2] [yr.sup.-1]) (Lobb et al. 1995) and may also cause accumulation of STP in lower landform positions. Manning et al. (2001) and Mulla (1993) documented greater concentrations of STP in lower landform areas. Thus, there is potential for greater concentrations of STP to occur in areas where runoff is likely, particularly if STP accumulations occur in the zone that interacts with runoff. If greater concentrations of STP in the soil surface increase the number of saturated P sorption sites, this can increase the potential for P transport by runoff (Schroeder et al. 2004;Vadas et al. 2005). Pote et al. (1996), Schroeder et al. (2004), and Casson et al. (2006) documented significant relationships between the degree of soil P saturation (DPS) and P concentrations in runoff.
As such, the use of traditional agronomic soil sampling methods to characterize concentrations of STP may not adequately represent the soil P pool that is susceptible to runoff. Although geostatistical techniques and different sizes of soil sampling grids have been used as alternative methods of characterizing STP variation within farm fields (Chang et al. 2003; Needelman et al. 2001), decreased grid sizes are needed when the variability of soil parameters increases (Wollenhaupt et al. 1994). Since the variability and concentration of STP are often unknown before soil sampling, this problem adds to the impracticality of expensive grid-based soil sampling strategies.
The use of topographic characteristics that relate to soil properties and determine the nature of erosion and runoff may provide a useful and intuitive means of characterizing field-scale variations in STP and DPS. MacMillan et al. (2000) and Pennock et al. (1987) used field slope and curvature to separate landform classes with different soil properties and fertility characteristics that may also influence STP variation. Another method of characterizing variations in topography that has been related to STP concentrations (Moore et al. 1993; Page et al. 2005) is the topographic index of Quinn et al. (1995).
The objectives of this study were to determine the field-scale variation of STP and DPS in soils and to relate these parameters to topographic characteristics in eight small watersheds in Alberta.
Materials and Methods
The field-scale microwatershed sites had relatively uniform management within each site and included one ungrazed grassland site near Stavely (STV); five cultivated, non-manured sites near Crowfoot Creek (CFT), Grande Prairie Creek (GPC), Renwick Creek (REN), Threehills Creek (THC), and Wabash Creek (WAB); and two cultivated, manured sites near Ponoka (PON) and the Lower Little Bow River (LLB) (figure 1). A range of management conditions were represented at the sites, from no-tillage to conventional tillage (i.e., fall tillage with a chisel plow or disc) and from manured to non-manured (table 1). The manured sites received solid cattle manure that was incorporated within 48 hours of application. Since the CFT and LLB sites were in areas with lower amounts of annual precipitation, larger areas were chosen to increase the opportunity for runoff collection in a companion study to relate soil P to concentrations of P in runoff (Little et al. 2006).
Soil Sampling Points. The microwater-shed boundaries of each site were identified using a digital elevation model (DEM). The soil sampling strategy within each micro-watershed was directed toward detecting differences in STP due to topography. To define topographic classes, the DEM was used to classify the microwatershed into upper, mid and lower landform classes (MacMillan et al. 2000). The DEM was also used to calculate a topographic index using the ln([alpha]/tan [beta]), where [alpha] is the upslope contributing area for a given point and [beta] is the local surface slope angle (Quinn et al. 1995).
A minimum of 18 points were sampled at each site, in six transects according to landform position (upper, mid, and lower). Additional sampling points were identified within each landform class to represent a range of topographic indices. The average sampling density was one sample per 1 to 5 ha (2 to 12 ac) (n = 22 to 48), with numbers of samples increasing with increased site area. The exception was the 2 ha (5 ac) STV site where only three sampling points were selected to represent the three landform positions. A total of 228 points were sampled among all sites. A Differential Global Positioning System, accurate to less than 1 m (39 in), was used to geo-reference all sampling points for the identification of landform class and topographic index.
A frame-excavation method of sampling representative proportions of banded P and soil was used in this study (Nolan et al. 2006). An 11 cm by 60 cm (4 to 24 in) steel frame, with a length that was adjusted to two times the fertilizer bandwidth was placed perpendicularly to the seed rows and fertilizer bands. Soil samples were excavated from the 0 to 2.5 cm, 2.5 to 5 cm, and 5 to 15 cm (0 to 1 in, 1 to 2 in, and 2 to 6 in) layers. One frame per sampling point was used for non-manured fields and two frames per sampling point were used at the manured and the STV sites. Composite samples were prepared for each depth from the two frame samples collected at each sampling point at the manured and STV sites. The excavated soil in each layer was thoroughly mixed in the field, and a 500 g (17.5 oz) subsample was removed and shipped in coolers with ice packs to the laboratory. The sites were sampled several times during a three-year period; however, only the samples collected in the fall of 2003 were used for this study Fall sampling in 2003 was completed after all field operations had been completed.
Soil-test Phosphorus. Soil samples were dried and ground to pass through a 2 mm (0.1 in) sieve, and a 5 g (0.2 oz) subsample was removed for STP analysis. Samples were analyzed using a modified Kelowna extraction method (0.015 M N[H.sub.4]F, 0.25 M HOAc, 0.25 M N[H.sub.4]OAc; Qian et al. 1991). Phosphorus concentrations in the extract were determined using the Murphy and Riley (1962) method and a Technicon AutoAnalyzer (Technicon Instrument Corp., Tarrytown, New York). Values of STP were calculated for the 0 to 5 cm (0 to 2 in) and 0 to 15 cm (0 to 6 in) soil layers by proportional weighting.
Degree of Phosphorus Saturation. A calcium chloride method (Casson et al. 2006) was used to measure the phosphorus sorption index (PSI) of each soil in the 0 to 2.5 cm (0 to 1 in) layer from a subsample of six transects representing upper, mid and lower landform classes (18 points) per site. The degree of P saturation (DPS) was calculated using the equation in Indiati and Sequi (2004), as the ratio of PSI to PSI plus STP.
Soil Characterization. The soil in the upper, mid, and lower landform positions of a single transect within each site was described according to the Canadian System of Soil Classification (Soil Classification Working Group 1998). Soil at each landform position was sampled by horizon and analyzed for organic matter (loss-on-ignition method, McKeague 1978), and texture (hydrometer method, Day 1965), and pH in a 1:2 mixture of soil: water (McKeague 1978).
Statistical Analysis. Analyses were completed using SAS version 9.1 (SAS Institute Inc. 2005) using the PROC MIXED procedure and a Fisher's protected LSD test for comparisons of three means, and a Tukey's adjustment for comparisons of more than six means. The PROC CORR procedure was used to relate measures of STP to topographic index. A significance level of 0.05 was used throughout this study.
Results and Discussion
Site Characteristics. The soils at the sites are typical of glaciated landscapes in Alberta, Canada, and the western Canadian Prairies. The soils developed under mixed open grassland, with some boreal aspen forest on calcareous clay loam till except for the northernmost GPC site, which was developed on a clay till. Surface textures were similar among landform classes, although there was some variation in organic matter content and pH among landform classes (table 2).
STP Variation among and within Sites. There was significant variation in mean STP concentration among the eight study sites (table 3). The ungrazed and unfertilized STV site had the lowest STP concentrations of the microwatersheds, with a mean STP of 5 mg [kg.sup.-1]. This value is similar to the 10 mg [kg.sup.-1] measured in other uncultivated soils in Alberta (Dormaar and Chang 1995). Mean STP concentrations in the non-manured but fertilized sites were significantly greater, with mean values from 24 to 39 mg [kg.sup.-1]. The narrow range of mean STP at the non-manured sites was within the range measured for other non-manured soils in Alberta (Penney et al. 2003). Mean STP at the manured sites was significantly greater than at the other sites, with values from 236 to 446 mg [kg.sup.-1], which are within the range reported for other manured soils in Alberta (Whalen and Chang 2001).
Greater variation in STP concentration was observed within sites than among sites with similar types of P inputs (i.e., non-manured vs. manured). In the non-manured sites, the mean STP concentration varied as much as 15 mg [kg.sup.-1] among sites, but the coefficient of variation (CV) within these sites ranged from 31% at the WAB site to 77% at the CFT site (table 3). The higher CV at the CFT site was attributed to a knoll that had been manured prior to the study, increasing the STP at that sample location to a much greater level than the rest of the field. When this value was removed from the data set, the resulting CV of 39% was within the range measured at the other non-manured sites (table 3). The CV at all sites, except for the CFT site, were within the 40 to 60% range reported for STP measured in core samples taken from conventionally-tilled non-manured fields in Alberta (Cameron et al. 1971). Standard deviations were greatest at the manured sites since manure application rates vary greatly and manure is a highly variable P source (Dou et al. 2001).
The statistical distribution of all the STP samples at each site showed that STP in the 0 to 15 cm (0 to 6 in) soil layer was positively skewed at all sites except for the GPC site (table 3). The skew was greatest at the CFT and REN sites and least at the GPC, WAB and LLB sites. The positive skew indicated that there were a few samples with STP concentrations that were much greater than the mean. Similar results were also found for the 0 to 2.5 (0 to 1 in) and 0 to 5 cm (0 to 2 in) layers (data not shown). A positive skew was also identified for STP results measured in Alberta (Penney et al. 1996) and in the United Kingdom (Page et al. 2005). If the points with high concentrations of STP are in lower landform positions or in areas with high topographic indices, they may have a large influence on P levels in runoff.
STP Variation with Depth of Soil Layer. Measures of STP in the 0 to 2.5 cm (0 to 1 in) layer may be a good indicator of the potential for P runoff from a site as this layer is the zone of interaction with runoff (Sharpley 1985). However, most samples of STP are taken for agronomic purposes from the 0 to 15 cm (0 to 6 in) depth, and sampling a shallow soil layer can be difficult since the roughness of the soil surface adds uncertainty. Although greater STP concentrations were measured in the 0 to 2.5 cm and 0 to 5 cm (0 to 2 in) soil layers than in the 0 to 15 cm layer at all sites, the increase was not significant at the manured LLB and PON sites (table 4). Although differences in STP by depth of soil layer were also not significant at the CFT site where a knoll had been manured, higher STP in the 0 to 2.5 cm and the 0 to 5 cm soil layers compared with the 0 to 15 cm layer were found at this site after the removal of the point sampled on the manured knoll (data not shown). The magnitude of STP increase in the 0 to 2.5 cm layer compared with the 0 to 15 cm layer ranged from 1.2 times at the conventionally tilled WAB site to two times at the reduced till (i.e., occasional fall cultivation) REN site. The increase of 1.6 times at the no-till THC site was less than the increase of up to three times reported for the 0 to 2 cm (0 to 0.8 in) layer compared with the 0 to 8 cm (0 to 3 in) layer by Guertal et al. (1991) in no-till conditions. These results confirm that P tends to be more concentrated at the soil surface due to its limited mobility in soils (Sharpley 1985), especially under reduced tillage management (Sharpley et al. 1993). Selles et al. (1999) also measured accumulations of STP in the surface 5 cm of soil after 12 years of no-till continuous wheat in western Canadian cropping systems. Andraski and Bundy (2003) and Page et al. (2005) reported similar findings in other cropping systems.
Degree of Phosphorus Saturation. Mean PSI values in the 0 to 2.5 cm (0 to 1 in) soil layer (table 5) were ordered differently among the study sites than were mean STP concentrations (table 3). The PSI at the non-manured GPC site (442 mg [kg.sup.-1]) was significantly greater than at all other sites. This was likely due to the greater clay content, which increases the number of exchange surfaces available to bind P, at the GPC site (table 2). Among the non-manured sites, there was a greater difference among the site means, as well as a greater range within each site (up to 323 mg [kg.sup.-1]) than was measured for STP. This finding suggests that there may be more potential to differentiate among sites and among points using PSI than using STP. The heavily manured PON site had a significantly lower PSI (49 mg [kg.sup.-1]) than all the other sites.
When STP and PSI were used to calculate DPS, the results (table 5) followed a pattern that was similar to the STP variation among the sites (table 3). There was significantly lower DPS at the STV site (5%) than at any other site, similar DPS at the non-manured sites (18 to 23%) except for the GPC site (10%), and significantly greater DPS at the two manured sites (60 to 91%, table 5). A critical STP change point, above which any added P may be more readily lost from soil via runoff, was identified at a DPS of 44% for Alberta soils (Casson et al. 2006). None of the sampling points within the non-manured sites had DPS levels that exceeded 44%, except for the manured knoll at the CFT site, which had a DPS value of 66%. The change point was exceeded at all points within the heavily manured PON site, where a maximum of 100% reflected the complete saturation of P sorption sites at some sampling points. If the areas of a field that are saturated with P are in critical source areas that contribute to runoff, there is an increased likelihood that P will be transported from the field by runoff.
Variation with Landform Class. Accumulations of P in low-lying areas in the western Canadian Prairies have been measured by other researchers (Manning et al. 2001; Penney et al. 2003), and it was anticipated that this might be found in the lower landform positions at the microwatershed sites. However, there were no differences in STP concentrations in the 0 to 15 cm (0 to 6 in) layer by landform class at five of the seven cultivated study sites (CFT, GPC, REN, WAB, and PON; figure 2). Although some field researchers have cautioned against using significance levels of 0.05 in variable field conditions due to the chance of failing to detect an effect when it exists (Pennock 2004; Peterman 1990), there were no differences in STP concentrations by landform class when a significance level of 0.15 was considered at these five sites. The lack of significance at the four non-manured sites may be attributable to decades of P fertilizer application at a consistent rate across all landform positions, and to the homogenizing effect of conventional tillage at the gently sloping GPC and WAB sites.
Of the two sites with significant differences in STP by landform class, the THC site had more STP in the lower than in the upper landform position, and the LLB site had significantly less STP in the lower than the upper landform position (figure 2). Since other landform classes within the THC site had STP concentrations as high as those measured in the lower landform positions (figure 3a), the choice of landform boundaries in the model of MacMillan et al. (2000) may not have been appropriate or factors other than landform class influenced the distribution of STP. For example, at the LLB site, significantly lower concentrations of STP in the lower landform positions were attributable to wet conditions that prevented the manure application in this part of the field.
It was anticipated that greater levels of STP might be found in the 0 to 2.5 cm (0 to 1 in) layers of the lower landform classes through the accumulation of P-rich material by erosion. However, the results generally did not support this hypothesis since no change from the pattern described for the 0 to 15 cm (0 to 6 in) soil layer was observed for the 0 to 2.5 cm or the 0 to 5 cm (0 to 2 in) layers, except at the THC site. Although differences in STP among landform classes were not significant for the shallower soil layers at the THC site, there was a trend toward a difference at p = 0.09 (data not shown).
The absence of a landform effect on STP distribution was also found by Needelman et al. (2001), who noted that the distribution of high STP concentrations in a 39.5 ha (98 ac) watershed was likely influenced by management practices and that there did not seem to be an impact of landform class, surface soil texture, or other natural factors. Cabot et al. (2004) and Page et al. (2005) also noted that management effects on the distribution of STP within a field can obscure possible accumulations of STP in lower landform positions due to erosion processes.
The PSI values were also similar within landform class in the subset of points measured at the study sites (figure 4a), except among the upper and lower landform positions at the PON site. These results reflect the similarity of soil properties within the study sites, as characterized in a transect representing the landform classes within each site, i.e., variation in clay content among landform classes was within 9% and organic matter levels were within 4% (table 2). Due to the similarities of PSI and STP with landform class at the sites, there were also few differences in DPS within any site except at the LLB site where STP differed among landform positions (figure 4b). The change point identified at 44% for Alberta soils (Casson et al. 2006) was not exceeded in any landform class at the non-manured sites, or in the lower landform class at the moderately manured LLB site, but it was exceeded in the upper and mid landform positions at the LLB site and in all three landform positions at the heavily-manured PON site.
Variation with Topographic Index. As the boundaries chosen to differentiate landform classes may be arbitrary, a continuous rather than a classed measure of topographic variation might better illuminate relationships with STP and DPS. In particular, greater STP accumulations may be found at sampling points with a high topographic index with large runoff contributing areas, than at points with a low topographic index where runoff contributing areas are small. However, no significant relationships were found between topographic index and STP, except for a weak, negative correlation at the GPC site (r = -0.58). The lack of correlation was most evident at the THC site, where a narrow range of STP levels corresponded to the widest range of topographic indices. The removal of an extreme point representing a manured knoll at the CFT site did not improve the relationship at that site. The topographic index was also poorly correlated with measures of STP in the 0 to 2.5 cm (0 to 1 in) and 0 to 5 cm (0 to 2 in) soil layers, and with DPS in the 0 to 2.5 cm layer. This confirms that the lack of a landform relationship with STP was correct and not due to inappropriate choices in landform categories.
Farm management information also gives some rationale for the lack of relationship between soil P and the topographic index. Multiple years of fertilizer additions applied consistently across the field as well as the homogenizing effect of conventional tillage at the GPC and WAB sites may have overwhelmed any STP redistribution by erosion. At the most variably managed LLB site, the cluster of points with a high topographic index but low STP concentration was sampled in a depression where manure was not applied in wet conditions. Points with high STP but low topographic indices within the lower landform positions at the LLB site (figure 3b) may have resulted from the overlap of the manure application equipment, from the high variability of manure application rates, and/or the use of manure as a P source. Although Moore et al. (1993) reported that the topographic index was significantly correlated with extractable soil P over a 5.4 ha (13 ac) toposequence in Colorado, the relationship was not strong (r = 0.53). Page et al. (2005) reported poor relationships between the distribution of STP and the topographic index within two similarly sized catchments (22 and 48 ha [54 to 119 ac]) in the United Kingdom, although the land management practices were more different within the United Kingdom catchments (i.e., mixed grassland, beef cattle, sheep and some maize) compared to the toposequence studied by Moore et al. (1993) as well as within the catchments of this study. Page et al. (2005) attributed the limited usefulness of the topographic index to the occurrence of "hot spots" due to applications of manure within the catchments. Similar observations of "patchy" distributions of STP within farm fields have been attributed to the location of old farmsteads (Schepers et al. 2000; Chang et al. 2003) and uneven applications of manure and municipal sludge (Cambardella and Karlen 1999; Page et al. 2005). However, Daniels et al. (2001) point out that a large number of samples, as is recommended for agronomic sampling methods, should offset the influence of a few extreme points.
Summary and Conclusions
Significant variation in the concentration of STP in the 0 to 15 cm (0 to 6 in) soil layer was measured among eight microwatersheds developed on glaciated landforms in Alberta. The variation was greater within sites than among sites with similar types of P inputs (i.e., non-manured vs. manured). Although it was expected that greater concentrations of STP might be found where P-rich sediments accumulated in lower landform positions, there was no difference in STP concentration by landform class at five of the seven cultivated study sites, and a topographic index was not correlated with STP. Rather, within-field variation in STP appeared to be more related to previous management history.
Accumulations of STP in 0 to 2.5 cm (0 to 1 in) were measured at all of the cultivated non-manured sites. Accumulations of STP represented an increase of up to two times the STP concentration in the 0 to 15 cm layer. As such, there appeared to be little advantage to soil sampling at depths shallower than the 0 to 15 cm (0 to 6 in) depth typically used for agronomic sampling. Since the microwatersheds in this study had soil properties that were similar among landform classes, there was little variation within sites in the PSI measured in the 0 to 2.5 cm layer. The similarities in PSI and STP also resulted in similar calculations of DPS in the 0 to 2.5 cm soil layer, and there was no reason to characterize DPS according to topographic characteristics at the study sites.
The lack of a topographic relationship for STP or DPS and small differences in STP by depth of soil layer suggests that where management and soil properties are similar, traditional agronomic sampling methods that use composite samples to represent average-yielding areas may be adequate to characterize STP levels as they relate to P concentrations in runoff from the field.
We wish to thank the staff of Alberta Agriculture and Food for assistance with soil sampling. Thanks also to Bob MacMillan, Wayne Pettapiece, Tony Brierley, Toby Entz, Rong-Cai Yang, Dennis Mikalson, Keith Toogood, and Phil Gibbs. We also thank the producers, landowners, and the municipalities for their cooperation, and the Alberta Agricultural Funding Consortium for partial funding.
Andraski, T.W., and L.G. Bundy. 2003. Relationships between phosphorus levels in soil and in runoff from corn production systems. Journal of Environmental Quality 32:310-316.
Brierly, J.A., T.C. Martin, and D.J. Spiess. 2001. Agricultural Region of Alberta Soil Inventory Database (AGRASID) Version 3.0. Soil Landcapes User's Manual. http://www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/sag3249?opendocument.
Cabot, P.E., S.K. Bower, and P.J. Nowak. 2004. Manure management in urbanizing settings. Journal of Soil and Water Conservation 59:235-243.
Cambardella, C.A., and D.L. Karlen. 1999. Spatial analysis of fertility parameters. Precision Agriculture 1:5-14.
Cameron, D.R., M. Nyborg, J.A. Toogood, and D.H. Laverty. 1971. Accuracy of field sampling for soil tests. Canadian Journal of Soil Science 51:165-175.
Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, and V.H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8:59-568.
Casson, J.P., D.R. Bennett, S.C. Nolan, B.M. Olson, and G.R. Ontkean. 2006. Degree of phosphorus saturation thresholds in manure-amended soils of Alberta. Journal of Environmental Quality 35:2212-2221.
Chang, J., D.E. Clay, C.G. Carlson, S.A. Clay, D.D. Malo, R. Berg, J. Kleinjan, and W. Wiebold. 2003. Different techniques to identify management zones impact nitrogen and phosphorus sampling variability. Agronomy Journal 95:1550-1559.
Cornish, P.S., R. Hallissey, and E. Hollinger. 2002. Is a rainfall simulator useful for estimating phosphorus runoff from pastures--A question of scale dependency. Australian Journal of Soil Research 42:953-959.
Crozier, C., G. Naderman, M.R. Tucker, and R.E. Sugg. 1999. Nutrient and pH stratification with conventional and no-till management. Communications in Soil Science and Plant Analysis 30:65-74.
Daniels, M.B., P. Delaune, P.A. Moore, Jr., A. Mauromoustakos, S.L. Chapman, and J.M. Langston. 2001. Soil phosphorus variability in pastures: Implications for sampling and environmental management strategies. Journal of Environmental Quality 30:2157-2165.
Day, P.R. 1965. Particle fractionation and particle size analysis. In Methods of Soil Analysis, Part 1, ed. C.A. Black, 545-567. Agronomy No. 9. Madison, WI: American Society of Agronomy.
Dormaar, J.F., and C. Chang. 1995. Effect of 20 annual applications of excess feedlot manure on labile soil phosphorus. Canadian Journal of Soil Science 75:507-512.
Dou, Z., D.T. Galligan, R.D. Allshouse, J.D. Toth, C.F. Ramberg, Jr., and J.D. Ferguson. 2001. Manure sampling for nutrient analysis: Variability and sampling efficacy. Journal of Environmental Quality 30:1432-1437.
Ecological Stratification Working Group. 1995. A National Ecological Framework for Canada. Agriculture and Agri-Food Canada, Ottawa/Hull.
Franzen, D.W., T. Nanna, and W.A. Norvell. 2006. A survey of soil attributes in North Dakota by landscape position. Agronomy Journal 98:1015-1022.
Gburek, W.J., and A.N. Sharpley. 1998. Hydrologic controls on phosphorus loss from upland agricultural watersheds. Journal of Environmental Quality 27:267-277.
Guertal, E.A., D.J. Eckert, S.J. Traina, and T.J. Logan. 1991. Differential phosphorus retention in soil profiles under no-till crop production. Soil Science Society of America Journal 55:410-413.
Haygarth, P.M., L.M. Condron, A.L. Heathwaite, B.L. Turner, and G.P. Harris. 2005. The phosphorus transfer continuum: Linking source to impact with an interdisciplinary and multi-scaled approach. Science of the Total Environment 344:5-14.
Heathwaite, A.L., and R.M. Dils. 2000. Characterising phosphorus loss in surface and subsurface hydrological pathways. Science of the Total Environment 251:523-538.
Indiati, R., and P. Sequi. 2004. Phosphorus intensity-quantity relationships in soils highly contrasting in phosphorus adsorption properties. Communications in Soil Science and Plant Analysis 35:131-143.
Kleinman, P.J.A., M.S. Srinivasan, C.J. Dell, J.P. Schmidt, A.N. Sharpley, and R.B. Bryant. 2006. Role of rainfall intensity and hydrology in nutrient transport via surface runoff. Journal of Environmental Quality 35:1248-1259.
Little, J.L., S.C. Nolan, and J.P. Casson. 2006. Relationships between soil and runoff phosphorus in small Alberta watersheds. Journal of Environmental Quality 36:1289-1300.
Lobb, D.A., R.G. Kachanoski, and M.H. Miller. 1995. Tillage translocation and tillage erosion on shoulder slope landscape positions measured using [.sup.137.C]s as a tracer. Canadian Journal of Soil Science 75:211-218.
MacMillan, R.A., W.W. Pettapiece, S.C. Nolan, and T.W. Goddard. 2000. A generic procedure for automatically segmenting landforms into landform elements using DEMs, heuristic rules and fuzzy logic. Fuzzy Sets and Systems 113:81-109.
Mallarino, A. 1996. Spatial variability patterns of phosphorus and potassium in no-tilled soils for two sampling scales. Soil Science Society of America Journal 60:1473-1481.
Manning, G., L.G. Fuller, R.G. Eilers, and I. Florinsky. 2001. Soil moisture and nutrient variation within an undulating Manitoba landscape. Canadian Journal of Soil Science 81:449-458.
McKeague, J.A. (ed.). 1978. Manual on soil sampling and methods of analysis. Ottawa, Ontario, Canada: Canadian Society of Soil Science.
Moore, I.D., P.E. Gessler, G.A. Nielson, and G.A. Peterson. 1993. Soil attribute prediction using terrain analysis. Soil Science Society of America Journal 57:443-452.
Morton, J.D., D.B. Baird, and M.J. Manning. 2000. A soil sampling protocol to minimise the spatial variability in soil test values in New Zealand hill country. New Zealand Journal of Agricultural Research 43:367-375.
Mulla, D.J. 1993. Mapping and managing spatial patterns in soil fertility and crop yield. In Proceedings of First Workshop, Soil Specific Crop Management, 15-26. Madison, WI: American Society of Agronomy, Crop Science Society of America, Soil Science Society of America.
Murphy, J., and J.P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27:31-36.
Needelman, B.A., W.J. Gburek, A.N. Sharpley, and G.W. Petersen. 2001. Environmental management of soil phosphorus: Modeling spatial variability in small fields. Soil Science Society of America Journal 65:1516-1522.
Nolan, S.C., B.M. Olson, and F.J. Hecker. 2006. A comparison of sampling methods for soil test-phosphorus. In Alberta Soil Phosphorus Limits Project. Volume 3: Soil Sampling, Manure Application, and Sorption Characteristics, 1-33. Lethbridge, Alberta, Canada.
Page, T., P.M. Haygarth, K.J. Beven, A. Joynes. T. Butler, C. Keeler, J. Freer, P.N. Owens, and G.A. Wood. 2005. Spatial variability of soil phosphorus in relation to the topographic index and critical source areas: Sampling for assessing risk to water quality. Journal of Environmental Quality 34:2263-2277.
Penney, D.P., S.C. Nolan, R.C. McKenzie, T.W. Goddard, and L. Kryzanowski. 1996. Yield and nutrient mapping for site specific fertilizer management. Communications in Soil Science and Plant Analysis 27:1265-1279.
Penney, D.P., K. Cannon, and D. Keyes. 2003. Preliminary analysis of five years of soil data from the AESA soil quality benchmark sites. In Proceedings of the 40th Alberta Soil Science Workshop, Edmonton, Alberta, Canada, February 18-20, 2003, 225-229.
Pennock, D.J., B.J. Zebarth, and E. De Jong. 1987. Landform classification and soil distribution in hummocky terrain, Saskatchewan, Canada. Geoderma 40:297-315.
Pennock, D.J. 2004. Designing field studies in soil science. Canadian Journal of Soil Science 84:1-10.
Peterman, R.M. 1990. Statistical power analysis can improve fisheries research and management. Canadian Journal of Fish and Aquatic Science 47:2-15.
Pote, D.H., T.C. Daniel, A.N. Sharplcy, P.A. Moore, Jr., D.R. Edwards, and D.J. Nichols. 1996. Relating extractable soil phosphorus to phosphorus losses in runoff. Soil Science Society of America Journal 60:855-859.
Qian, P., J. Liang, and R. Karamanos. 1991. Comparison of several extractants for available phosphorus and potassium. In Soils and Crops Workshop, Saskatoon, Saskatchewan, Canada, February 21-22, 1991, 91-100.
Quinn, P.F., K.J. Beven, and R. Lamb. 1995. The ln([alpha]/tan [beta]) index: How to calculate it and how to use it within the topmodel framework. Hydrological Processes 9:161-182.
Quinton, J.N., P. Strauss, N. Miller, E. Azazoglu, M. Yli-Halla, and R. Uusitalo. 2003. The potential for soil phosphorus tests to predict phosphorus losses in overland flow. Journal of Plant Nutrition and Soil Science 166:432-437.
SAS Institute Inc. 2005. The SAS System for WindowsTM, release 9.1. SAS Institute Inc., Cary, North Carolina.
Selles, F., B.G. McConkey, and C.A. Campbell. 1999. Distribution and forms of P under cultivator and zerotillage for continuous- and fallow-wheat cropping systems in the semi-arid Canadian prairies. Soil and Tillage Research 51:47-59.
Schepers, J.S., M.R. Schlemmer, and R.B. Ferguson. 2000. Site-specific considerations for managing phosphorus. Journal of Environmental Quality 29:125-130.
Schroeder, P.D., D.E. Radcliffe, M.L. Cabrera, and C.D. Belew. 2004. Relationship between soil test phosphorus and phosphorus in runoff: Effects of soil series variability. Journal of Environmental Quality 33:1452-1463.
Sharpley, A.N. 1985. Depth of surface soil-runoff interaction as affected by rainfall, soil slope, and management. Soil Science Society of America Journal 49:1010-1015.
Sharpley, A.N., T.C. Daniel, and D.R. Edwards. 1993. Phosphorus movement in the landscape. Journal of Production Agriculture 6:492-500.
Sharpley, A.N., P.J.A. Kleinman, R.W. McDowell, M. Gitau, and R.B. Bryant. 2002. Modeling phosphorus transport in agricultural watersheds: Processes and possibilities. Journal of Soil and Water Conservation 57:425-439.
Soil Classification Working Group. 1998. The Canadian System of Soil Classification. Agriculture and Agri-Food Canada Publication 1646 (revised). Ottawa, Ontario; NRC Research Press.
Tisdale, S.L., W.L. Nelson, and J.D. Beaton. 1985. Soil Fertility and Fertilizers. New York, NY: Macmillan Publishing Company.
United States Department of Agriculture, Natural Resources Conservation Service. 1999. Soil Taxonomy. Agriculture Handbook No. 436. Washington, DC: US Government Printing Office.
Vadas, P.A., P.J.A. Kleinman, A.N. Sharpley, and B.L. Turner. 2005. Relating soil phosphorus to dissolved phosphorus in runoff: A single extraction coefficient for water quality monitoring. Journal of Environmental Quality 34:572-580.
Wollenhaupt, N.C., R.P. Wolkowksi, and M.K. Clayton. 1994. Mapping soil test phosphorus and potassium for variable-rate fertilizer application. Journal of Production Agriculture 7:441-448.
Whalen, J.K., and C. Chang. 2001. Phosphorus accumulation in cultivated soils from long-term annual applications of cattle feedlot manure. Journal of Environmental Quality 30:229-237.
Wright, C.R., M. Amrani, M.A. Akbar, D.J. Heaney, and D.S. Vanderwel. 2006. Determining phosphorus release rates to runoff from selected Alberta soils using laboratory rainfall simulation. Journal of Environmental Quality 35:806-814.
Sheilah C. Nolan is a soils research agrologist, Joanne L. Little is the head of the Water Quality Unit, Janna P. Casson is a soil and water specialist, Frank J. Hecker is the head of the Irrigated Crop Development Unit, and Barry M. Olson is a research scientist in the Agricultural Stewardship Division, Alberta Agriculture and Food, Alberta, Canada.
Table 1 Characteristics and management information of fields closest to the outlet in the eight microwatershed sites. Area CDN US Landform Site (ha) classification* classification[dagger] type[double dagger] Ungrazed grassland site STV 2 O.BLC TH M1h Non-manured sites CFT 248 O.DBC TH U1h GPC 50 SZ.DGC AlfA U1h-M1l REN 26 O.BLC TH M1m THC 51 O.BLC TH M1m WAB 33 O.DGC AlfA U1h Manured sites LLB# 88 O.DBC TH U1l PON 30 E.BLC AlbA H1l Type of Site Slope Management[section] Crop P application Ungrazed grassland site STV 0% to 25% NA NA None Non-manured sites CFT 1% to 4% NT Barley Banded with seed GPC 1% to 4% CT Barley Banded with seed REN 1% to 8% RT Wheat Banded with seed THC 0% to 6% NT Wheat Banded with seed WAB 1% to 4% CT Wheat Banded with seed Manured sites LLB# 1% to 2% CT Corn silage Manure PON 0% to 5% CT Corn silage Manure Site Added phosphorus (kg [ha.sup.-1]) Ungrazed grassland site STV 0 Non-manured sites CFT 22 GPC 10 REN 28 THC 15 WAB 15 Manured sites LLB# Every 3 years PON 1 to 2 x per year * CDN soil classification symbols follow the Canadian System of Soil Classification (Soil Classification Working Group 1998): 0 = orthic; SZ = solonetzic; E = eluviated; DB = dark brown; BL = black; DG = dark gray; C = chernozem. [dagger] US soil classification symbols follow the Soil Taxonomy system (USDA Natural Resources Conservation Service 1999): TH = Typic Haplustoll, AlfA = Alfic Agricryoll, AlbA = Albic Agricryoll. [double dagger] Landform type symbols follow the Agricultural Region of Alberta Soil Inventory Database (AGRASID) Version 3.0 (Brierly et al. 1998): U = undulating; M = rolling; H = hummocky; l = low relief; m = moderate relief; h = high relief (relative for each landform). [section] CT = conventional tillage; RT = reduced tillage; NT = no tillage before seeding. # Irrigated using a center pivot. NA = not applicable. Table 2 Soil properties in the surface horizon in a transect of upper, mid, and lower landform positions. Upper Mid Lower Organic Organic Organic Site Clay matter pH Clay matter pH Clay matter pH Ungrazed grassland STV 17.6% 10.0% 6.3 13.7% 13.9% 6.5 15.6% 12.2% 6.3 Non-manured CFT 19.1% 3.6% 8.0 20.7% 5.3% 6.4 19.8% 5.4% 6.5 GPC 26.4% 7.5% 6.5 29.0% 7.5% 6.0 26.4% 7.6% 6.0 REN 23.1% 3.7% 5.9 15.3% 6.6% 5.7 14.0% 7.0% 6.1 THC 23.9% 8.8% 6.2 22.6% 10.0% 6.0 18.3% 10.9% 6.2 WAB 20.4% 4.5% 5.6 20.4% 4.3% 5.9 25.8% 5.3% 6.2 Manured LLB 28.2% 4.1% 8.0 26.4% 4.5% 7.7 20.0% 3.2% 7.9 PON 14.9% 8.0% 6.5 12.4% 9.6% 6.5 15.9% 9.4% 7.6 Table 3 Soil-test phosphorus at all sampling points in the 0 to 15 cm soil layer after fall management in 2003. Min. Max. Median Site n (mg [kg.sup.-1]) (mg [kg.sup.-1]) (mg [kg.sup.-1]) Ungrazed grassland STV 3 5 6 5 Non-manured CFT[dagger] 48 13 220 33 GPC 22 3 52 34 REN 28 14 62 23 THC 27 11 68 24 WAB 27 16 57 31 Manured LLB 45 51 630 241 PON 22 240 786 430 Mean* Skew sd Site (mg [kg.sup.-1]) (mg [kg.sup.-1]) (mg [kg.sup.-1]) CV Ungrazed grassland STV 5e 1.6 1 17% Non-manured CFT[dagger] 39cd 4.9 30 77% GPC 35c -0.6 12 36% REN 24d 3.2 9 36% THC 27cd 1.6 14 53% WAB 32cd 0.6 10 31% Manured LLB 236b 0.8 113 48% PON 446a 1.1 127 28% Notes: sd = standard deviation. CV = coefficient of variation. * Mean values within the column followed by the same letter are not significantly different (p [less than or equal to] 0.05). [dagger] A knoll was manured at the site. Table 4 Soil-test phosphorus mean, standard deviation, and coefficient of variation values at all points sampled in three soil layers. 0 to 2.5 cm Mean* sd Site n Prob. > F (mg [kg.sup.-1]) (mg [kg.sup.-1]) CV Ungrazed grassland STV 3 0.0403 10ab 4 38% Non-manured CFT 48 0.3619 47 33 71% GPC 28 0.0009 47a 14 31% REN 28 <0.0001 48a 13 27% THC 27 0.0004 44a 17 38% WAB 27 0.0243 39a 12 31% Manured LLB 45 0.9050 245 127 52% PON 22 0.1002 544 184 34% 0 to 5 cm 0 to 15 cm Mean* sd Mean* Site (mg [kg.sup.-1]) (mg [kg.sup.-1]) CV (mg [kg.sup.-1]) Ungrazed grassland STV 9a 2 22% 5b Non-manured CFT 46 36 77% 39 GPC 46a 14 31% 35b REN 44a 13 28% 24b THC 39a 17 43% 27b WAB 39a 11 29% 32b Manured LLB 245 122 50% 236 PON 509 161 32% 446 0 to 15 cm sd Site (mg [kg.sup.-1]) CV Ungrazed grassland STV 1 17% Non-manured CFT 30 77% GPC 12 36% REN 9 36% THC 14 53% WAB 10 31% Manured LLB 113 48% PON 127 28% Notes: sd = standard deviation. CV = coefficients of variation. * Mean values within a row followed by the same letter are not significantly different (p [less than or equal to] 0.05). Table 5 Mean phosphorus sorption index and degree of phosphorus saturation in the 0 to 2.5 cm soil layer in six transects per site. Phosphorus sorption index Min. Max. Mean* Site n (mg [kg.sup.-1]) (mg [kg.sup.-1]) (mg [kg.sup.-1]) Ungrazed grassland STV[dagger] 3 185 207 193b Non-manured CFT 18 114 216 146d GPC 17 301 624 442a REN 18 87 249 194b THC 18 110 202 153cd WAB 18 142 211 172bc Manured LLB 18 79 190 133d PON 18 0 138 49e Phosphorus sorption index Degree of phosphorus saturation sd Site (mg [kg.sup.-1]) CV Min. Max. Mean* sd CV Ungrazed grassland STV[dagger] 12 6% 3% 7% 5%e 2 31% Non-manured CFT 27 18% 10% 66% 23%c 13 56% GPC 80 18% 5% 14% 10%d 2 24% REN 43 22% 13% 39% 21%c 7 32% THC 30 20% 15% 41% 23%c 7 32% WAB 18 11% 12% 30% 18%c 5 26% Manured LLB 29 22% 31% 84% 60%b 15 26% PON 40 83% 80% 100% 91%a 7 8% Notes: CV = coefficient of variation. * Means within the same column followed by the same letter are not significantly different at p [less than or equal to] 0.05. [dagger] Only one transect in 2 ha area.
|Printer friendly Cite/link Email Feedback|
|Author:||Nolan, S.C.; Little, J.L.; Casson, J.P.; Hecker, F.J.; Olson, B.M.|
|Publication:||Journal of Soil and Water Conservation|
|Date:||Nov 1, 2007|
|Previous Article:||Erosion control and storm water quality from straw with PAM, mulch, and compost blankets of varying particle sizes.|
|Next Article:||An integrated economic-hydrologic modeling framework for the watershed evaluation of beneficial management practices.|