Critical Source Areas (CSAs) of phosphorus (P) are areas within a watershed that have a high propensity to export P to surface waters. CSAs contain two factors: source and transport factors. Source factors include soil P status and fertilizer and manure inputs, while transport factors include hydrologic and erosion processes that mobilize P. The aim of this study was to: 1) identify CSAs of P in an agricultural watershed and the stormflow dynamics controlling P export and 2) to delineate CSAs of P at the agricultural field scale using georeferenced soil test P (STP) and a digital elevation model (DEM) in a geographic information system (GIS). Soil test P (STP) along with dissolved reactive P (DRP), particulate P (PP), and total P (TP) in soil water, groundwater, and surface runoff were monitored in three small (< 8 ha) agricultural watersheds located in Decatur, Illinois, each situated within a separate experimental field. Further, volumetric water content (VWC) was continuously monitored on topographic positions, e.g. foot slopes, hill slopes, and shoulder slopes, to determine topographic position influence on soil moisture distribution. Repeated measures mixed models analysis showed that foot slopes (32.2%) had significantly higher VWC than hill slope (29.6%) and shoulder slopes (30.9%) during the growing season, while foot slopes (38.9%) and hill slopes (38.9%) had significantly higher VWC than shoulder slopes (34.9%) during the dormant season. Persistent shallow groundwater tables were implicated to control spatial and temporal VWC moisture distribution. Both foot slopes and hill slopes were implicated as transport areas. Repeated measures mixed models analysis also showed that foot slopes (73 kg ha&minus1) had significantly higher STP than hill slopes (28.9 kg ha&minus1) and shoulder slopes (33.8 kg ha&minus1) most likely due to the erosion and deposition of sediment from upper slopes to lower slopes. Foot slopes were consequently classified as source areas. A surface runoff event revealed near stream saturation and flushing of soil moisture from upper slopes to lower slopes, indicating that the watersheds are variable source area driven. The peak of PP on the rising limb of the hydrograph was attributed to near stream sediment mobility while the peak of DRP on the falling limb was attributed to flushing of upper slope soil moisture via subsurface flow. GIS delineation of CSAs at the agricultural field scale was conducted to pinpoint precise locations within a field to implement precision P management. The topographic position index (TPI) along with a modified version of the slope classification model &mdash both of which were created by Weiss (2001) and automated by Jenness (2006) &mdash were used to delineate foot slopes, hill slopes, shoulder slopes, and flat areas within a 91.2 ha agricultural field from a DEM. Transport factors were, again, identified as foot slopes and hill slopes. Further, georeferenced STP data collected in spring 2010, fall 2010, and fall 2011 were averaged and interpolated using ordinary kriging to generate a single surface that represented three year spatial soil P status within the agricultural field. Source factors were identified as areas in the field that were excessive in soil P for corn-soybean production. A CSA model was created that identified areas where both source factors and transport factors overlapped. CSAs of P occurred on 2.3 ha of the agricultural field and occurred near grass waterways and roadside drainage ditches. A one way analysis of variance (ANOVA) along with a Tukey mean separation procedure of soil P on the four topographic positions was used to characterize soil P spatial dependencies on landscape attributes associated with topographic position. Foot slopes (79.5 kg ha&minus1) and flat areas (92.9 kg ha&minus1) had significantly greater soil P than hill slopes (59.8 kg ha&minus1) and shoulder slopes (49.8 kg ha&minus1) due to depositional and sink attributes. Depositional attributes exhibit concave curvature, e.g. foot slopes. This curvature effectively reduces the velocity of surface runoff so that sediment bound P suspended in surface runoff can be deposited on the soil surface. Sink areas accrue P inputs but do not lose P to erosion via surface runoff. These areas exhibit linear, non-sloping planes, e.g. flat areas, that are not conducive to surface runoff. Although topographic position explains the spatial dependencies of source and transport factors, the CSA model was able to pinpoint where CSAs of P spatially occur within the agricultural field which can allow for precision P management.

Since agriculture began, field management has been at the forefront of expanding food production beyond previous limitations. Agricultural productivity is closely related to the physical, chemical, and biological properties of the soil. Landscape position and field management are among primary factors affecting these soil properties. Delineation of topographic positions of the field surface by shape (i.e., convex, concave, and linear) characterizes areas that may accumulate or lose soil and nutrients either during a discrete event or cumulatively over several growing seasons. Increased soil compaction, degradation of soil structure, and erosion have all been attributed to declining agricultural production. In addition to the physical disturbance from cultivation, erosion and deposition of soil components in different landscape positions explain a large part of the heterogeneity of soil properties across an agriculture field. In response to this, conservation tillage techniques, precision agriculture, and other novel management strategies have been developed to reduce negative impacts conventional row crop production such as nutrient pollution and compaction while optimizing farmer inputs. The objective of this project was to evaluate effects of topographic position and conservation tillage techniques on soil physical, chemical, and biological properties on the field scale as well as correlate certain soil attributes with suspended soil runoff collected during the sprinkle infiltration test. Soil fertility sampling was completed every fall from 2011 to 2014 and additional sampling of soil physical properties was taken in the spring between 2013 and 2014. Differences between fall conservation tillage treatments, no-till (NT), AerWay® aerator (AA), and Great Plains Turbo-Till® (GP), and topographic positons, concave, convex and linear were analyzed. Sediment runoff and earthworm biomass were also collected in the fall in 2014. Results indicated a significant increase of soil organic matter (12%-24%), water stable aggregates (78%-98%), phosphorus (43%-76%), and cation exchange capacity (28%-35%) within concave over the convex landscape positions. Soil strength was significantly lower in the field managed with the GP vertical tillage disk compared with the AA field to a depth of 27.5 cm and the NT field to depth of 17.5 cm. Crop residue coverage (percent covered) was more complete in the NT field (12%) and the GP field (3%) compared with the AA field. Suspended sediment runoff was negatively correlated with water-stable aggregates, Ca, and Mg, but positively correlated with earthworm biomass. Extractable nutrients and soil physical properties were also strongly affected by air temperature and precipitation throughout the study period. Characterizing soil properties within topographic positions has potential applications in precision agriculture management, such as reducing excessive fertilization, and identifying areas of increased pollution potential. Evaluation of the tandem effects of conservation tillage tools and topographic position within central Illinois is important in order for the optimization of production and conservation of resources. Physical disturbance from tillage and the transport of sediment from eroded areas to depositional topographic positions are key factors influencing the variability of soil properties, crop productivity, and potential sediment-borne nutrient pollution within individual agricultural fields.

This research work concentrate on developing digital soil maps to support field based plant phenotyping research. We have developed soil organic matter content (OM), cation exchange capacity (CEC), natural soil drainage class, and tile drainage line maps using topographic indices and aerial imagery. Various prediction models (universal kriging, cubist, random forest, C5.0, artificial neural network, and multinomial logistic regression) were used to estimate the soil properties of interest.