**Spatial Patterns of Water and Nitrogen Response Within Corn Production Fields**

Jerry L. Hatfield

*National Laboratory for Agriculture and the Environment USA* 

## **1. Introduction**

Agricultures role on environmental quality has been debated for many decades and although there has been advances in our understanding of the linkage between agricultural management and environmental quality, there is still much we don't understand about the combination of field-scale and watershed scale management changes (Hatfield et al., 2009). There is increasing interest in developing solutions to environmental quality problems originating from agricultural management; however, the challenge remains on how we integrate the pieces of a very complex puzzle together to achieve solutions which transcend space and time scale. Within the Midwestern United States, the reoccurrence of the hypoxic zone within the Gulf of Mexico and the large increase in the size of the hypoxic zone after the 1993 floods in the Midwest focused attention on the role of agriculture in nonpoint source pollution. Burkart and James (1999) evaluated the nitrogen (N) balance for the Mississippi River Basin and concluded that mineralization of soil organic matter and application of commercial fertilizer was two primary contributors to N load. Jaynes et al. (1999) after examination of a small watershed (5400 ha) in central Iowa found that nitrate-N losses averaged 20 kg ha-1 for this watershed and reached a level in excess of 40 kg ha-1 during 1993. Hatfield et al. (1998) found that drainage from Walnut Creek was the primary transport pathway for nitrate and the annual loads were related to precipitation differences among years. Hatfield et al. (1999) found that for Walnut Creek watershed that drainage through the subsurface drain lines accounted for approximately half of the annual precipitation with evapotranspiration accounting for the other half. This movement of water through the soil profile and the solubility of nitrate in water produce large amounts of nitrate-N loss (Jaynes et al, 1999). These results suggested a strong link between precipitation, crop water use patterns, and nitrate losses. Occurrence of the hypoxic zone has prompted an increased level of debate about the need to reduce N inputs into agricultural systems. Opponents of this conclusion argue that over the past 20 years the input of N fertilizer has not increased, soil organic matter levels haven't changed, and crop production levels have increased suggesting that the efficiency of the agronomic production system has increased and any change in nitrate loss would be difficult to achieve. Hatfield et al. (2009) performed an analysis of the temporal changes in the nitrate-N concentrations in the Raccoon River watershed in central Iowa and observed the changes in nitrate-N concentrations since the 1970's were dominated more by changes in land use practices

Spatial Patterns of Water and Nitrogen Response Within Corn Production Fields 75

related to soil types and N management. These results have prompted a series of studies developed to further elucidate these interactions. Studies conducted on water use and N application has been done within a single soil type or management zone. There is a lack of information on the interactions among soil, crop water use, and N management that would help develop an understanding of better N management in cropping systems typical of the Midwest United States. The increasing concern about the role of agricultural practices on water quality have prompted us to ask a series of questions about the interactions between crop water use, N use, yield, and soils. The objectives of our studies have been to quantify the interactions among N application, crop water use, yield, and soils for central Iowa production fields on the Clarion-Nicollet-Webster soil association and to determine the impact of these interactions on N management and potential offsite impacts from drained

Spatial analysis of field scale changes have recently been evaluated by Inman et al. (2008). They determined the relationship between early season NDVI, soil color-based management zones, and relative corn yields. Values of NDVI were collected at the eight-leaf growth stage and regression models explained between 25 to 82% of the variability in relative yield where relative yield was defined as the ratio of the observed yield to the maximum yield for the field as defined by Brouder et al. (2000). Inman et al. (2008) found coupling NDVI with the color-based soil zones didn't increase the ability to explain yield within the field. A complimentary study conducted by Massey et al. (2008) on ten years of site-specific data for corn, soybean (*Glycine max* (L.) Merr.), and grain sorghum (*Sorghum bicolor* L.) for a 36.5 ha field in central Missouri with claypan soils to quantify temporal changes in crop yield response. They developed profitability maps for the field and found that large areas of the field had negative profit due to areas of the field in which there was significant topsoil erosion. Brock et al. (2005) showed that high yielding management zones in a corn-soybean rotation were associated with poorly drained level soils while low yielding zones were associated with eroded or more sloping soils. Sadler et al. (2000a and 2000b) conducted a field scale study on drought stressed corn and found that although there was a relationship between soil map units and grain yield these relationships did not explain the reason for the yield variation. They found that variation among sites within soils was significant and suggested that improved understanding of yield variation would require more attention to within season observations of crop water stress. These studies have shown that there is a large amount of variation present within fields and that our understanding of the reasons for these patterns of variation would improve the knowledge base for precision agriculture

There have been several studies that have related remote sensing indices to crop yield and detection of N status in plant leaves; however, there has been little research on the spatial patterns within the field that could lead to improved understanding of the changes that occur within the growing season. There is a lack of understanding of the interactions between the spatial patterns of crop water use and N response. This study was conducted to evaluate the spatial and temporal patterns of crop water use and couple these observations with observations collected from N strips within fields as part of a N evaluation study. The objective of this study was to evaluate the spatial patterns within different fields observed by remote sensing and yield maps collected at harvest with yield monitors to determine the

information content contained in spatial analyses of agricultural fields.

agricultural lands.

applications.

than on nutrient management. This change in land use affected the soil water balance and the patterns of water movement within fields and throughout the season rather than application rates as the primary transport mechanism.

Crop yield response to N has been the focus of agronomic studies and has been primarily conducted on small plots under relatively controlled conditions. However, the advent of application systems to differentially apply N across fields has opened up new possibilities for quantifying N response at a scale which producers would have confidence of the information being applicable to their operations. Understanding field scale responses to N management practices is critical to the development of precision agriculture tools and methods for determining the patterns to change application rates within a field. One method that has been applied to the precision application of N has been to use leaf color or reflectance as a measure of N status. Detection of N stress in crop plants using leaf color has created opportunities for field-scale evaluation of N response. Some of the methods that have been employed are the Soil-Plant Analyses Development (SPAD) chlorophyll meter, color photography, or canopy reflectance factors to assess N variation across corn fields (Schepers et al. 1992, 1996; Blackmer et al. 1993, 1994, 1996a, 1996b; Blackmer and Schepers, 1996). These methods have been based on comparisons of an adequately-fertilized strip in the same field with a strip with an altered N rate. This approach eliminates the requirements for prior knowledge of the relationship between nutrient concentration and crop reflectance.

There have continued to be advances in the use of remote sensing methods to estimate N status in crops. Lee et al. (2008) used a single waveband at 0.735 μm to quantify N status in rice (*Oryza sativa* L.). In their method they used the first derivative of the reflectance at 0.735 μm as difference of the reflectance at 0.74 – 0.73 μm and dividing by ten. They found this to be as accurate as other indices including the normalized difference vegetative index (NDVI) expressed as (RNIR – RRED)/(RNIR – RRED) where RNIR reflectance in the nearinfrared waveband (0.78 – 0.79 μm) and RRED the reflectance in the 0.61 – 0.68 μm waveband. They also compared a simple ratio vegetative index expressed as RNIR/RRED. They used these wavelengths to create maps of canopy N status across fields at the panicle formation stage. They showed the variation in canopy N status but didn't evaluate the spatial patterns within fields. There continues to be advancements in the use of remote sensing to detect N status in crops, Chen et al. (2010) developed a double-peak canopy N index to predict plant N concentration in both corn (*Zea mays* L.) and wheat (*Triticum aestivum* L.) and found this worked well for N status; however, did not extend this approach to assess field variation. This is an extension of an earlier method proposed by Haboudane et al. (2002) in which they proposed leaf chlorophyll indices could be used to predict leaf chlorophyll content for application to precision agriculture. One of the lingering questions that remain is the degree of variation present in N response across a corn production field.

Spatial variation of crop yield across fields has prompted a series of questions about the role of nutrient management. Jaynes and Colvin (1997) showed that yield variation within central Iowa fields was related to precipitation differences among years. However, in their study there was not a measure of crop water use within the field with the implied assumption that precipitation was uniform across the field. Hatfield and Prueger (2001) and Hatfield et al. (2007) found a large amount of spatial variation in water use across fields

than on nutrient management. This change in land use affected the soil water balance and the patterns of water movement within fields and throughout the season rather than

Crop yield response to N has been the focus of agronomic studies and has been primarily conducted on small plots under relatively controlled conditions. However, the advent of application systems to differentially apply N across fields has opened up new possibilities for quantifying N response at a scale which producers would have confidence of the information being applicable to their operations. Understanding field scale responses to N management practices is critical to the development of precision agriculture tools and methods for determining the patterns to change application rates within a field. One method that has been applied to the precision application of N has been to use leaf color or reflectance as a measure of N status. Detection of N stress in crop plants using leaf color has created opportunities for field-scale evaluation of N response. Some of the methods that have been employed are the Soil-Plant Analyses Development (SPAD) chlorophyll meter, color photography, or canopy reflectance factors to assess N variation across corn fields (Schepers et al. 1992, 1996; Blackmer et al. 1993, 1994, 1996a, 1996b; Blackmer and Schepers, 1996). These methods have been based on comparisons of an adequately-fertilized strip in the same field with a strip with an altered N rate. This approach eliminates the requirements for prior knowledge of the relationship between

There have continued to be advances in the use of remote sensing methods to estimate N status in crops. Lee et al. (2008) used a single waveband at 0.735 μm to quantify N status in rice (*Oryza sativa* L.). In their method they used the first derivative of the reflectance at 0.735 μm as difference of the reflectance at 0.74 – 0.73 μm and dividing by ten. They found this to be as accurate as other indices including the normalized difference vegetative index (NDVI) expressed as (RNIR – RRED)/(RNIR – RRED) where RNIR reflectance in the nearinfrared waveband (0.78 – 0.79 μm) and RRED the reflectance in the 0.61 – 0.68 μm waveband. They also compared a simple ratio vegetative index expressed as RNIR/RRED. They used these wavelengths to create maps of canopy N status across fields at the panicle formation stage. They showed the variation in canopy N status but didn't evaluate the spatial patterns within fields. There continues to be advancements in the use of remote sensing to detect N status in crops, Chen et al. (2010) developed a double-peak canopy N index to predict plant N concentration in both corn (*Zea mays* L.) and wheat (*Triticum aestivum* L.) and found this worked well for N status; however, did not extend this approach to assess field variation. This is an extension of an earlier method proposed by Haboudane et al. (2002) in which they proposed leaf chlorophyll indices could be used to predict leaf chlorophyll content for application to precision agriculture. One of the lingering questions that remain is the degree of variation present in N response across a

Spatial variation of crop yield across fields has prompted a series of questions about the role of nutrient management. Jaynes and Colvin (1997) showed that yield variation within central Iowa fields was related to precipitation differences among years. However, in their study there was not a measure of crop water use within the field with the implied assumption that precipitation was uniform across the field. Hatfield and Prueger (2001) and Hatfield et al. (2007) found a large amount of spatial variation in water use across fields

application rates as the primary transport mechanism.

nutrient concentration and crop reflectance.

corn production field.

related to soil types and N management. These results have prompted a series of studies developed to further elucidate these interactions. Studies conducted on water use and N application has been done within a single soil type or management zone. There is a lack of information on the interactions among soil, crop water use, and N management that would help develop an understanding of better N management in cropping systems typical of the Midwest United States. The increasing concern about the role of agricultural practices on water quality have prompted us to ask a series of questions about the interactions between crop water use, N use, yield, and soils. The objectives of our studies have been to quantify the interactions among N application, crop water use, yield, and soils for central Iowa production fields on the Clarion-Nicollet-Webster soil association and to determine the impact of these interactions on N management and potential offsite impacts from drained agricultural lands.

Spatial analysis of field scale changes have recently been evaluated by Inman et al. (2008). They determined the relationship between early season NDVI, soil color-based management zones, and relative corn yields. Values of NDVI were collected at the eight-leaf growth stage and regression models explained between 25 to 82% of the variability in relative yield where relative yield was defined as the ratio of the observed yield to the maximum yield for the field as defined by Brouder et al. (2000). Inman et al. (2008) found coupling NDVI with the color-based soil zones didn't increase the ability to explain yield within the field. A complimentary study conducted by Massey et al. (2008) on ten years of site-specific data for corn, soybean (*Glycine max* (L.) Merr.), and grain sorghum (*Sorghum bicolor* L.) for a 36.5 ha field in central Missouri with claypan soils to quantify temporal changes in crop yield response. They developed profitability maps for the field and found that large areas of the field had negative profit due to areas of the field in which there was significant topsoil erosion. Brock et al. (2005) showed that high yielding management zones in a corn-soybean rotation were associated with poorly drained level soils while low yielding zones were associated with eroded or more sloping soils. Sadler et al. (2000a and 2000b) conducted a field scale study on drought stressed corn and found that although there was a relationship between soil map units and grain yield these relationships did not explain the reason for the yield variation. They found that variation among sites within soils was significant and suggested that improved understanding of yield variation would require more attention to within season observations of crop water stress. These studies have shown that there is a large amount of variation present within fields and that our understanding of the reasons for these patterns of variation would improve the knowledge base for precision agriculture applications.

There have been several studies that have related remote sensing indices to crop yield and detection of N status in plant leaves; however, there has been little research on the spatial patterns within the field that could lead to improved understanding of the changes that occur within the growing season. There is a lack of understanding of the interactions between the spatial patterns of crop water use and N response. This study was conducted to evaluate the spatial and temporal patterns of crop water use and couple these observations with observations collected from N strips within fields as part of a N evaluation study. The objective of this study was to evaluate the spatial patterns within different fields observed by remote sensing and yield maps collected at harvest with yield monitors to determine the information content contained in spatial analyses of agricultural fields.

Spatial Patterns of Water and Nitrogen Response Within Corn Production Fields 77

any missing data using an approach similar to the method described by Hernandez-Ramirez (2010). Generally the length of any missing data was less than 3 hours. The amount of

Crop transpiration rates were estimated from an energy balance model that determined the soil water evaporation rate based on the leaf area index of the crop and previous precipitation amounts following the approach described by Ritchie and Burnett (1971). Soil water evaporation rates were estimated from a surface energy balance model based on crop residue cover amounts and the energy balance. Precipitation for these studies was available from a tipping bucket raingauge located at a meteorological station within 1 km of the field

Crop growth and development were measured in a variety of ways. In 1997 measurements were made of yield at harvest. Beginning in 1998 and 1999, a more intensive plant regime of weekly plant measurements consisting of leaf area, phenological stage, number of leaves, dry weight, and plant height were made on 10 plants from each plot and each plot was replicated three times. In 2000, the frequency was decreased to four destructive plant samplings to represent the 6-leaf stage, 12-leaf stage, tasseling, and mid-grain fill. Leaf chlorophyll measurements were made on 30 plants in each plot with a leaf chlorophyll meter at two times per week commencing with the 6 leaf stage and continuing through late grain fill. The upper leaf was measured at the mid-leaf position until the tassel appeared then the leaf immediately above the ear position was measured. Leaf carbon and N contents were determined on dry, ground samples from same stages as the 2000 plant samples were collected. For 1999 and 2000 experiments, stalk sugar content was measured with a refractive method using sap collected from freshly cut stalks. This was done on ten plants from each plot. Grain quality parameters of protein, oil, and starch were measured on subsamples of grain collected from the hand-harvest samples. Field yields were measured with yield-monitors mounted on the producers combine. These data were registered with a

Data analyses for these studies are based on crop yield, total seasonal water use, and N application. Water use and N use efficiency was determined by the ratio of crop yield to either seasonal crop transpiration or N application rates. Intensive plant sampling data are not described in this report but were used to understand the dynamics of plant response to changes in within season N management decisions. Likewise, the leaf chlorophyll and stalk sugar content data were used to guide decisions in the 1999 and 2000 experiments. These

Experiments were conducted in 12 different fields located in central Iowa from 2000 to 2002 (Table 1). These fields varied in size from 15 to 130 ha. Each of the field experiments was similar with strips arranged in the field with different N rates and in three of the study fields, variable planting rates were also used as a treatment variable. The strips were a minimum of 50 m wide covering at least 60 rows of corn. The arrangement of the N treatments was randomized across the field and strips were treated as replicates. Nitrogen was applied as UAN (Urea and Ammonium Nitrate) as a preplant treatment in all fields and incorporated into the soil. Rates of N application were determined by using

data sets represent a complete analysis of crop-soil-water-N interactions.

missing data for these experiments was less than 3% of the total data record.

sites.

GPS unit to obtain field locations.

**2.2 Spatial nitrogen studies** 
