Elsevier

Biosystems Engineering

Volume 113, Issue 2, October 2012, Pages 103-111
Biosystems Engineering

Research Paper
Environmental and economic evaluation of N fertilizer rates in a maize crop in Italy: A spatial and temporal analysis using crop models

https://doi.org/10.1016/j.biosystemseng.2012.06.012Get rights and content

Crop simulation models have the potential to assess the risk associated with the selection of a specific N fertilizer rate, by integrating the effects of soil-crop interactions on crop growth under different pedo-climatic and management conditions. The objective of this study was to simulate the environmental and economic impact (nitrate leaching and N2O emissions) of a spatially variable N fertilizer application in an irrigated maize field in Italy. The validated SALUS model was run with 5 nitrogen rates scenarios, 50, 100, 150, 200, and 250 kg N ha−1, with the latter being the N fertilization adopted by the farmer. The long-term (25 years) simulations were performed on two previously identified spatially and temporally stable zones, a high yielding and low yielding zone. The simulation results showed that N fertilizer rate can be reduced without affecting yield and net return. The marginal net return was on average higher for the high yield zone, with values ranging from 1550 to 2650  ha−1 for the 200 N and 1485 to 2875  ha−1 for the 250 N. N leaching varied between 16.4 and 19.3 kg N ha−1 for the 200 N and the 250 N in the high yield zone. In the low yield zone, the 250 N had a significantly higher N leaching. N2O emissions varied between 0.28 kg N2O ha−1 for the 50 kg N ha−1 rate to a maximum of 1.41 kg N2O ha−1 for the 250 kg N ha−1 rate.

Highlights

► The best amount of N fertilizer to apply on a field varies spatially and temporally. ► A crop simulation model was used to select optimal N rate over space and time. ► Net marginal return increased using the N rate chosen by model. ► N leaching and N2O emission were reduced with the N rate selected by the model.

Introduction

Increasing maize yield has been one of the main challenges for agronomists and researchers worldwide for the last 50 years (e.g. Egli, 2008, Hafner, 2003). In areas where Nitrogen (N) fertilizer is affordable or subsidized, there is an increased probability that farmers will apply in large quantities, potentially imposing a high environmental impact, including nitrate leaching (Basso and Ritchie, 2005, Giola et al., 2012, Martin et al., 2006, Syswerda et al., 2012), ammonia volatilization, nitrous oxide (N2O) emissions and soil acidification (Chen et al., 2008, Grace et al., 2011, Spiertz, 2010). The environmental impact that agriculture exerts is gaining more attention from society. For example, the European Union (EU) Nitrates Directive (91/676/EEC) aims to preserve the quality of groundwater through promotion of good farming practises to increase N use efficiency through a reduction in direct the application of N.

Determining the optimum amount of N fertilizer to meet plant needs while simultaneously minimizing environmental impacts is not an easy task (Robertson & Vitousek, 2009). The optimum N fertilizer rate varies within the same field with each growing season as a result of the heterogeneity of soil properties (which in turn affects soil water content) and inter- and intra-annual weather patterns (Basso, Bertocco, Sartori, & Martin, 2007). In most cases, farmers apply N fertilizer without considering the within field variation of soil properties. The concept of ‘precision farming’ was introduced in the early 1990s, with yield monitors being the most important technological tool for a successful application of precision farming (Pierce & Sadler, 1997). Since then, much research has been conducted in the search for site-specific and optimized application rates for several input resources, such as fertilizers and pesticides.

Some research has examined and found reasonable consistency between variable rate N fertilizer applications and the factors affecting nitrogen variability. These factors that have shown to have high influence are elevation, apparent electrical conductivity (ECa), and soil texture (Fraisse et al., 2001, Godwin and Miller, 2003, Kyveryga et al., 2011, Ruffo et al., 2005, Welsh et al., 2003). Walter, Bausch, and Brodahl (2012) recommended using ECa maps as a method of obtaining reference strip normalizing values in fields with spatially variable sandy soils.

Delin, Linden, and Berglund (2005) reported that the potential for improvement of yield or nitrogen efficiency by site-specific nitrogen fertilization is only relevant if the causes of within-field variation are predictable before fertilization. Approaches to derive uniform management zones have been described by Mulla, 1991, Schepers et al., 2004, Chang et al., 2004, Miao et al., 2006, Basso et al., 2007, and Basso, Ritchie, Cammarano, and Sartori (2011). Specific studies by Lawes and Roberston (2011) found that information needed for precision fertilizer management may neither be feasible nor easy to interpret. However, the complete decision making process on the application of N at the farm level must consider the specifics of information necessary to aid in decision making, including the cost–benefit of acquiring this information. The decision making process on N-application therefore becomes an integral and sustained element of the farm management information system as demonstrated by numerous studies (Basso et al., 2003, Doole et al., 2009, Fountas et al., 2006, Janssen and van Ittersum, 2007, Lawes and Roberston, 2011, Sørensen et al., 2010). An improved understanding of the factors affecting the determination of the optimal N-application in terms of both external influences (e.g. cost of fertilizers, chemicals, fuels, etc.) as well as the on-farm and in-field influences will assist farmers in achieving higher yields at lower costs.

The complexity of decision making is illustrated by the fact that even if farmers have a spatial map of soil properties, the decision about the amount of N fertilizer to apply on the field is taken without any prior knowledge of future weather conditions. A feasible approach to cope with such uncertain future information is to quantify the uncertainty under different scenarios as part of a predictive decision support system (Basso et al., 2011, Fountas et al., 2006).

Crop simulation models can quantify the interaction between multiple stresses and crop growth under different environmental and management conditions (Basso, Ritchie, Pierce, Jones, & Braga, 2001; Basso et al., 2011, Batchelor et al., 2002, Schnebelen et al., 2004). Using long-term historical weather data, the models can be used to develop alternative management strategies for optimizing productivity and maximizing profit as well as capturing the diversity of environments that can be encountered at a given farm. Crop simulation models are rarely used in precision farming because of the costs of obtaining detailed site-specific field attributes, as inputs are prohibitive, except in few case (Basso et al., 2001, Basso et al., 2011; Basso, Fiorentino, Cammarano, Cafiero, & Dardanelli, 2012; Batchelor et al., 2002, Booltink et al., 2001, Cora et al., 1999, Link et al., 2008, Miao et al., 2006; Wong & Asseng, 2006). Models can help farmers in a strategic way assessing the probability that a certain outcome will occur for those measured pedo-climatic condition and management practises. Models have also being shown to be useful in a tactical management of N fertilizer rate associated with more easily observed variables (i.e. water availability based on rainfall amounts). Basso, Ritchie, et al. (2011) demonstrated that N fertilizer amount needs to be different over space and time depending on the amount of water stored in the soil profile. In the past, best management of N fertilizer and irrigation recommendations have mainly aimed at increasing crop yield giving the environmental impact (potential for NO3 leaching and N2O emissions) a lower priority. In order to obtain information on nitrate leaching data over a long period of time, long-term experiments need to be implemented. This kind of information is normally not available due to the prohibitive cost of the long-term experiment and nitrate leaching collection protocol. A much more practical way of obtaining these data is through the use of crop simulation models. Several models are able to predict nitrate leaching potential under different fertilization strategies. Asseng et al. (1998) used the APSIM model to predict the leaching potential under different initial soil water and inorganic soil N showing that the soil water and the soil inorganic N content at the beginning of each season had no effect on grain yield, implying that pre-seed soil NO3 was mainly lost from the soil by leaching.

The objective of this study was to simulate the environmental and economic impact (nitrate leaching and N2O emissions) of spatially variable N fertilizer applications in an irrigated maize field in Italy using a validated crop model. The research aims at demonstrating the importance of using crop models for selecting best N management strategies from the economic and environmental perspective over space and time and in places where previous knowledge or long-term experiments are not present to reduce the risks the farmers face when selection N fertilizer strategies.

Section snippets

Site description

The study was carried out on an 8-ha field with a near zero slope, located close to Rovigo (44° 4′ 12″ N, 11° 47′ 22″ E, 6 m.a.s.l.) NE Italy grown with continuous maize for 5 years (1998–2002). The soil type was clay according to the USDA particle-size distribution limits, defined as FAO Ombric and Thionic Histosols. The climate of the area was characterized by an average annual rainfall of 700 mm (in years 1997–2008), distributed mostly in autumn and spring. The annual average temperature, for

Results

The differences between simulated and measured maize yield for the six years of study are shown in Fig. 1. There is a high correlation between simulated and measured yield demonstrating the consistency of the model used (y = 1.02x  493; r2 = 0.92; RMSE = 398.7 kg ha−1). The long-term simulation of yield at different N rates showed significant differences in yields as a function of the quantity of N applied to the crop (Table 1). An application rate of 50 N showed lower yield with values ranging from 6.1

Discussions

This research presented a modelling approach for selecting the most sustainable N fertilizer rate on two spatially and temporally stable zones of a maize field. The alternative to this approach, which accounts for the interaction between soil, plant, climate and management for 25 years is a long-term study, where despite the enormous value in having such an experiment, the cost for running and maintain are prohibitive. By running a validated model with a long-term weather data recorded from a

Conclusions

The potential of using a crop model as a decision tool for improving farmer’s economic return by maximizing yield or reducing input while protecting the environment was demonstrated. The crop growth simulation model allowed for deriving valuable information needed to decide on the optimal N rate to apply on spatially and temporally stable zones of a field. The long-term simulation of maize yield and N leaching showed the variability of yield response at different N rates. Derived differences

Acknowledgements

This research was partially funded by Veneto Agricoltura and the MIUR-PRIN 2008 project (Prof. Bruno Basso coordinator).

References (54)

  • S. Fountas et al.

    A model of decision-making and information flows for information-intensive agriculture

    Agricultural Systems

    (2006)
  • P. Giola et al.

    Impact of manure and slurry applications on soil nitrate in a maize-triticale rotation: field study and long term simulation analysis

    European Journal of Agronomy

    (2012)
  • R.J. Godwin et al.

    A review of the technologies for mapping within-field variability

    Biosystems Engineering

    (2003)
  • P.R. Grace et al.

    The contribution of maize cropping in the Midwest USA to global warming: a regional estimate

    Agricultural Systems

    (2011)
  • S. Hafner

    Trends in maize, rice, and wheat yields for 188 nations over the past 40 years: a prevalence of linear growth

    Agriculture, Ecosystems and Environment

    (2003)
  • S. Janssen et al.

    Assessing farm innovations and responses to policies: a review of bio-economic farm models

    Agricultural Systems

    (2007)
  • R.A. Lawes et al.

    Whole farm implications on the application of variable rate technology to every cropped field

    Field Crop Research

    (2011)
  • U. Sehy et al.

    Nitrous oxide fluxes from maize fields: relationship to yield, site-specific fertilization, and soil conditions

    Agriculture, Ecosystems and Environment

    (2003)
  • C.G. Sørensen et al.

    A user-centric approach for information modelling in arable farming

    Computers and Electronics in Agriculture

    (2010)
  • S.P. Syswerda et al.

    Long-term nitrate loss along an agricultural intensity gradient in the Upper Midwest USA

    Agriculture, Ecosystems and Environment

    (2012)
  • H.F.M. Ten Berge et al.

    Limits to nitrogen use on grassland

    Environmental Pollution

    (2002)
  • J.P.G. Welsh et al.

    Developing strategies for spatially variable nitrogen application in cereals, part II: wheat

    Biosystems Engineering

    (2003)
  • S. Asseng et al.

    Use of the APSIM wheat model to predict yield, drainage, and NO3− leaching for a deep sand

    Australian Journal of Agricultural Research

    (1998)
  • B. Basso et al.

    Impact of compost, manure and inorganic fertilizer on nitrate leaching and yield for a 6-year maize-alfalfa rotation in Michigan

    Agriculture, Ecosystem and Environment

    (2005)
  • B. Basso et al.

    Simulating tillage impacts on soil biophysical properties using the SALUS model

    Italian Journal of Agronomy

    (2006)
  • B. Basso et al.

    Evaluation of variable depth tillage: economic aspects and simulation of long term effects on soil organic matter and soil physical properties

  • B.S. Blackmore

    The interpretation of trends from multiple yield maps

    Computers and Electronics in Agriculture

    (2000)
  • Cited by (36)

    • Combining remote sensing-derived management zones and an auto-calibrated crop simulation model to determine optimal nitrogen fertilizer rates

      2023, Agricultural Systems
      Citation Excerpt :

      It is therefore crucial to maintain NFUE within this resource use efficiency range when determining optimal N rates to ensure long-term soil health and avoid harmful environmental consequences. To date, the combined application of MZs and CSMs to maximize profit and minimize environmental impacts has been successfully conducted in maize, barley and wheat (Basso et al., 2011; Basso et al., 2012; Cammarano et al., 2020a; McNunn et al., 2019). However, this combination has not been applied in the cotton industry.

    • Assessing the impacts of pre-growing-season weather conditions on soil nitrogen dynamics and corn productivity in the U.S. Midwest

      2022, Field Crops Research
      Citation Excerpt :

      The lack of understanding and quantification of weather impacts on soil N status during pre-growing-seasons restricts the development of more informed N fertilizer management strategies for current and changing climate conditions. The inability to control weather and challenges of measuring spatially and temporally variable N fluxes and underlying processes at field scales are major obstacles to evaluating the holistic impact of weather conditions on soil N cycling and crop yield (Basso et al., 2012; Batchelor et al., 2002; Fountas et al., 2006; Puntel et al., 2016). Process-based agroecosystem models such as APSIM (Holzworth et al., 2014), DNDC (Giltrap et al., 2010; Li et al., 1992), and EPIC (Williams, 1995), which simulate the carbon-water-nutrient balance in the soil-plant systems can be a viable approach to overcome the measurement challenge (Bassu et al., 2014; Chen et al., 2010; Dietzel et al., 2016; Peng et al., 2020; Tubiello and Ewert, 2002).

    View all citing articles on Scopus
    View full text