Understanding the foundations of fertilizer recommendations

Why fertilizer recommendations matter 

At least 17 elements are currently recognized as essential for plant growth, development, and reproduction (Figure 1). Most of these nutrients are supplied by the soil, whereas much of the oxygen in plant tissues are derived from atmospheric carbon dioxide, and hydrogen is obtained primarily from water. While the total nutrient content of most soils greatly exceeds crop requirements, only a small portion exists in plant-available forms. As a result, nutrient deficiencies are a common yield-limiting factor in agricultural production. 

Fertilizers can prevent these deficiencies, but they increase production costs and, when mismanaged, can contribute to environmental degradation. Fertilizer recommendations are designed to maximize crop productivity, improve farm profitability, and minimize environmental impacts by providing guidance on the nutrient source, application rate, timing, and placement needed to meet crop demand during critical growth stages. This article explains how these recommendations are developed.

What is a fertilizer recommendation?

Fertilizer recommendations are research-based guidelines that outline nutrient management practices for supplying nutrients to a crop in order to achieve yield, quality, and economic objectives while minimizing nutrient losses to the environment. The principles of 4R nutrient stewardship, which emphasize the right source, right rate, right time, and right placement of nutrients, are integral components of modern fertilizer recommendations (Mikkelsen, 2011; Johnston & Bruulsema, 2014).

The source refers to the fertilizer products used to supply nutrients, including soluble, granular, fluid, and controlled-release fertilizers. Nitrogen, for example, may be supplied as dry fertilizers such as urea, ammonium sulfate, calcium ammonium nitrate, or diammonium phosphate, or as liquid fertilizers including aqua ammonia, urea ammonium nitrate, and urea solutions. The rate specifies how much fertilizer is needed to supplement the nutrients supplied from the soil. The timing of fertilizer application affects nutrient use efficiency, yield, quality, and environmental impact. Timing options such as pre-plant, starter, and split application can help producers meet different goals. Choosing between placement methods such as ground (broadcast and band), foliar, fertigation, and injection depends on the crop, soil, location, and nutrient goals, as well as available equipment/operation feasibility. 

Site-specific fertilizer prescriptions can be developed by combining these principles with soil test results and field-specific management information. 

Why nutrient behavior matters

Nutrient mobility dictates how elements move through the soil profile and determines the risk of environmental loss or root access. Because soil colloids tend to hold a negative charge, anions such as nitrate and sulfate do not bind to soil particles and can move relatively freely in soil water. Consequently, mobile nutrients are generally more susceptible to leaching and runoff than relatively immobile nutrients. Conversely, many nutrients such as phosphorus, potassium, zinc, and iron interact strongly with soil minerals and organic matter, limiting their movement through the soil profile and making root interception and diffusion important uptake mechanisms.

Soil pH is one of the most important factors affecting nutrient availability. In highly acidic or alkaline soils, many nutrients become less soluble and therefore less available for plant uptake. Under favorable soil conditions, mobile nutrients primarily reach roots through mass flow, whereas relatively immobile nutrients are supplied mainly by diffusion and root interception. Because nutrients differ greatly in their mobility and availability, fertilizer recommendations must be tailored to individual nutrients, crops, soils, and production systems. 

Building the foundation: Soil testing and field research

Soil testing

The accuracy of fertilizer recommendations depends on the ability of soil tests to predict crop response to fertilizer application. This prediction depends on two linked steps: obtaining a representative soil sample and interpreting the soil test value using locally calibrated recommendations.

Soil testing is the starting point for most fertilizer recommendations. Its purpose is to provide an index of the soil's nutrient-supplying capacity before fertilizer decisions are made. However, laboratory results can only represent the sample submitted for analysis. Therefore, sampling depth, timing, and consistency are critical and should follow state or regional guidelines. 

For relatively immobile nutrients and soil properties such as phosphorus, potassium, pH, and organic matter, samples are typically collected from the surface soil. For more mobile nutrients such as nitrogen and sulfur, profile soil samples are recommended because these nutrients move throughout the root zone. Keeping sampling procedures consistent across fields and over time is essential to avoid misleading estimates of soil nutrient status. The sampling strategy should also match the scale of the nutrient management decision. Composite samples are commonly collected from relatively uniform areas of a field, whereas grid or zone sampling may be used to support site-specific nutrient management.

After sampling, the next step is assessing soil nutrient status through laboratory analysis using standardized soil-testing procedures. In the United States, soil testing methods are commonly based on regional guidelines developed by organizations such as the Southern Extension and Research Activity Information Exchange Group for Soil, Plant, and Water Analysis (SERA-IEG-6), the North Central Region Soil Testing and Plant Analysis Committee (NCERA-13), and the Western Coordinating Committee on Nutrient Management (WERA-103). Standardization efforts are further supported by organizations such as the North American Proficiency Testing Program (NAPT) and the Agricultural Laboratory Proficiency Program (ALP) of the Agricultural Laboratory Testing Association (ALTA), which help ensure analytical consistency among laboratories. 

A soil test does not measure the total amount of a nutrient present in the soil. Instead, it extracts a fraction that is empirically related to crop nutrient availability and crop response. Therefore, soil test values should be interpreted as relative indices of nutrient supply rather than absolute measures of plant-available nutrients. In general, lower soil test values indicate a greater probability of crop response to fertilization, whereas higher values indicate a lower probability of response.

Because soil test interpretations are method and region specific, fertilizer recommendations may differ among regions, laboratories, and land grant universities, especially when different extractants, crops, soils, or calibration datasets are used. Therefore, soil test results must be interpreted using recommendations developed for the same extraction method, crop, soil conditions, and region whenever possible. Soil testing should be understood as an integrated process that includes representative sampling, laboratory analysis, and locally calibrated interpretation. Inconsistency among any of these components can lead to misleading fertilizer recommendations.     

Linking measurements to crop response

Plant measurements, including tissue nutrient concentration, nutrient uptake, biomass accumulation, chlorophyll readings, Normalized Difference Vegetation Index (NDVI), canopy development, visual deficiency symptoms, and cell sap tests, provide additional information on crop nutrient status. These measurements help distinguish nutrient deficiencies from other constraints to crop growth, such as poor root development, water stress, or unfavorable weather conditions. Yield response data provide the final evidence needed to evaluate fertilizer effectiveness. By comparing fertilized treatments with unfertilized control or with lower fertilizer rates, researchers can determine whether fertilizer application increased yield and whether that response was large enough to be agronomically or economically meaningful. Together, soil tests, plant measurements, and yield response data provide the scientific basis for developing and refining fertilizer recommendation systems.

Why mobile nutrients are different

Managing mobile nutrients is fundamentally different from managing immobile ones because their availability is never static. Nitrogen is the best-known example with sulfur following similar principles. Because these nutrients are not held strongly by the soil, their availability changes continuously throughout the growing season. Their supply changes continuously because nutrients are added through natural and management processes and removed by crop uptake and environmental losses. A pre-season soil test captures only a moment in a constantly shifting balance. Recommendations must therefore account for crop demand and all sources and losses throughout the entire growing season, an approach fundamentally different from the soil-test correlation methods used for immobile nutrients.

Crop nutrient demand

The first step in most mobile nutrient recommendation systems is estimating crop nutrient demand. Nutrient demand is commonly estimated from expected yield and the quantity of nutrient required to produce that yield, which reflects the predictable amounts of nutrients crops need to support growth, development, and grain production. Although yield-goal approaches are simple and widely adopted, their accuracy depends on realistic yield expectations and sound assumptions regarding nutrient use efficiency. Estimating crop nutrient demand establishes how much nutrient must be supplied by the soil–plant system and, ultimately, by fertilizer.

Field nutrient supply

Once crop nutrient demand has been estimated, the next step is determining how much nutrient the field can supply. Field nutrient supply includes nutrients derived from soil reserves, organic matter mineralization, previous crops, biological fixation, manure, irrigation water, and atmospheric deposition. Their contributions vary considerably among fields, years, and management systems.

Many recommendation systems can be summarized by the following relationship:

Fertilizer Requirement = Crop Nutrient Demand − Field Nutrient Supply

This nutrient balance framework forms the basis of nitrogen recommendation systems used worldwide (Stanford, 1973). By accounting for nutrients supplied from multiple sources, recommendations can be adjusted to better match crop needs while improving nutrient use efficiency.

Fertilizer response and economic optimization

Field research remains the foundation for developing mobile nutrient recommendations. Fertilizer response trials conducted across multiple locations and years quantify crop response to nutrient applications and provide the data needed to estimate economically efficient fertilizer rates. Crop response to fertilizer generally follows a diminishing returns curve, where each additional unit of nutrient produces progressively smaller yield increases, meaning the rate that maximizes yield often exceeds the rate that maximizes profit.

The Economic Optimum Nitrogen Rate (EONR) identifies the fertilizer rate that maximizes profit rather than yield. Building on this concept, the Maximum Return to Nitrogen (MRTN) approach uses large regional databases of fertilizer response trials to estimate economically optimal nitrogen rates under varying crop and fertilizer price scenarios (Nafziger et al., 2022). Both approaches recognize that sound fertilizer recommendations must account for agronomic response and economic return simultaneously.

Adaptive nutrient management

Because mobile nutrient availability is strongly influenced by weather and environmental conditions, many recommendation systems incorporate adaptive management strategies. Split applications, sidedress applications, and in-season adjustments allow producers to better synchronize nutrient availability with crop demand while reducing nutrient losses. 

Advances in remote sensing and optical technologies have further expanded these capabilities. Technologies such as NDVI sensors, active canopy sensors, drone imagery, and other optical sensing tools can assess crop nutrient status and identify within-field variability, allowing fertilizer rates to be adjusted during the growing season. A notable example is the algorithm developed by Raun et al. (2005), which combined NDVI measurements, yield prediction, and estimates of crop responsiveness to nitrogen to calculate variable fertilizer rates within production fields. Compared with conventional practices, the approach improved nitrogen use efficiency by more than 15% and helped establish the foundation for modern sensor-based nutrient management. 

Strengths and limitations of nutrient balance approaches

The primary strength of nutrient balance approaches is their ability to integrate crop nutrient demand with nutrient supply from multiple sources, improving nutrient use efficiency while avoiding unnecessary fertilizer applications.

A major limitation is the uncertainty associated with estimating nutrient supply because biological processes and management practices vary among fields, years, and cropping systems. In addition, yield goals are sometimes set above realistic expectations, which can result in fertilizer rates that exceed actual crop needs. Consequently, nutrient balance approaches are often supplemented with soil and plant measurements, economic analyses, and adaptive management strategies to improve recommendation accuracy and avoid unnecessary applications.

Recommendations for relatively immobile nutrients

Why immobile nutrients are different

Some nutrients, such as phosphorus, potassium, and many other micronutrients, move only short distances through the soil. Because plant roots must intercept these nutrients, soil fertility levels play a major role in determining nutrient availability and crop response to fertilization. As a result, fertilizer recommendations for relatively immobile nutrients are primarily based on soil test results rather than estimates of crop nutrient demand or environmental processes.

Immobile nutrients often react strongly with soil minerals and organic matter and sorb tightly to soil particles. For example, phosphorus can become temporarily unavailable by reacting with iron, aluminum, and calcium compounds in the soil. Similarly, many micronutrients can be oxidized or precipitated into forms that are less available for plant uptake. These interactions influence nutrient availability, making soil properties such as pH, texture, and organic matter important considerations in fertilizer recommendations.

Because immobile nutrients generally remain in the soil for extended periods of time, fertilizer applications can provide benefits beyond a single growing season. When nutrient applications exceed crop removal, a residual nutrient bank can accumulate in the soil and support future crops. This residual effect makes soil testing a reliable tool for predicting crop response to fertilization.

Correlation and calibration

The value of a soil test depends on its ability to predict crop response to fertilizer application. Soil-testing laboratories measure nutrient concentrations, but these values have little practical meaning unless research relates them to crop response. Soil test correlation and calibration research provides the scientific foundation needed to interpret soil test results and develop fertilizer recommendations. Together, these steps form the basis of soil test correlation and calibration used to develop fertilizer recommendations (Figure 3).

Soil test–crop response relationships

Correlation determines whether a soil test successfully predicts nutrient availability and crop response. This is particularly important because soil tests do not directly measure the amount of each nutrient a plant will absorb. Instead, they estimate the fraction of nutrients that may be available for plant uptake under specific soil conditions. Correlation research evaluates whether nutrients extracted by a soil test are sufficiently related to plant uptake and fertilizer response to support fertilizer recommendations. This transforms a soil test from a laboratory measurement into an agronomically meaningful indicator of nutrient availability.

Relative yield

To compare results across many locations and years, crop performance is often expressed as relative yield. Relative yield is calculated as the yield obtained without fertilizer divided by the maximum yield achieved with fertilization and expressed as a percentage. Using relative yield allows data from diverse environments, crops, and growing conditions to be combined into a common framework for developing fertilizer recommendations.

Critical soil test levels

As soil test values increase, crop response to fertilization generally decreases. The critical soil test level is the value above which a crop is expected to achieve near-maximum yield without additional fertilization. Many recommendation systems define critical soil test levels as the soil test value associated with approximately 90 to 95% relative yield. Below this threshold, fertilizer applications are more likely to increase yield and profitability; above it, the probability of a response declines substantially. Critical soil test levels are established through extensive field research and form the basis of many fertilizer recommendation systems.

Soil test interpretation categories

Because soil test values represent probabilities rather than guarantees, laboratories typically group results into interpretation categories such as very low, low, medium, high, and very high. These categories reflect the likelihood of an economic or agronomic response to fertilizer application. Soils in lower categories generally have a greater probability of responding to fertilizer, whereas soils in higher categories are less likely to show measurable yield or profit increases.

Fertilizer calibration

Once soil test–crop response relationships have been established, calibration research is used to determine fertilizer requirements at different soil fertility levels. Researchers conduct fertilizer rate studies across a range of soil test values and develop yield response curves that relate nutrient application rates to crop yield. These response curves typically show diminishing yield responses, where yield increases become progressively smaller as fertilizer rates increase.

Rather than relying on a single critical value, many recommendation systems recognize a critical range where the probability of response transitions from high to low. Calibration research uses these relationships to develop recommendation categories and fertilizer guidelines that translate soil test values into practical nutrient management recommendations.

Phosphorus and potassium recommendations

Phosphorus and potassium are the most widely studied examples of soil-test-based fertilizer recommendations. Because both nutrients are relatively immobile in soil, recommendations rely heavily on soil testing. Common phosphorus soil tests include Bray-1, Olsen, and Mehlich-3, while exchangeable potassium is frequently measured using ammonium acetate or Mehlich-3 extraction methods. These methods are calibrated for specific regions and soil conditions to ensure accurate prediction of crop response. Critical soil test levels vary considerably among regions because of differences in climate, soils, crops, fertilizer management, and calibration research (Figure 4).

Many phosphorus and potassium recommendation systems follow either sufficiency or a build-and-maintain philosophy. Sufficiency approaches recommend fertilizer only when soil test levels indicate a high probability of economic response and focus on maximizing profitability for the current crop. In contrast, build-and-maintain approaches seek to increase soil test values to a target range and then replace nutrients removed through harvest to maintain long-term soil fertility. When soil test levels are below the target range, fertilizer applications may exceed crop removal to gradually build soil nutrient reserves.

Economic considerations also influence phosphorus and potassium recommendations. The optimal fertilizer rate depends not only on agronomic response, but also on fertilizer costs, crop prices, and expected return on investment. In some situations, applying less than the yield-maximizing rate may be economically justified, whereas in others, maintaining adequate soil fertility may provide long-term benefits that outweigh short-term costs.

Micronutrient recommendations

Micronutrients are required in smaller quantities than macronutrients but remain essential for plant growth and development. Deficiencies are generally less common because soil supplies are often adequate, although their occurrence varies among regions, soil types, and cropping systems. Micronutrient availability is strongly influenced by soil properties, particularly pH, organic matter, and drainage.

Crop response to micronutrient fertilization is often more crop-specific than for phosphorus or potassium. Consequently, recommendations are typically based on a combination of soil tests, plant tissue analysis, crop sensitivity, and local research. Foliar applications can help correct deficiencies during the growing season, whereas soil applications generally provide a longer-term nutrient supply. Chelated micronutrient fertilizers are often used in calcareous or high-pH soils to maintain nutrient availability.

Strengths and limitations of the sufficiency approach

The primary strength of the sufficiency approach is that it is directly linked to observed crop response from field research, allowing fertilizer to be targeted where the probability of an economic response is greatest. Decades of correlation and calibration research have made this approach simple, widely adopted, and scientifically robust.

However, soil test interpretations represent probabilities rather than guarantees. Crop response can still vary because of weather, management, and environmental conditions. In addition, sufficiency-based recommendations focus on the current crop and do not account for long-term nutrient removal or changes in soil fertility. For this reason, many producers combine sufficiency-based recommendations with complementary approaches such as crop removal calculations, nutrient budgeting, and build-and-maintain fertility programs to support long-term nutrient management.

Alternative and complementary recommendation approaches

Although fertilizer recommendations are primarily based on soil testing, nutrient balance, and fertilizer response research, complementary approaches can improve nutrient management decisions. The fertilizer rate that maximizes yield is often greater than the rate that maximizes profit because each additional unit of nutrient typically produces progressively smaller yield increases. Fertilizer costs, crop prices, expected response, production risk, and long-term management objectives all influence fertilizer decisions. Consequently, fertilizer recommendations should be viewed as agronomic guidelines that must ultimately be evaluated within the economic context of each farming operation.

Crop removal and build-and-maintain

Crop removal estimates nutrient removal in harvested products based on crop yield and nutrient concentration (crop removal = yield × nutrient concentration). This information is used alongside soil test levels to determine fertilizer rates in build-and-maintain fertility programs, where fertilizer applications are designed not only to meet immediate crop needs, but also to maintain or increase long-term soil fertility. When soil test values are below target levels, fertilizer rates may exceed crop removal to gradually build soil fertility. When soil tests are within the desired range, nutrient applications typically approximate crop removal to maintain soil fertility over time.

Nutrient budgets and mass balance

Nutrient budgets compare nutrient inputs and outputs over time. Inputs include fertilizer, manure, biological nitrogen fixation, irrigation water, and atmospheric deposition, whereas outputs include harvested crops, runoff, erosion, leaching, and gaseous losses. Positive nutrient balances may indicate nutrient accumulation or increased risk of environmental loss, while negative balances can suggest long-term soil nutrient depletion. Nutrient budgets are particularly useful for evaluating sustainability and nutrient use efficiency at the field, farm, or regional scale. However, because many nutrient inputs and losses are difficult to measure accurately, nutrient budgets are generally better suited for long-term assessment than for determining annual fertilizer rates.

Nutrient uptake and crop demand

Crop nutrient uptake patterns help identify periods of greatest nutrient demand. For example, corn takes up much of its nitrogen between the V6 and silking growth stages, which informs the timing of sidedress applications to better synchronize nutrient availability with peak crop demand. Synchronizing nutrient availability with crop demand can improve nutrient use efficiency and reduce losses, particularly for mobile nutrients. However, nutrient uptake should not be confused with fertilizer requirement. Crops obtain nutrients from many sources, including soil reserves, organic matter mineralization, crop residues, manure, and biological fixation. As a result, total nutrient uptake often exceeds the amount of fertilizer that must be applied.

Plant tissue analysis

Plant tissue analysis provides a direct assessment of crop nutrient status and can help identify nutrient deficiencies, imbalances, or hidden hunger before visible symptoms develop. Tissue testing is particularly valuable for diagnosing in-season nutrient deficiencies and evaluating nutrient status during crop growth. Results must be interpreted carefully because nutrient concentrations vary with crop species, growth stage, plant part sampled, environmental conditions, and other stresses. Consequently, tissue analysis is most effective when used together with soil testing, field observations, and local agronomic knowledge.

Cation balance approaches

The base cation saturation ratio (BCSR) approach proposes that crop performance can be optimized by maintaining specific proportions of calcium, magnesium, and potassium on the soil cation exchange complex. The concept originated from the work of Bear et al. (1945) and was later promoted by William Albrecht, who suggested target saturation ranges for individual cations. Although the BCSR approach remains popular in some advisory systems, decades of research have found little evidence that maintaining specific cation ratios improves crop yield when nutrients are already present at sufficient levels (Kopittke & Menzies, 2007). Most modern fertilizer recommendation systems, therefore, focus on ensuring adequate nutrient availability rather than achieving predetermined cation saturation ratios.

Emerging tools and future directions

Fertilizer recommendation systems are becoming increasingly data-driven, site-specific, and adaptive. New tools allow agronomists to integrate soil test information, field variability, crop sensing, and environmental data into fertilizer recommendations that better reflect local conditions.

Decision support systems and research databases

Large research databases and digital tools are strengthening the scientific foundation of fertilizer recommendations. The Fertilizer Recommendation Support Tool (FRST), launched nationally in 2024, uses a database containing 2,828 fertilizer response trials from 44 states and Puerto Rico to support soil test interpretations. Other initiatives, such as the Crop Nutrient Data platform, are compiling large datasets to support future fertilizer recommendation systems. The database continues to expand with thousands of observations on crop nutrient concentrations, nutrient omission trials, crop residues, and yield responses.

Precision agriculture

Precision agriculture recognizes that soil properties and crop productivity often vary within the same field. Technologies including yield monitors, remote sensing, geographic information systems (GIS), and variable-rate application equipment allow fertilizer rates to be adjusted according to local conditions. A review by Cai et al. (2025) found that these technologies were among the primary drivers of improved nutrient management and nutrient use efficiency.

Artificial intelligence and machine learning

Artificial intelligence (AI) and machine learning are increasingly being explored to integrate soil, weather, crops, and management information into fertilizer recommendations. For example, a recent study reported that a TabNet deep-learning model achieved prediction accuracies exceeding 95% for both fertilizer and crop recommendations (Venkateswara & Padmanaban, 2025). However, not all studies have found consistent results. Using data from commercial farms in the Netherlands, another study found that economically optimal fertilizer rates predicted by machine-learning models varied by 13 to 32% depending on the model and input variables used (Tanaka et al., 2024). Together, these studies highlight the potential of AI while emphasizing the continued need for agronomic validation and uncertainty assessment before widespread adoption.

New fertilizer technologies

New fertilizer technologies, including nanofertilizers, biofertilizers, and enhanced-efficiency fertilizers, are being developed to improve nutrient use efficiency and reduce environmental losses. By better matching nutrient release with crop demand, these products may become important components of future site-specific fertilizer recommendation systems (Maaz et al., 2025).

Toward integrated recommendations

The future of fertilizer recommendations will likely involve integrating multiple sources of information rather than relying on a single measurement or approach. Soil tests, crop measurements, field history, weather, economics, and digital technologies each contribute valuable information for fertilizer recommendations. Together, these advances may support more site-specific recommendations that improve nutrient use efficiency, profitability, and environmental stewardship while recognizing that no single recommendation system is likely to fit all soils, crops, climates, and management conditions.

Conclusion 

Fertilizer recommendations are built on decades of soil testing, field research, and agronomic knowledge. Different recommendation approaches have been developed to account for differences in nutrient behavior, ranging from soil test calibration for relatively immobile nutrients to nutrient balance and economic optimization for mobile nutrients. Complementary tools such as plant tissue analysis, nutrient budgets, precision agriculture, and decision support systems further improve fertilizer recommendations.

Although fertilizer recommendation systems continue to evolve through advances in data science, precision agriculture, and large research databases, they remain grounded in field research and crop response data. No single recommendation approach is appropriate for all soils, crops, climates, and management systems. The most effective fertilizer recommendations combine sound science, local validation, and practical agronomic experience to improve nutrient use efficiency, support profitable crop production, and promote environmental stewardship.