Conference PaperPDF Available

Generation of Site-specific Nitrogen Response Curves for Winter Wheat using Deep Learning

Authors:

Abstract and Figures

Nitrogen fertilizer response (N-response) curves are tools used to support farm management decisions. The conventional approach to model an N-response curve is to fit crop yield in response to a range of N fertilizer rates as a quadratic or exponential function. The purpose of the model is to identify the profit-maximizing N rate given the costs of nitrogen and the price paid for the crop yield. We show that N-response curves are not only field-specific but also site-specific and, as such, economic optimal (profit-maximizing) rates should be calculated for each field each year promoting the use of on-field precision experiments (OFPE) utilizing precision agriculture technologies. We propose a methodology that allows deriving N-response curves automatically instead of using parametric curve-fitting approaches. Thus, we obtain a specific non-parametric N-response curve for each 10 m x 10 m cell of a grid virtually draped on the field. First, we train a convolutional neural network called Hyper3DNetReg using remotely sensed data collected during the early stage of the winter wheat growing season (March) to predict crop harvest yield values. The neural network models the behavior of the field under different environmental and terrain conditions. Then, we use the trained prediction model to obtain an N-response curve per cell by simulating what would be the yield response given a range of nitrogen rate values between 0 and 150 pounds per acre (lbs/ac). Results show that the shape of the N-response curve depends on the region of the field from which it was calculated. Related work will address the problem of generating prescription maps that merge the site-specific economic optimal rates calculated from our N-response curves while also minimizing the overall fertilizer applied and the number of jumps between consecutive cells' nitrogen rates.
Content may be subject to copyright.
The authors are solely responsible for the content of this paper, which is not a refereed publication. The citation of this work should
state that it is from the Proceedings of the 15th International Conference on Precision Agriculture. EXAMPLE: Last Name, A. B. &
Coauthor, C. D. (2018). Title of paper. In Proceedings of the 15th International Conference on Precision Agriculture (unpaginated,
online). Monticello, IL: International Society of Precision Agriculture.
Generation of Site-specific Nitrogen Response
Curves for Winter Wheat using Deep Learning
Giorgio Morales1, John Sheppard1, Amy Peerlinck1, Paul Hegedus2, Bruce
Maxwell2
1Gianforte School of Computing, Montana State University, Bozeman, MT.
2Land Resources & Environmental Science, Montana State University, Bozeman, MT.
A paper from the Proceedings of the
15th International Conference on Precision Agriculture
June 26-29, 2022
Minneapolis, Minnesota, United States
Abstract.
Nitrogen fertilizer response (N-response) curves are tools used to support farm management
decisions. The conventional approach to model an N-response curve is to fit crop yield in
response to a range of N fertilizer rates as a quadratic or exponential function. The purpose of
the model is to identify the profit maximizing N rate given the costs of nitrogen and the price
paid for the crop yield. We show that N-response curves are not only field-specific but also site-
specific and, as such, economic optimal (profit maximizing) rates should be calculated for each
field each year prompting the use of on-field precision experiments (OFPE) utilizing precision
agriculture technologies. We propose a methodology that allows deriving N-response curves
automatically instead of using parametric curve fitting approaches. Thus, we obtain a specific
non-parametric N-response curve for each 10 m x 10 m cell of a grid virtually draped on the
field. First, we train a convolutional neural network called Hyper3DNetReg using remote sensed
data collected during the early stage of the winter wheat growing season (March) to predict crop
harvest yield values . The neural network models the behavior of the field under different
environmental and terrain conditions. Then, we use the trained prediction model to obtain an N-
response curve per cell by simulating what would be the yield response given a range of
nitrogen rate values between 0 and 150 pounds per acre (lbs/ac). Results show that the shape
of the N-response curve depends on the region of the field from which it was calculated. Related
work will address the problem of generating prescription maps that merge the site-specific
economic optimal rates calculated from our N-response curves while also minimizing the overall
fertilizer applied and the number of jumps between consecutive cells’ nitrogen rates.
Keywords.
Nitrogen response curves, Site-specific, Deep learning.
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
2
Introduction
The economic optimal nitrogen rate (EONR) is defined as the nitrogen rate beyond which there
is no actual profit for the farmers. Estimation of the EONR is usually obtained after fitting pre-
selected, parametric yield response functions to crop yield data. Some traditional approaches
assume plateau-type, quadratic, and exponential functions (Bullock & Bullock, 1994; Watkins, et
al., 2010; Kablan, et al., 2017). Other approaches are based on basic agronomic principles, as is
the case of the Liebig response functions (Ackello-Ogutu, Paris, & Williams, 1985). It is worth
mentioning that different crop yield data models can produce different EONR estimations
(Meisinger, Schepers, & Raun, 2008).
Previous works have documented that the EONR is highly dependent on the type of crop, soil,
environmental conditions, and other factors (Nyiraneza, et al., 2010; Tremblay, et al., 2012).
Some work has tried to account for the field and year variability by introducing stochasticity on
crop yield models that used conventional functional forms, obtaining better results than their
deterministic counterparts (Tembo, et al., 2008; Tumusiime, et al., 2011; Boyer, et al., 2013).
However, the variability of the nitrogen response (N-response) functional form across the fields
has not been widely addressed by previous work (Anselin, Bongiovanni, & Lowenberg-DeBoer,
2004; Maxwell, et al., 2018).
We argue that trying to fit a single N-response curve for an entire field implies the strong
assumption that the field is homogeneous and behaves similarly everywhere. However, it is
reasonable to consider that this assumption does not always hold when there exists variability in
the terrain of the field (e.g., terrain slope and soil composition). Thus, in this work, we consider
that N-response curves are not only field-specific and year-specific but also site-specific.
In particular, we propose to use a convolutional neural network (CNN) called Hyper3DNetReg
(Morales & Sheppard, 2021) that maps the spatial features into predicted yield values. The CNN
acts as a complex crop yield data model whose implicit functional form is non-parametric and is
learned from the data. We use an early-yield prediction dataset and show how different N-
response curves can be generated for different regions of a field using the learned yield prediction
model. In addition, we explain that having site-specific N-response curves allows for site-specific
N recommendations.
Proposed Methodology
Datasets
In previous work, we presented a curated early-yield prediction dataset of winter wheat (Morales
& Sheppard, 2021; Hegedus, 2022). In this case, the early-yield prediction of winter wheat can be
viewed as a regression problem where its explanatory variables are determined by the set of
features collected during the growing season (March). Below, we report the list of features that
were used:
Nitrogen rate applied.
Terrain slope.
Terrain elevation.
Terrain aspect.
Topographic position index (TPI).
Sentinel-1 Vertical Transmit-Vertical Receive Polarization (VV) backscatter coefficients.
Sentinel-1 Vertical Transmit-Horizontal Receive Polarization (VH) backscatter
coefficients.
The yield value in bushels per acre (bu/ac) is considered the response variable of the regression
problem. The yield is measured during the harvest season (August) using a combine harvester
equipped with a yield monitor that collects georeferenced values from the field. Hence, the data
acquired from the winter wheat fields in March is used to predict crop yield values in August of
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
3
the same year.
Fig 1. Image rasters corresponding to the G1 field. (a) Yield raster. (b) Automatic extraction of
𝟓 × 𝟓
m pixel patches.
Each field was divided into a grid where each cell represents a region of
10 ×10
m. By doing this,
all the data was aggregated on a scale of 10 m. Fig 1a shows a yield raster for one of the fields
of our dataset from 2018 where each pixel represents an area of 10 m x 10 m.
We chose Hyper3DNetReg as the yield prediction model due to its ability to exploit spatial
information from its two-dimensional (2-D) input patches by using 2-D and 3-D convolutional
filters. Thus, the collected data rasters need to be pre-processed to create the training datasets
used to fit the Hyper3DNetReg models. Specifically, we automatically extract square patches
using a
5 × 5
pixel window allowing a maximum overlap of 0.75, as shown in Fig 1b. This
approach differs from traditional methods that rely on models trained on 1-D features regardless
of the spatial distribution (Gonzalez-Sanchez, Frausto-Solis, & Ojeda-Bustamante, 2014; Kim &
Lee, 2016; Wei, et al., 2020). Table 1 shows the total number of extracted samples and the years
of observation for each field used in our experiments. Here, F1 and G1 are two fields of winter
wheat from two different farms in Montana.
Table 1. Number of samples and years of observation for each field
Field
# Samples 1st
Year
# Samples 2nd
Year
# Samples 3rd
Year
Observed Years
F1
408
316
317
2016, 2018, 2020
G1
484
497
614
2016, 2018, 2020
Yield Prediction
In this section, we use the collected datasets to train and test the Hyper3DNetReg models. These
models are field-specific; that is, they are trained on data of a given field from previous years
(2016 and 2018 in this case) and used to predict yield maps using data from the last observed
year (2020 in this case) of the same field.
In our previous work (Morales & Sheppard, 2021), we showed that 2-D deep regression models
yielded better results than regression models that use a single output. Therefore, in this work, we
only use Hyper3DNetReg models with output windows of
5 × 5
pixels (i.e., the input window size
matches the output window size), as shown in Fig 2.
After training the Hyper3DNetReg models using data from 2016 and 2018, they are tested on
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
4
data from 2020. Here, the predicted yield maps are obtained after averaging the resulting
overlapping output patches. We refer the reader to our previous work (Morales & Sheppard, 2021)
for further details on how the predicted yield maps are generated. Table 2 shows the metrics
obtained after comparing the generated predicted maps and their corresponding ground truth.
The metrics that are used for comparison are root mean square error (
𝑅𝑀𝑆𝐸
), root median square
error (
𝑅𝑀𝑒𝑑𝑆𝐸
), and average structural similarity using a windows size of 3 (
𝑆𝑆𝐼𝑀3
) and 11
(
𝑆𝑆𝐼𝑀11
). Fig 3 illustrates the comparison between the ground-truth maps and the predicted yield
maps that results from applying our Hyper3DNetReg network. The performance values shown in
Table 2 are slightly better than those shown in the original paper due to more careful
hyperparameter tuning.
Fig 2. Hyper3DNetReg yield prediction model using an
𝟓 × 𝟓
pixel output window.
Table 2. Yield prediction results
Field
𝑅𝑀𝑆𝐸
𝑅𝑀𝑒𝑑𝑆𝐸
𝑆𝑆𝐼𝑀3
𝑆𝑆𝐼𝑀11
F1
10.74
6.93
41.63
61.71
G1
14.88
8.94
20.51
42.27
Fig 3. Yield prediction, square error map, and structural similarity map of the (a) F1 field and (b) G1 field.
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
5
N-response Curves
A site-specific Hyper3DNetReg model is a mapping from the feature space to the yield value
space. In our previous work, we demonstrated that the Hyper3DNetReg models trained for
different fields performed better than other traditional and more recent machine learning methods
(e.g., linear regression, stacked autoencoders, and 3-D CNNs). This implies that our
Hyper3DNetReg network models the mapping from the feature space to the yield value space
better than other approaches. Therefore, here we propose a method to generate N-response
curves that are specific to each location of the field using Hyper3DNetReg models. To do so, we
apply the learned Hyper3DNetReg model at each location of the field with varying nitrogen rate
values between 0 and 150 pounds per acre (lbs/ac) with a step size of 3 lbs/ac.
For example, Fig 4a depicts the original nitrogen rate map applied on F1 in 2020 and the
corresponding yield prediction. In contrast, Fig 4b, Fig 4c, and Fig 4d show the results of applying
constant nitrogen rates of 50, 75, and 100 lbs/ac, respectively.
Fig 4. Yield prediction for the G1 field using different N rates. (a) N rate values applied on the field during the growing
season (March 2020). (b) Uniform N rate of 50 lbs/ac. (c) Uniform N rate of 75 lbs./ac. (d) Uniform N rate of 100 lbs./ac.
From the yield maps generated after simulating using a wide range of N rates for the entire field,
we obtain an N-response curve for each
10 ×10
m cell of the field. Fig 5 and Fig 6 show six
different N-response curves generated from the F1 and G1 fields, respectively. Note that, in both
cases, different regions of the field react differently to the nitrogen rate applied. Specifically, some
regions of the fields reach a plateau for high N rates while others experience a rapid decline.
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
6
Fig 5. N-response curves generated for different regions of the F1 field.
Fig 6. N-response curves generated for different regions of the G1 field.
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
7
From Fig 5 and Fig 6, it is observed that the slope and the position of the local maximum of the
generated N-response curves vary depending on the field location. This is important because
traditional methods suggest that the EONR point can be found as the N rate at which the first
derivative of the N-response curve which depends directly on the slope is equal to a common
yield-nitrogen price ratio (Bullock & Bullock, 1994). Thus, considering that the EONR point is
estimated from a single N-response curve and that we are obtaining different N-response curves
for different regions of the field, we would be able to find different EONR points for different regions
of the field
It is worth mentioning that traditionally the EONR calculation assumes concave functional forms
with a single local optimum (e.g., quadratic and exponential functions). On the other hand, the N-
response curves generated by our non-parametric approach may have more than one local
optimum. This can be seen in two of the N-response curves shown in Fig 5 that have at least two
local maxima. Therefore, the traditional EONR calculation is not sufficient for these cases.
Furthermore, we argue that uncertainty plays a crucial role in yield prediction and EONR
estimation. Uncertainty may arise due to representational bias, training data variance, and
parameter uncertainty (Pearce, Brintrup, Zaki, & Neely, 2018). It can be quantified using
prediction intervals (PIs) that consist of an estimate of the upper and the lower bounds within
which a prediction will fall with a certain probability. For future work, we will present an N-response
curve generation method coupled with non-parametric PIs generated by the modified
Hyper3DNetReg models automatically.
Finally, we should consider that the EONR values obtained from our site-specific N-response
curves may differ significantly between neighboring cells. We refer to this phenomenon as
“jumps.” Hence, jumps put a strain on the farmer’s equipment, so it becomes necessary to
minimize the jumps when determining the prescription maps (Peerlinck, Sheppard, Pastorino, &
Maxwell, 2019). Therefore, future work will focus on generating prescription maps that merge the
site-specific economic optimum points calculated from our N-response curves while also
minimizing the overall fertilizer applied and the number of jumps between consecutive cells’
nitrogen rates.
Conclusion
Traditionally, EONR estimation is carried out assuming parametric N-response curves for an
entire field. In this work, we have proposed that N-response curves should be site-specific, and
their functional form can be learned from data.
In our experiments, we have used the Hyper3DNetReg convolutional neural network, which is an
accurate yield prediction model that can be used to generate N-response curves. Thus, our initial
results using two different winter wheat fields showed that different regions of the field have
different responses to the N fertilizer. Moreover, the N-response curves generated by our method
may have more than a single local optimum. Therefore, conventional EONR calculation methods
cannot be applied directly.
Future work will address the need of incorporating uncertainty quantification into the N-response
curve generation by using automatically generated prediction intervals. Furthermore, these
curves will be used to generate fertilizer prescription maps based on multi-objective optimization
while considering other factors such as the minimization of jumps.
Acknowledgements
The authors thank the team members of the On-Field Precision Experiment (OFPE) project for
their comments throughout the development of this work. This research was supported by a
USDA-NIFA-AFRI Food Security Program Coordinated Agricultural Project, titled “Using
Precision Technology in On-farm Field Trials to Enable Data-Intensive Fertilizer Management,”
Proceedings of the 15th International Conference on Precision Agriculture
June 26-29, 2022, Minneapolis, Minnesota, United States
8
(Accession Number 2016-68004-24769), and also by the USDA-NRCS Conservation Innovation
Grant from the On-farm Trials Program, titled “Improving the Economic and Ecological
Sustainability of US Crop Production through On-Farm Precision Experimentation” (Award
Number NR213A7500013G021).
References
Ackello-Ogutu, C., Paris, Q., & Williams, W. A. (1985). Testing a von Liebig Crop Response Function against Polynomial
Specifications. American Journal of Agricultural Economics, 67(4), 873-880.
Anselin, L., Bongiovanni, R., & Lowenberg-DeBoer, J. (2004). A Spatial Econometric Approach to the Economics of Site-
Specific Nitrogen Management in Corn Production. merican Journal of Agricultural Economics, 86(3), 675-687.
Boyer, C. N., Larson, J. A., Roberts, R. K., McClure, A. T., Tyler, D. D., & Zhou, V. (2013). Stochastic Corn Yield
Response Functions to Nitrogen for Corn after Corn, Corn after Cotton, and Corn after Soybeans. Journal of
Agricultural and Applied Economics, 45(4), 669681.
Bullock, D. G., & Bullock, D. S. (1994). Quadratic and Quadratic-Plus-Plateau Models for Predicting Optimal Nitrogen
Rate of Corn: A Comparison. Agronomy Journal, 86(1), 191-195.
Gonzalez-Sanchez, A., Frausto-Solis, J., & Ojeda-Bustamante, W. (2014). Predictive ability of machine learning methods
for massive crop yield prediction. Spanish Journal of Agricultural Research, 12(2), 313-328.
Hegedus, P. (2022). Optimizing Site-Specific Nitrogen Fertilizer Management Based on Maximized Profit and Minimized
Pollution. PhD Thesis, Montana State University, Department of Land Resources and Environmental Sciences,
Bozeman, Montana.
Kablan, L. A., Chabot, V., Mailloux, A., Bouchard, M.-È. a., & Bruulsema, T. (2017). Variability in Corn Yield Response to
Nitrogen Fertilizer in Eastern Canada. Agronomy Journal, 109(5), 2231-2242.
Kim, N., & Lee, Y. (2016). Machine Learning Approaches to Corn Yield Estimation Using Satellite Images and Climate
Data: A Case of Iowa State. Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and
Cartography, 34(4), 383-390.
Maxwell, B., Hegedus, P., Davis, P., Bekkerman, A., Payn, R., Sheppard, J., . . . Izurieta, C. (2018). Can optimization
associated with on-farm experimentation using site-specific technologies improve producer management
decisions? Proceedings of the 14th International Conference on Precision Agriculture. Montreal, Quebec,
Canada.
Meisinger, J., Schepers, J., & Raun, W. (2008). Crop nitrogen requirement and fertilizeation. Nitrogen in agricultural
systems, 49, 563-612.
Morales, G., & Sheppard, J. W. (2021). Two-dimensional deep regression for early yield prediction of winter wheat. SPIE
Future Sensing Technologies, 11914, pp. 49-63.
Nyiraneza, J., N'Dayegamiye, A., Gasser, M. O., Giroux, M., Grenier, M., Landry, C., & Guertin, S. (2010). Soil and Crop
Parameters Related to Corn Nitrogen Response in Eastern Canada. Agronomy Journal, 102(5), 1478-1490.
Pearce, T., Brintrup, A., Zaki, M., & Neely, A. (2018). High-Quality Prediction Intervals for Deep Learning: A Distribution-
Free, Ensembled Approach. Proceedings of the 35th International Conference on Machine Learning, (pp.
4072-4081).
Peerlinck, A., Sheppard, J., Pastorino, J., & Maxwell, B. (2019). Optimal Design of Experiments for Precision Agriculture
Using a Genetic Algorithm. 2019 IEEE Congress on Evolutionary Computation (CEC), (pp. 1838-1845).
Wellington, New Zealand.
Te mb o , G . , B r o rs e n , B . W . , E p pl i n , F. M. , & To st ã o , E . ( 2 0 08 ) . Cr o p I n pu t Re s p on s e F u n ct i o ns w it h St o c ha s t ic P la t e au s .
American Journal of Agricultural Economics, 90(2), 424434.
Tremblay, N., Bouroubi, Y. M., Bélec, C., Mullen, R. W., Kitchen, N. R., Thomason, W. E., . . . Ortiz-Monasterio, I. (2012).
Corn Response to Nitrogen is Influenced by Soil Texture and Weather. Agronomy Journal, 104(6), 1658-1671.
Tumusiime, E., Brorsen, B., Biermacher, J. T., Mosali, J., Johnson, J., & Locke, J. (2011). Determining Optimal Levels of
Nitrogen Fertilizer Using Random Parameter Models. Journal of Agricultural and Applied Economics, 43(4),
541-552.
Watkins, K. B., Hignight, J. A., Norman, R. J., Roberts, T. L., Slaton, N. A., Wilson, C. E., & Frizzell, D. L. (2010).
Comparison of Economic Optimum Nitrogen Rates for Rice in Arkansas. Agronomy Journal, 102(4), 1099-
110 8.
Wei, M., Maldaner, L., Ottoni, P., & Molin, J. (2020). Carrot Yield Mapping: A Precision Agriculture Approach Based on
Machine Learning. AI, 1(2), 229-241.
... In the case of agricultural applications, fitting a single N-response curve for an entire field implies that the field is homogeneous and behaves similarly at every location. Nevertheless, recent works suggest that the N-response functional form varies across the fields given the variability of factors such as terrain slope and soil composition [6], [7]. ...
... If this held, the set of passive features would only shift the response curves vertically but would not alter their shape. However, there is no guarantee for this assumption to always hold, as seen in some precision agriculture applications [6], [7]. Thus, we argue that it is essential to acknowledge all possible sources of variability in analyzing the shape of response curves. ...
Preprint
Full-text available
Response curves exhibit the magnitude of the response of a sensitive system to a varying stimulus. However, response of such systems may be sensitive to multiple stimuli (i.e., input features) that are not necessarily independent. As a consequence, the shape of response curves generated for a selected input feature (referred to as "active feature") might depend on the values of the other input features (referred to as "passive features"). In this work we consider the case of systems whose response is approximated using regression neural networks. We propose to use counterfactual explanations (CFEs) for the identification of the features with the highest relevance on the shape of response curves generated by neural network black boxes. CFEs are generated by a genetic algorithm-based approach that solves a multi-objective optimization problem. In particular, given a response curve generated for an active feature, a CFE finds the minimum combination of passive features that need to be modified to alter the shape of the response curve. We tested our method on a synthetic dataset with 1-D inputs and two crop yield prediction datasets with 2-D inputs. The relevance ranking of features and feature combinations obtained on the synthetic dataset coincided with the analysis of the equation that was used to generate the problem. Results obtained on the yield prediction datasets revealed that the impact on fertilizer responsivity of passive features depends on the terrain characteristics of each field.
... This model list is not exhaustive but meant to represent a spectrum of approaches. Different models may be incorporated into the OFPE framework and current research includes exploration of other machine learning approaches such as AdaBoost, Stacked Autoencoders, and Convolutional Networks [34][35][36]. ...
... This implied that the Hyper3DNetReg network modeled the mapping from the feature space to the yield value space better than other approaches. Therefore, a method to generate N-response curves that were specific to each location of the field using Hyper3DNetReg models was proposed [35]. Initial results using two different winter wheat fields showed that different regions of the field have different responses to the N fertilizer ( Figure 5) and thus allowed further refinement of approaches to identify site-specific input management. ...
Preprint
Full-text available
Precision agriculture and open-source data repositories provide a plethora of field-specific ecological data about agroecosystems, but few mechanisms have been developed to turn that information into management recommendations for crop production. The On-Farm Precision Experiments (OFPE) framework is an agroecological model-based methodology to improve crop manager’s abilities to make field-scale agronomic input decisions. This work evaluates the use of field-specific experiments that employ open-source data and the data emanating from precision agriculture technologies to gain local knowledge of the spatial and temporal variability in agroeconomic performance at the sub-field scale. Quantification of the temporal variability in crop response to inputs (e.g., crop seeding rates, crop rotations, fertilizers, other soil amendments, pesticides, etc.) allows for estimation of the probability that a future management scenario will outcompete another, in terms of crop yield, crop quality, farmer net return, or environmental quality. The challenge is to integrate OFPE into applied management with minimal disruption of stakeholder practices while drawing on historic knowledge about the field and economic constraints. OFPE is the basis of a decision support system that includes a six-step cyclical process that harnesses precision agriculture technology to apply experiments and gather field-specific data, incorporates modern data management and analytical approaches, and generates management recommendations as probabilities of outcomes. The OFPE framework allows field managers to assess the tradeoffs in agronomic input management between the maximization of crop production, quality and profits from production while considering environmental effects.
... Fertilizer responsivity can be characterized using nitrogen fertilizer-yield response (N-response) curves. These curves exhibit the estimated crop yield values corresponding to a specific field site in response to all admissible fertilizer rates, typically ranging between 0 and 150 pounds per acre (lbs/acre) [14], [15]. The shape of the N-response curve indicates the site's responsiveness to fertilizer, with a flat curve suggesting low responsivity and a steep curve suggesting high responsivity. ...
Conference Paper
Full-text available
In Precision Agriculture, the utilization of management zones (MZs) that take into account within-field variability facilitates effective fertilizer management. This approach enables the optimization of nitrogen (N) rates to maximize crop yield production and enhance agronomic use efficiency. However, existing works often neglect the consideration of responsivity to fertilizer as a factor influencing MZ determination. In response to this gap, we present a MZ clustering method based on fertilizer responsivity. We build upon the statement that the responsivity of a given site to the fertilizer rate is described by the shape of its corresponding N fertilizer-yield response (N-response) curve. Thus, we generate N-response curves for all sites within the field using a convolutional neural network (CNN). The shape of the approximated N-response curves is then characterized using functional principal component analysis. Subsequently, a counterfactual explanation (CFE) method is applied to discern the impact of various variables on MZ membership. The genetic algorithm-based CFE solves a multi-objective optimization problem and aims to identify the minimum combination of features needed to alter a site's cluster assignment. Results from two yield prediction datasets indicate that the features with the greatest influence on MZ membership are associated with terrain characteristics that either facilitate or impede fertilizer runoff, such as terrain slope or topographic aspect.
... The predictive capabilities of DL can be leveraged for site-specific nutrient management and precision agriculture. With detailed knowledge of the spatial variability of soil properties and nutrient content, farmers can implement precise agricultural practices by applying the right type and amount of nutrients at the right place and time (Chang et al., 2021;Morales, Sheppard, Peerlinck, Hegedus, & Maxwell, 2022). This approach not only enhances agricultural productivity but also minimises the environmental impact by reducing over-fertilisation and nutrient runoff. ...
Article
Deep Learning (DL) has emerged as a pivotal modelling tool in various domains because of its proficiency in learning distributed representations. Numerous DL algorithms have recently been proposed and applied to non-destructive testing (NDT) methods in agriculture. This study aimed to review the state-of-the-art applications of DL algorithms in NDT by analysing the application of DL to specific NDT applications and highlighting their contributions and challenges. It first presents a comprehensive overview of various NDT techniques that have been combined with DL in agricultural product evaluation, and then briefly describes their applications in diverse NDT tasks, such as image classification, object detection, image retrieval, and semantic segmentation. Second, this study addresses the ongoing challenges associated with data collection and fusion, model complexity, computational requirements, and robustness. Finally, future research directions are examined, underscoring the potential of novel neural network architectures and cross-disciplinary collaborations. This review aims to provide a clear understanding of the current state of DL-based NDT in agricultural product examinations and its prospects for the future.
... Fertilizer responsivity can be characterized using nitrogen fertilizer-yield response (N-response) curves. These curves exhibit the estimated crop yield values corresponding to a specific field site in response to all admissible fertilizer rates, typically ranging between 0 and 150 pounds per acre (lbs/acre) [14], [15]. The shape of the N-response curve indicates the site's responsiveness to fertilizer, with a flat curve suggesting low responsivity and a steep curve suggesting high responsivity. ...
Preprint
Full-text available
In Precision Agriculture, the utilization of management zones (MZs) that take into account within-field variability facilitates effective fertilizer management. This approach enables the optimization of nitrogen (N) rates to maximize crop yield production and enhance agronomic use efficiency. However, existing works often neglect the consideration of responsivity to fertilizer as a factor influencing MZ determination. In response to this gap, we present a MZ clustering method based on fertilizer responsivity. We build upon the statement that the responsivity of a given site to the fertilizer rate is described by the shape of its corresponding N fertilizer-yield response (N-response) curve. Thus, we generate N-response curves for all sites within the field using a convolutional neural network (CNN). The shape of the approximated N-response curves is then characterized using functional principal component analysis. Subsequently, a counterfactual explanation (CFE) method is applied to discern the impact of various variables on MZ membership. The genetic algorithm-based CFE solves a multi-objective optimization problem and aims to identify the minimum combination of features needed to alter a site's cluster assignment. Results from two yield prediction datasets indicate that the features with the greatest influence on MZ membership are associated with terrain characteristics that either facilitate or impede fertilizer runoff, such as terrain slope or topographic aspect.
... The models listed are not exhaustive but are meant to represent a spectrum of approaches with different underlying assumptions about the approach. Different models may be incorporated into the OFPE methodology and current research includes the exploration of other machine learning approaches such as AdaBoost, stacked autoencoders, and convolutional networks [40][41][42]. ...
Article
Full-text available
Few mechanisms turn field-specific ecological data into management recommendations for crop production with appropriate uncertainty. Precision agriculture is mainly deployed for machine efficiencies and soil-based zonal management, and the traditional paradigm of small plot research fails to unite agronomic research and effective management under farmers’ unique field constraints. This work assesses the use of on-farm experiments applied with precision agriculture technologies and open-source data to gain local knowledge of the spatiotemporal variability in agroeconomic performance on the subfield scale to accelerate learning and overcome the bias inherent in traditional research approaches. The on-farm precision experimentation methodology is an approach to improve farmers’ abilities to make site-specific agronomic input decisions by simulating a distribution of economic outcomes for the producer using field-specific crop response models that account for spatiotemporal uncertainty in crop responses. The methodology is the basis of a decision support system that includes a six-step cyclical process that engages precision agriculture technology to apply experiments, gather field-specific data, incorporate modern data management and analytical approaches, and generate management recommendations as probabilities of outcomes. The quantification of variability in crop response to inputs and drawing on historic knowledge about the field and economic constraints up to the time a decision is required allows for probabilistic inference that a future management scenario will outcompete another in terms of production, economics, and sustainability. The proposed methodology represents advancement over other approaches by comparing management strategies and providing the probability that each will increase producer profits over their previous input management on the field scale.
Chapter
A key criterion for determining the health of the wheat crop is the nitrogen content. In agricultural modeling, the leaf color chart (LCC) is used to compare the color of the leaf to its matching color in order to track the success of the crop. Nitrogen is essential for plant development and growth, as well as for disease defense. In this paper, support vector machine, convolutional neural network, k-nearest neighbor, and random forest are compared, to assess the nitrogen deficiency level in accordance with the leaf color chart level. A mobile phone camera was used to obtain many samples of photos of wheat leaves. The training, testing, and validation datasets were used in this study. There are total of 1347 images. The comparison is performed using orange data mining tool. According to the findings, convolutional neural networks and random forests are more accurate than support vector machines and k-nearest neighbors.
Conference Paper
Full-text available
Response curves exhibit the magnitude of the response of a sensitive system to a varying stimulus. However, response of such systems may be sensitive to multiple stimuli (i.e., input features) that are not necessarily independent. As a consequence, the shape of response curves generated for a selected input feature (referred to as “active feature”) might depend on the values of the other input features (referred to as “passive features”). In this work we consider the case of systems whose response is approximated using regression neural networks. We propose to use counterfactual explanations (CFEs) for the identification of the features with the highest relevance on the shape of response curves generated by neural network black boxes. CFEs are generated by a genetic algorithm-based approach that solves a multi-objective optimization problem. In particular, given a response curve generated for an active feature, a CFE finds the minimum combination of passive features that need to be modified to alter the shape of the response curve. We tested our method on a synthetic dataset with 1-D inputs and two crop yield prediction datasets with 2-D inputs. The relevance ranking of features and feature combinations obtained on the synthetic dataset coincided with the analysis of the equation that was used to generate the problem. Results obtained on the yield prediction datasets revealed that the impact on fertilizer responsivity of passive features depends on the terrain characteristics of each field.
Article
Full-text available
Carrot yield maps are an essential tool in supporting decision makers in improving their agricultural practices, but they are unconventional and not easy to obtain. The objective was to develop a method to generate a carrot yield map applying a random forest (RF) regression algorithm on a database composed of satellite spectral data and carrot ground-truth yield sampling. Georeferenced carrot yield sampling was carried out and satellite imagery was obtained during crop development. The entire dataset was split into training and test sets. The Gini index was used to find the five most important predictor variables of the model. Statistical parameters used to evaluate model performance were the root mean squared error (RMSE), coefficient of determination (R 2) and mean absolute error (MAE). The five most important predictor variables were the near-infrared spectral band at 92 and 79 days after sowing (DAS), green spectral band at 50 DAS and blue spectral band at 92 and 81 DAS. The RF algorithm applied to the entire dataset presented R 2 , RMSE and MAE values of 0.82, 2.64 Mg ha −1 and 1.74 Mg ha −1 , respectively. The method based on RF regression applied to a database composed of spectral bands proved to be accurate and suitable to predict carrot yield.
Conference Paper
Full-text available
Deep neural networks are a powerful technique for learning complex functions from data. However, their appeal in real-world applications can be hindered by an inability to quantify the uncertainty of predictions. In this paper, the generation of prediction intervals (PI) for quantifying uncertainty in regression tasks is considered. It is axiomatic that high-quality PIs should be as narrow as possible, whilst capturing a specified portion of data. In this paper we derive a loss function directly from this high-quality principle that requires no distributional assumption. We show how its form derives from a likelihood principle, that it can be used with gradient descent, and that in ensembled form, model uncertainty is accounted for. This remedies limitations of a popular model developed on the same high-quality principle. Experiments are conducted on ten regression benchmark datasets. The proposed quality-driven (QD) method is shown to outperform current state-of-the-art uncertainty quantification methods, reducing average PI width by around 10%.
Article
Full-text available
Core Ideas A 8‐yr study of corn N fertilization on high‐yielding fields in Québec, eastern Canada. Grain yield response to N rates varied among site‐years. The economically optimal N rate was affected by soil textural classes, planting date, and rainfall. Averaged across textures, planting date, and weather, economically optimal N rate was 195 kg N ha ⁻¹ . Nitrogen applications at rates above the current N recommendation increased grain yield. Corn ( Zea mays L.) yield response to N has been found to vary spatially within a field. The objective of this study was to examine how grain corn yield response to N varies with planting date, soil texture, and spring weather across sites and years in the Montérégie region. Trials were conducted from 2002 to 2004 and 2006 to 2010, at 11 sites with 23 hybrids and four N application rates, for a total of 45 site‐years. Each site‐year involved five or six N rates ranging from 80–90 to 240 kg N ha ⁻¹ . Grain yield response to N rates varied among site‐years. Trials were separated into two groups based on optimal and late planting dates. Significant differences in grain yield among the applied N rates were observed in all of the site‐years planted at optimal dates (from 8.8–14.7 Mg ha ⁻¹ ), and in most of those planted late (8.5–12.8 Mg ha ⁻¹ ). Economic optimum nitrogen rates (EONR) ranged less widely for site‐years planted on optimal dates (180–237 kg N ha ⁻¹ ) than for those planted late (132–237 kg N ha ⁻¹ ). The EONR was affected by soil textural classes and rainfall. On coarse‐textured soils, more N was needed to optimize grain yield in years with wet growing seasons. These results suggest that the current N recommendations for corn in Quebec should consider the variability in response associated with site‐specific effects of planting date, soil texture, and weather.
Article
Full-text available
Remote sensing data has been widely used in the estimation of crop yields by employing statistical methods such as regression model. Machine learning, which is an efficient empirical method for classification and prediction, is another approach to crop yield estimation. This paper described the corn yield estimation in Iowa State using four machine learning approaches such as SVM (Support Vector Machine), RF (Random Forest), ERT (Extremely Randomized Trees) and DL (Deep Learning). Also, comparisons of the validation statistics among them were presented. To examine the seasonal sensitivities of the corn yields, three period groups were set up: (1) MJJAS (May to September), (2) JA (July and August) and (3) OC (optimal combination of month). In overall, the DL method showed the highest accuracies in terms of the correlation coefficient for the three period groups. The accuracies were relatively favorable in the OC group, which indicates the optimal combination of month can be significant in statistical modeling of crop yields. The differences between our predictions and USDA (United States Department of Agriculture) statistics were about 6-8 %, which shows the machine learning approaches can be a viable option for crop yield modeling. In particular, the DL showed more stable results by overcoming the overfitting problem of generic machine learning methods. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http:// creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Article
Full-text available
Deterministic and stochastic yield response plateau functions were estimated to determine the expected profit-maximizing nitrogen rates, yields, and net returns for corn grown after corn, cotton, and soybeans. The stochastic response functions were more appropriate than their deterministic counterparts, and the linear response stochastic plateau described the data the best. The profit-maximizing nitrogen rates were similar for corn after corn, cotton, and soybeans, but relative to corn after corn, the expected corn yield plateaus increased by 12% and 16% after cotton and soybeans, respectively. Expected net returns increased for corn after cotton and soybeans relative to corn after corn.
Article
Full-text available
Soil properties and weather conditions are known to affect soil N availability and plant N uptake; however, studies examining N response as affected by soil and weather sometimes give conflicting results. Meta-analysis is a statistical method for estimating treatment effects in a series of experiments to explain the sources of heterogeneity. In this study, the technique was used to examine the influence of soil and weather parameters on N response of corn (Zea mays L.) across 51 studies involving the same N rate treatments that were performed in a diversity of North American locations between 2006 and 2009. Results showed that corn response to added N was significantly greater in fine-textured soils than in medium-textured soils. Abundant and well-distributed rainfall and, to a lesser extent, accumulated corn heat units enhanced N response. Corn yields increased by a factor of 1.6 (over the unfertilized control) in medium-textured soils and 2.7 in fine-textured soils at high N rates. Subgroup analyses were performed on the fine-textured soil class based on weather parameters. Rainfall patterns had an important effect on N response in this soil texture class, with yields being increased 4.5-fold by in-season N fertilization under conditions of "abundant and well-distributed rainfall." These findings could be useful for developing N fertilization algorithms that would prescribe N application at optimal rates taking into account rainfall pattern and soil texture, which would lead to improved crop profitability and reduced environmental impacts.
Article
Full-text available
An important issue for agricultural planning purposes is the accurate yield estimation for the numerous crops involved in the planning. Machine learning (ML) is an essential approach for achieving practical and effective solutions for this problem. Many comparisons of ML methods for yield prediction have been made, seeking for the most accurate technique. Generally, the number of evaluated crops and techniques is too low and does not provide enough information for agricultural planning purposes. This paper compares the predictive accuracy of ML and linear regression techniques for crop yield prediction in ten crop datasets. Multiple linear regression, M5-Prime regression trees, perceptron multilayer neural networks, support vector regression and k-nearest neighbor methods were ranked. Four accuracy metrics were used to validate the models: the root mean square error (RMS), root relative square error (RRSE), normalized mean absolute error (MAE), and correlation factor (R). Real data of an irrigation zone of Mexico were used for building the models. Models were tested with samples of two consecutive years. The results show that M5- Prime and k-nearest neighbor techniques obtain the lowest average RMSE errors (5.14 and 4.91), the lowest RRSE errors (79.46% and 79.78%), the lowest average MAE errors (18.12% and 19.42%), and the highest average correlation factors (0.41 and 0.42). Since M5-Prime achieves the largest number of crop yield models with the lowest errors, it is a very suitable tool for massive crop yield prediction in agricultural planning.
Conference Paper
Crop yield prediction is one of the tasks of Precision Agriculture that can be automated based on multi-source periodic observations of the fields. We tackle the yield prediction problem using a Convolutional Neural Network (CNN) trained on data that combines radar satellite imagery and on-ground information. We present a CNN architecture called Hyper3DNetReg that takes in a multi-channel input image and outputs a two-dimensional raster, where each pixel represents the predicted yield value of the corresponding input pixel. We utilize radar data acquired from the Sentinel-1 satellites, while the on-ground data correspond to a set of six raster features: nitrogen rate applied, precipitation, slope, elevation, topographic position index (TPI), and aspect. We use data collected during the early stage of the winter wheat growing season (March) to predict yield values during the harvest season (August). We present experiments over four fields of winter wheat and show that our proposed methodology yields better results than five compared methods, including multiple linear regression, an ensemble of feedforward networks using AdaBoost, a stacked autoencoder, and two other CNN architectures.
Article
Information regarding the relationships between soil properties and the economic optimum N rate for crop yield is needed to ensure profitable use of N fertilizer. This study was conducted in 2007 and 2008 at 62 field sites in Quebec (eastern Canada) to assess corn (Zea mays L.) response to N fertilizer, to calculate the economically optimum nitrogen rate (EONR) and corn yield (EOY), and to relate these two parameters with soil and crop-based parameters. Yield response to N fertilizer rates (0-250 kg N ha(-1)) at each site was fitted to a linear, quadratic, or quadratic-plus-plateau model. The EONR and yield (EOY) were related to 12 soil and crop-based parameters, and corn heat units (CHU). The quadratic-plus-plateau model best described the yield-fertilizer relationship at 43 of the 62 field sites. The values of EOY varied from 7.4 to 13.3 Mg ha(-1) in 2007 and from 5.2 to 11.2 Mg ha(-1) in 2008, while EONR was between 73 and 235 kg N ha(-1) in 2007 and from 48 to 200 kg N ha(-1) in 2008. Correlation and principal component analysis showed that dissolved nitrogen (DNc) and dissolved carbon (DOCc) extracted with cold water and pre-sidedress nitrate analyzed using nitrate test strips (PSNTts) were significantly and negatively related to EONR. In both years, PSNTts was consistently related to EONR, and CHU with EOY. While it remains challenging to predict EONR due to site-specific variability and fluctuations in growing conditions, the PSNTts test shows promise in predicting the EONR for corn production in Quebec.