Wheat Yield Prediction with Machine Learning based on MODIS and Landsat NDVI Data at Field Scale

Tuğaç, Murat Güven; Özbayoğlu, A. Murat; TORUNLAR, Harun; KARAKURT, Erol

doi:10.30897/ijegeo.1128985

Cited by 5 publications

(3 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The approximate average increase in the grain yields of spring wheat was about 6.7 t ha-1, with an increase of 0.1 in NDVI value. Tuğaç et al (2022) found that the highest correlation between NDVI and yield was during the flowering period (R²= 0.63). They also found that the best prediction performance was achieved with the MLP model for MODIS, with a root mean square error (RMSE) ranging from 0.23-0.65 t ha -1 .…”

Section: Discussionmentioning

confidence: 99%

Cereal yield forecasting in semi-arid region of Algeria using MODIS-NDVI

Boulaaras,

Bouregaa

2024

jaa

View full text Add to dashboard Cite

The prediction of cereals yields today is very important for global food security and helps decision-makers in the import-export operations of countries, especially with the rise world population. The advent of remote sensing technologies in precision farming systems has made cereal yield predictions possible, providing valuable insights into the temporal and spatial variations in cereal conditions across both large and small-scale crop lands. Among the various vegetation indices used to analyze these conditions, the normalized difference of vegetation index (NDVI) has emerged as a key indicator. The main objective of this study is to evaluate the possibility of using MODIS-NDVI data to forecast the yield of cereal crops (wheat and barley) in semi-arid region of Algeria (Setif). Additionally, identify the optimal timing for reliable and accurate crop yield forecasts. The remote sensing data utilized in this study covered the growing seasons from February to June, from 2002 to 2022. The results indicated a strong correlation between cereal grain yield and NDVI from late February to mid-March, with R² values ranging from 0.55 to 0.82 for the two cereal species. The RMSE of the NDVI based prediction model ranged from 0.01 t ha-1 to 0.276 t ha-1. The approximate average increase in the grain yield of barley and wheat lies between 0.659 to 0.746 t ha-1 with an increase of 0.1 in NDVI value. These results demonstrate the effectiveness of using MODIS-NDVI data for cereal yield forecasting in semi-arid region of Algeria, offering valuable predictions two to three months before the harvest.

show abstract

Section: Discussionmentioning

confidence: 99%

Cereal yield forecasting in semi-arid region of Algeria using MODIS-NDVI

Boulaaras,

Bouregaa

2024

jaa

View full text Add to dashboard Cite

show abstract

“…The results of this study also demonstrated that in 2022, RVI and NDVI showed a significant positive correlation with seed yield, with the highest correlation coefficient observed at the heading stage. However, as the growth stages progressed, the correlation coefficients gradually declined, which may be related to changes in leaf chlorophyll content and water content [ 42 ]. In conclusion, using data at the heading stage for smooth bromegrass seed yield prediction achieved the highest performance.…”

Section: Discussionmentioning

confidence: 99%

Using Machine Learning Methods Combined with Vegetation Indices and Growth Indicators to Predict Seed Yield of Bromus inermis

Ou,

Jia,

Sun

et al. 2024

Plants

View full text Add to dashboard Cite

Smooth bromegrass (Bromus inermis) is a perennial, high-quality forage grass. However, its seed yield is influenced by agronomic practices, climatic conditions, and the growing year. The rapid and effective prediction of seed yield can assist growers in making informed production decisions and reducing agricultural risks. Our field trial design followed a completely randomized block design with four blocks and three nitrogen levels (0, 100, and 200 kg·N·ha−1) during 2022 and 2023. Data on the remote vegetation index (RVI), the normalized difference vegetation index (NDVI), the leaf nitrogen content (LNC), and the leaf area index (LAI) were collected at heading, anthesis, and milk stages. Multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) regression models were utilized to predict seed yield. In 2022, the results indicated that nitrogen application provided a sufficiently large range of variation of seed yield (ranging from 45.79 to 379.45 kg ha⁻¹). Correlation analysis showed that the indices of the RVI, the NDVI, the LNC, and the LAI in 2022 presented significant positive correlation with seed yield, and the highest correlation coefficient was observed at the heading stage. The data from 2022 were utilized to formulate a predictive model for seed yield. The results suggested that utilizing data from the heading stage produced the best prediction performance. SVM and RF outperformed MLR in prediction, with RF demonstrating the highest performance (R2 = 0.75, RMSE = 51.93 kg ha−1, MAE = 29.43 kg ha−1, and MAPE = 0.17). Notably, the accuracy of predicting seed yield for the year 2023 using this model had decreased. Feature importance analysis of the RF model revealed that LNC was a crucial indicator for predicting smooth bromegrass seed yield. Further studies with an expanded dataset and integration of weather data are needed to improve the accuracy and generalizability of the model and adaptability for the growing year.

show abstract

“…Here, critical phenological states which may hold information necessary to estimate yield amount may be missed. Conversely, at farm-level, [13] deduced that using Landsat improved wheat prediction suggesting the appropriateness of higher spatial resolution for small spatial units.…”

Section: Benefiting From Improved Spatial Resolutionmentioning

confidence: 98%

High Spatial Resolution for Crop Yield Prediction in Large Farming Systems: A Necessity or Additional Overhead

Ofori-Ampofo,

Kuzu,

Xiang Zhu

2023

IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

Concerning county-level yield predictions in large farming systems, relying on coarse-resolution satellite images is customary. However, these images lack sufficient textural detail to accurately summarise spatial information. Our work evaluates the advantage of enhanced spatial resolution by conducting a comparative analysis between coarse-resolution, high-temporal-frequency MODIS data and relatively highresolution, low-temporal-frequency Landsat data for predicting corn yield in the USA. We benchmark this comparison using several models in a spatial versus non-spatial input data case. According to our findings, incorporating high-spatial resolution in this context did not yield any significant benefits, as it comes at the cost of reduced temporal revisit. Root mean square error of about 13 bushels per acre can be achieved 2-4 weeks before harvest for less-intense drought years. Extreme drought-struck years are, however difficult to anticipate.

show abstract

Wheat Yield Prediction with Machine Learning based on MODIS and Landsat NDVI Data at Field Scale

Cited by 5 publications

References 54 publications

Cereal yield forecasting in semi-arid region of Algeria using MODIS-NDVI

Cereal yield forecasting in semi-arid region of Algeria using MODIS-NDVI

Using Machine Learning Methods Combined with Vegetation Indices and Growth Indicators to Predict Seed Yield of Bromus inermis

High Spatial Resolution for Crop Yield Prediction in Large Farming Systems: A Necessity or Additional Overhead

Contact Info

Product

Resources

About