Next Article in Journal
Improving Daily Precipitation Estimates by Merging Satellite and Reanalysis Data in Northeast China
Previous Article in Journal
Evaluating MULTIOBS Chlorophyll-a with Ground-Truth Observations in the Eastern Mediterranean Sea
Previous Article in Special Issue
UAV-Based Phenotyping: A Non-Destructive Approach to Studying Wheat Growth Patterns for Crop Improvement and Breeding Programs
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Editorial

An Overview of Within-Season Agricultural Monitoring from Remotely Sensed Data

by
Ruyin Cao
School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Remote Sens. 2024, 16(24), 4706; https://doi.org/10.3390/rs16244706
Submission received: 9 December 2024 / Accepted: 13 December 2024 / Published: 17 December 2024
(This article belongs to the Special Issue Within-Season Agricultural Monitoring from Remotely Sensed Data)

1. Introduction

Remote sensing data have been widely used to monitor various agricultural activities, such as crop distribution mapping, crop phenology extraction, farmland soil moisture monitoring, crop diseases prevention, and crop ideotype breeding. In some practical applications, agricultural information should be acquired in a timely and accurate manner to benefit agricultural management and decision making. For example, it is preferable to acquire crop spatial distributions and to generate reliable yield forecasts before crop harvests within the growing season. Therefore, there is an increasing interest to conduct within-season (in-season) agricultural monitoring using remotely sensed data proven by satellites, drones, or field instruments. Considering this context, we organized a Special Issue entitled “Within-Season Agricultural Monitoring from Remotely Sensed Data” in the Remote Sensing journal. This Special Issue particularly prefers research works related to real-time agricultural monitoring.
This Special Issue has received a total of 19 submissions, ultimately publishing 12 research papers, all of which have undergone a rigorous review process. The objective of this Editorial is to offer an overview of the latest findings and contributions from the studies in this Special Issue. I also provided a short outlook on the research topic of within-season agricultural monitoring.

2. Overview of Contribution and Future Perspectives

The 12 papers in this issue focused on remote sensing of crops. I used the words in the abstract of the 12 papers to generate a cloud of the words (Figure 1). The words “yield”, “growth”, and ”mapping” were found to have high occurrence frequency. Actually, the topics of the published papers cover crop discrimination and distribution mapping (five papers), yield estimation of multiple crops (four papers), crop phenotyping (two papers), and crop phenological stages detection (one paper). Research progress is summarized henceforth, according to research topics.
First, most papers in this Special Issue investigated crop mapping using remotely sensed data, and each paper has its characteristic in terms of mapping methods, crop types, study areas, and remotely sensed data. Wei et al. [1] investigated the potential to identify paddy rice, corn, and soybean at early growing stages in the Songnen Plain of China. They generated combination sets of key identification features for different crops and tested four machine-learning classifiers, including regression tree, random forest, gradient boosting decision tree, and support vector. The results showed that paddy rice in the Songnen Plain can be reliably identified in early May with produce accuracy and user accuracy above 0.95, and corn and soybean can be identified in early July. Gradient boosting decision tree and random forest classifiers have better performance. Wang et al. [2] proposed to integrate regional historical crop planting data with crop spectral information for early-season crop mapping. The authors assumed that crop types can firstly be estimated before the growing season using crop rotation patterns, and then this preseason estimate can be further improved when crop spectral information is available. Their experiments revealed that if the same crop was planted continuously over years or planted with regular time intervals, it can better predict crop types. Early-season mapping of corn and soybean greatly benefits from the fusion of crop rotations and satellite-derived spectral information. Zhan et al. [3] revealed the challenges of Sentinel-2 time-series normalized difference vegetation index (NDVI) data for crop mapping in cloudy areas due to limited cloud-free sentinel-2 images during the growing season. To address this challenge, they generated a temporally continuous NDVI time series by blending MODIS NDVI time-series data with the Sentinel-2 NDVI time-series data. Their results confirmed that the temporally continuous NDVI time series can provide critical phenological characteristics, which improve crop classification performance. Cui et al. [4] focused on crop monitoring in the special high groundwater level mining areas. Considering the growing environment of coal mining subsidence water areas, they employed Sentinel-1 C-band synthetic aperture radar (SAR) and Sentinel-2 optical time-series data. Their results suggest the superiority of the radar vegetation index to track the early growth processes of paddy rice. The effectiveness of combining Sentinel-1 SAR and Sentinel-2 optical data for early-season crop mapping was also demonstrated by Liu et al. [5]. The addition of Sentinel-1 SAR data was found to improve the classification accuracy of winter canola by 2–4% in overall accuracy and 1–2% in F1-score. In addition, winter canola can be identified with an F1-score above 0.85 at the date 130 days before the ripening stage.
Second, crop yield prediction remains a hot topic in the agricultural remote sensing community. Belmahi et al. [6] analyzed the relationship between grain yields and MODIS NDVI values in a local area of Morocco and observed a linear correlation in which an increase of 0.1 in NDVI could take an increase of 490 to 870 kilograms per ha in grain yields. Although their results suggest the importance of remotely sensed data for yield estimations, the lack of other types of data (e.g., meteorological data) lead to only a medium value of correlation coefficient in their regression model. By using both vegetation index data and meteorological data, Zhao et al. [7] developed an adaptive boosting-long short-term memory (AdaBoost-LSTM) ensemble model to estimate winter wheat yields in the Huang-Huai-Hai Plain of China. They found that prediction variables acquired between February and April can generate reliable yield estimates, highlighting the possibility to acquire reginal winter wheat yields at the time one to two months prior to the maturity stage. Peng et al. [8] proposed an interesting strategy to predict crop yields by combining weather forecasts and remotely sensed data. Their method first generated short-term weather forecasts (25 days), and they further incorporated these predicted meteorological data into a deep-learning model, which achieved significant improvements for in-season wheat yield predictions compared to only the remotely sensed data. I expect that this strategy can be more useful in the future, with improved accuracy in weather forecasts. In addition to yield predictions of main grains, one paper in this Special Issue estimated fresh yields of spring tea, an important cash crop in China, using unmanned aerial vehicle images [9].
Third, crop phenology provides important information in agricultural management. In the applications of crop mapping and yield estimations, most of the models need crop phenology information. In this Special Issue, Rodigheri et al. [10] compared different algorithms to detect crop sowing and harvesting dates from MODIS vegetation index time-series data. However, the root mean square error (RMSE) values for the estimates of sowing and harvesting dates were found to be over 15 days, suggesting the challenges to estimate the two phenological stages at the crop field scale. Zhang et al. [11] and Zahra et al. [12] made efforts to link crop spectrum with plant traits using unmanned aerial vehicle and field instruments as high-throughput phenotyping techniques. Specifically, Zhang et al. [11] proposed a simple and low-cost method to reduce spectral variances in field measurements caused by varying diurnal imaging time. Zahra et al. [12] investigated the effectiveness of using NDVI to characterize the wheat stay-green trait. Since the wheat stay-green trait is directly related with grain yields, NDVI is further suggested to be used for crop ideotype breeding.
In future studies, I expect research progress in the following directions.
(1)
Crop mapping: It is necessary to identify crop types at the earliest time in the growing season, but this may depend on specific regions and crop types. For example, previous studies have identified winter canola during the flowering period because of the distinct color characteristic of canola. It is encouraging that Liu et al. [5] proposed the method to identify winter canola at an early overwinter stage in the Jianghan Plain of China. More efforts in different regions and for different crops are needed.
(2)
Yield prediction: Currently, machine learning-based methods have been widely employed for yield prediction. However, two main issues remain. First, crop yield samples, particularly at the field scale, are still lacking. Most crop yield samples are only available at the county scale provided by public datasets, which cannot meet the requirement to train deep learning-based model to predict yields for different crop fields. I expect that more crop yield data at the field scale can be freely released. Second, it is still difficult to interpret the yield predictions given by deep learning-based models, which limit the generalization of these models to other regions. Combining crop growth models and deep learning-based models may further improve crop yield predictions.
(3)
Crop phenology: As one of the most important crop parameters, crop phenology is expected to be acquired accurately and in real time. Collecting field phenology samples is time consuming, laborious, and inefficient. As a result, most studies do not have the intention to share field phenology data [10]. Constructing public crop phenology datasets would benefit the research community. In addition, high-quality vegetation index time-series data cannot be acquired in cloudy areas. Spatiotemporal data fusion may be a useful technology [13], but its effectiveness for crop phenology detection needs more evaluations. Another technology to address data problems is SAR-optical data fusion, which showed promising results in recent studies [14].
Considering the general interests to agricultural monitoring, we continuously organized the second edition entitled “Real-Time Agricultural Monitoring from Remotely Sensed Data” (https://www.mdpi.com/journal/remotesensing/special_issues/2D595ZXXXT. Assessed on 16 December 2024). We welcome more contributions for submission to this Special Issue.

Funding

This work was funded by the National Natural Science Foundation of China (42271379) and the Sichuan Science and Technology Program (Application number: 25GJHZ0125).

Acknowledgments

We thank all authors, reviewers, and assistant editors for their contribution to the Special Issue entitled “Within-Season Agricultural Monitoring from Remotely Sensed Data”.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Wei, M.; Wang, H.; Zhang, Y.; Li, Q.; Du, X.; Shi, G.; Ren, Y. Investigating the Potential of Crop Discrimination in Early Growing Stage of Change Analysis in Remote Sensing Crop Profiles. Remote Sens. 2023, 15, 853. [Google Scholar] [CrossRef]
  2. Wang, Q.; Yang, B.; Li, L.; Liang, H.; Zhu, X.; Cao, R. Within-Season Crop Identification by the Fusion of Spectral Time-Series Data and Historical Crop Planting Data. Remote Sens. 2023, 15, 5043. [Google Scholar] [CrossRef]
  3. Zhan, W.; Luo, F.; Luo, H.; Li, J.; Wu, Y.; Yin, Z.; Wu, Y.; Wu, P. Time-Series-Based Spatiotemporal Fusion Network for Improving Crop Type Mapping. Remote Sens. 2024, 16, 235. [Google Scholar] [CrossRef]
  4. Cui, R.; Hu, Z.; Wang, P.; Han, J.; Zhang, X.; Jiang, X.; Cao, Y. Crop Classification and Growth Monitoring in Coal Mining Subsidence Water Areas Based on Sentinel Satellite. Remote Sens. 2023, 15, 5095. [Google Scholar] [CrossRef]
  5. Liu, T.; Li, P.; Zhao, F.; Liu, J.; Meng, R. Early-Stage Mapping of Winter Canola by Combining Sentinel-1 and Sentinel-2 Data in Jianghan Plain China. Remote Sens. 2024, 16, 3197. [Google Scholar] [CrossRef]
  6. Belmahi, M.; Hanchane, M.; Krakauer, N.Y.; Kessabi, R.; Bouayad, H.; Mahjoub, A.; Zouhri, D. Analysis of Relationship between Grain Yield and NDVI from MODIS in the Fez-Meknes Region, Morocco. Remote Sens. 2023, 15, 2707. [Google Scholar] [CrossRef]
  7. Zhao, Y.; He, J.; Yao, X.; Cheng, T.; Zhu, Y.; Cao, W.; Tian, Y. Wheat Yield Robust Prediction in the Huang-Huai-Hai Plain by Coupling Multi-Source Data with Ensemble Model under Different Irrigation and Extreme Weather Events. Remote Sens. 2024, 16, 1259. [Google Scholar] [CrossRef]
  8. Peng, D.; Cheng, E.; Feng, X.; Hu, J.; Lou, Z.; Zhang, H.; Zhao, B.; Lv, Y.; Peng, H.; Zhang, B. A Deep–Learning Network for Wheat Yield Prediction Combining Weather Forecasts and Remote Sensing Data. Remote Sens. 2024, 16, 3613. [Google Scholar] [CrossRef]
  9. He, Z.; Wu, K.; Wang, F.; Jin, L.; Zhang, R.; Tian, S.; Wu, W.; He, Y.; Huang, R.; Yuan, L.; et al. Fresh Yield Estimation of Spring Tea via Spectral Differences in UAV Hyperspectral Images from Unpicked and Picked Canopies. Remote Sens. 2023, 15, 1100. [Google Scholar] [CrossRef]
  10. Rodigheri, G.; Sanches, I.D.; Richetti, J.; Tsukahara, R.Y.; Lawes, R.; Bendini, H.d.N.; Adami, M. Estimating Crop Sowing and Harvesting Dates Using Satellite Vegetation Index: A Comparative Analysis. Remote Sens. 2023, 15, 5366. [Google Scholar] [CrossRef]
  11. Zhang, J.; Ma, D.; Wei, X.; Jin, J. Visible and Near-Infrared Hyperspectral Diurnal Variation Calibration for Corn Phenotyping Using Remote Sensing. Remote Sens. 2023, 15, 3057. [Google Scholar] [CrossRef]
  12. Zahra, S.; Ruiz, H.; Jung, J.; Adams, T. UAV-Based Phenotyping: A Non-Destructive Approach to Studying Wheat Growth Patterns for Crop Improvement and Breeding Programs. Remote Sens. 2024, 16, 3710. [Google Scholar] [CrossRef]
  13. Cao, R.; Xu, Z.; Chen, Y.; Chen, J.; Shen, M. Reconstructing High-Spatiotemporal-Resolution (30 m and 8-Days) NDVI Time-Series Data for the Qinghai–Tibetan Plateau from 2000–2020. Remote Sens. 2022, 14, 3648. [Google Scholar] [CrossRef]
  14. Chen, Y.; Cao, R.; Liu, S.; Peng, L.; Chen, X.; Chen, J. A new deep learning-based model for reconstructing high-quality NDVI time-series data in heavily cloudy areas: Fusion of Sentinel 1 and 2 data. Int. J. Digit. Earth 2024, 17, e2407941. [Google Scholar] [CrossRef]
Figure 1. World cloud showing the word occurrence frequency in the abstracts of the published papers in the Special Issue. A bigger size of a word indicates a higher occurrence frequency, and the word color has no special meaning.
Figure 1. World cloud showing the word occurrence frequency in the abstracts of the published papers in the Special Issue. A bigger size of a word indicates a higher occurrence frequency, and the word color has no special meaning.
Remotesensing 16 04706 g001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Cao, R. An Overview of Within-Season Agricultural Monitoring from Remotely Sensed Data. Remote Sens. 2024, 16, 4706. https://doi.org/10.3390/rs16244706

AMA Style

Cao R. An Overview of Within-Season Agricultural Monitoring from Remotely Sensed Data. Remote Sensing. 2024; 16(24):4706. https://doi.org/10.3390/rs16244706

Chicago/Turabian Style

Cao, Ruyin. 2024. "An Overview of Within-Season Agricultural Monitoring from Remotely Sensed Data" Remote Sensing 16, no. 24: 4706. https://doi.org/10.3390/rs16244706

APA Style

Cao, R. (2024). An Overview of Within-Season Agricultural Monitoring from Remotely Sensed Data. Remote Sensing, 16(24), 4706. https://doi.org/10.3390/rs16244706

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop