Abstract
This article delves into the intricate details of a robust machine learning analysis that was conducted on a plethora of environmental data, including precipitation, temperature, soil moisture, and vegetation index data, in three distinct regions of Turkey, namely the Aegean, Southeastern Anatolia, and the Mediterranean. The core objective of this scientific inquiry is to shed light on the quintessential determinants that wield a profound influence on agricultural productivity, with an explicit focus on the soil's moisture, temperature fluctuations, and precipitation patterns. It is of paramount importance to fathom the intricacies and multifarious dimensions of these pivotal determinants to enrich our understanding of the entangled dynamics between the ecosystem and crop cultivation. The intricate nature of soil's moisture is a multifaceted interplay, encompassing water availability and the delicate interconnectedness between precipitation, soil structure, and vegetative growth, which can instigate a series of biological and chemical reactions. Furthermore, the study underscores the significance of monitoring the normalized difference vegetation index (NDVI) as a critical indicator of vegetation growth and yield. The outcomes of this study are truly fascinating and highlight the enormous potential of AI-powered tools, which incorporate advanced machine learning and deep learning models, in elevating and optimizing crop management practices, thereby leading to heightened crop productivity and profitability while promoting sustainability. The revolutionary discoveries made in this study underscore the tremendous potential of artificial intelligence (AI) technologies to propel and elevate the agricultural forecasting and management processes, resulting in a more sustainable, productive, and efficient agricultural industry that bestows substantial environmental, social, and economic advantages.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Change history
24 June 2024
A Correction to this paper has been published: https://doi.org/10.1007/s13762-024-05840-0
References
Alkan M, Karaman M, İlkan CC (2021) Precision agriculture and its potential impact on agricultural productivity: evidence from Turkey. Agric Econ Rev 22(1):95–105
Anderson D, Jackson P (2019) Impact of IoT-based monitoring systems on crop yield: a California case study. J Agric Technol 35(4):1024–1037
Bozkurt M, Koc AB (2020) The role of smart agriculture in sustainable development: a review. J Clean Prod 258:120882
Brambilla M, Romano E, Toscano P, Cutini M, Biocca M, Ferré C, Comolli R, Bisaglia C (2021) From conventional to precision fertilization: a case study on the transition for a small-medium farm. Agri Engi 3(2):438–446. https://doi.org/10.3390/agriengineering3020029
Chen J, Xu X, Zhang X, Wu S, **e J (2019a) Application of machine learning in crop yield prediction: a review. Comput Electron Agric 162:219–227
Chen L, Gong Y, Liu J (2019b) Smart agriculture for food security and rural development: New technologies and opportunities. J Clean Prod 222:785–790
Costa F, Silva L (2021) Precision fertilizer application: A Brazilian perspective. J Agric Sci Technol 33(5):789–800
Eroğlu E (2021) Precision agriculture in Turkey: status, challenges, and opportunities. Precis Agric 22(2):335–348
Garcia M et al (2019) Enhancing water use efficiency in Californian agriculture through smart irrigation systems. Am J Environ Eng 29(2):210–225
Gebbers R, Adamchuk VI (2010) Precision agriculture and food security. Science 327(5967):828–831
Guzel A, Ozkan B (2021) The impact of smart agriculture on agricultural productivity: evidence from Turkey. Agric Econ Rural Develop 18(1):49–62
Hegde R, Prasad VK, Kumar R (2019) Machine learning-based identification and diagnosis of plant diseases: a review. Arch Comput Methods Eng 26(4):1073–1088
https://ipad.fas.usda.gov/cropexplorer/util/new_get_psd_data.aspx?regionid=metu
Kelebek MB, Batibeniz F, Önol B (2021) Exposure assessment of climate extremes over the Europe–Mediterranean region. Atmosphere 12(5):633. https://doi.org/10.3390/atmos12050633
Koc AB, Bal M, Gurgen FS (2020) An artificial intelligence-based approach for plant disease detection. J Environ Manage 257:109973
Korkmaz S, Cetin B (2020) Smart agriculture: an overview of technologies and applications. J Agric Sci Technol 22(3):503–516
Koutroulis AG, Papadimitriou LV, Tsihrintzis VA (2019) Smart agriculture: a review of technologies and sustainable approaches for food security. J Clean Prod 233:1288–1303
Lee H et al (2018) The effectiveness of sensor-based technology in pest management: A South Korean farming perspective. Asian J Agric Res 16(3):555–562
Liu L, Zhang X, Zhao W, Li Y, Wang D, He H (2020) Smart agriculture: a review. Engineering 6(6):644–658
Ministry of Agriculture and Forestry (2021) Smart agriculture program. Retrieved from https://www.tarimorman.gov.tr/TarimVeKirsalKalkinmaRehberi/Konular/smart-agriculture-program
Nguyen V et al (2020) Satellite imagery in crop health monitoring: a Vietnamese study. J Remote Sens Agric 18(1):88–102
Öztemiz E, Şahin E (2020) Internet of things in agriculture: an empirical study from Turkey. Int J Adv Comput Sci Appl 11(7):384–390
Patel S, Kumar A (2020) Advancements in precision farming: a study in Indian agricultural context. Int J Smart Agric 22(1):134–145
Qin Y, Yu H, Wang X (2021) Precision agriculture technology and its impact on the environment: a review. Environ Pollut 277:116707
Republic of Turkey Ministry of Agriculture and Forestry (2021) National smart agriculture strategy (2021–2025). Retrieved from https://www.tarimorman.gov.tr/Haberler/2021/Turkiyenin-Ulusal-Akili-Tarim-Stratejisi-yayimlandi
Republic of Turkey Ministry of Agriculture and Forestry (2022) Digital agriculture market. Retrieved from https://www.tarimorman.gov.tr/Birimler/Bas%C4%B1n-Yay%C4%B1n-ve-Halkla-%C4%B0
Sarihan V, Tekgunduz S (2021) Predicting corn yield using machine learning algorithms in Turkey. Agric Meteorol 307:108520
Şenol S, Şimşek Ö, Balcı S (2021) Precision agriculture applications in hazelnut cultivation in Turkey: an economic analysis. Turk J Agric for 45(1):37–45
Silva E, Moita G, de Sousa AB, Monteiro A (2019) Smart agriculture for sustainable rural development. Environ Sci Pollut Res 26(13):13212–13223
Tahiri AZ, Carmi G, Ünlü M (2020) Promising water management strategies for arid and semiarid environments. In: Loures L, Ergen M (eds) Landscape architecture - processes and practices towards sustainable development. IntechOpen. https://doi.org/10.5772/intechopen.87103
Turkish Statistical Institute (2022) Agriculture and forestry statistics 2021. Retrieved from https://data.tuik.gov.tr/Bulten/Index?p=Agriculture-and-Forestry-Statistics-2021-37404
TurkStat (2021) Agriculture statistics. Retrieved from https://tuik.gov.tr/PreTablo.do?alt_id=1007.
World Bank (2017) Harvesting prosperity: technology and productivity growth in agriculture. Retrieved from https://openknowledge.worldbank.org/bitstream/handle/10986/28539/9781464810109.pdf
Zhang X, Huang Y, Liu L, Feng H (2019) A review of smart agriculture research. Sustainability 11(21):6109
Zhou W, Gong Y, Chen L (2019) The economic impact of smart agriculture: a case study of China. J Clean Prod 235:1016–1023
Acknowledgements
We thank the anonymous referees for their suggestions.
Funding
No funding was received.
Author information
Authors and Affiliations
Contributions
M.U.A. and E.N. contributed to conceptualization and resources and wrote the manuscript. E.N. contributed to methodology and investigation. . M.U.A. supervised the study.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
This article does not contain any studies with human participants performed by any of the authors.
Additional information
Editorial responsibility: M. Shabani.
Appendix: implementation & testing
Appendix: implementation & testing
\({\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}}\_{\varvec{t}}{\varvec{e}}{\varvec{s}}{\varvec{t}}\_{\varvec{s}}{\varvec{p}}{\varvec{l}}{\varvec{i}}{\varvec{t}}\): This function is used to split the dataset into training and testing sets.
\({\varvec{D}}{\varvec{e}}{\varvec{c}}{\varvec{i}}{\varvec{s}}{\varvec{i}}{\varvec{o}}{\varvec{n}}{\varvec{T}}{\varvec{r}}{\varvec{e}}{\varvec{e}}{\varvec{R}}{\varvec{e}}{\varvec{g}}{\varvec{r}}{\varvec{e}}{\varvec{s}}{\varvec{s}}{\varvec{o}}{\varvec{r}}\): This is the class for creating a decision tree regression model.
\({\varvec{m}}{\varvec{e}}{\varvec{a}}{\varvec{n}}\_{\varvec{s}}{\varvec{q}}{\varvec{u}}{\varvec{a}}{\varvec{r}}{\varvec{e}}{\varvec{d}}\_{\varvec{e}}{\varvec{r}}{\varvec{r}}{\varvec{o}}{\varvec{r}}\) and \({\varvec{r}}2\_{\varvec{s}}{\varvec{c}}{\varvec{o}}{\varvec{r}}{\varvec{e}}\): These are metrics used to evaluate the performance of regression models.
The dataset was split into two parts: \({\varvec{X}}\) represents the features (attributes) of the data, and \({\varvec{y}}\) represents the target variable (wheat crop yield). The \({\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}}\_{\varvec{t}}{\varvec{e}}{\varvec{s}}{\varvec{t}}\_{\varvec{s}}{\varvec{p}}{\varvec{l}}{\varvec{i}}{\varvec{t}}\) function divides the data into training and testing sets. Here, 80% of the data is used for training (\({\varvec{X}}\_{\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}},\) \({\varvec{y}}\_{\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}})\) and 20% is reserved for testing (\({\varvec{X}}\_{\varvec{t}}{\varvec{e}}{\varvec{s}}{\varvec{t}},\) \({\varvec{y}}\_{\varvec{t}}{\varvec{e}}{\varvec{s}}{\varvec{t}}\)). The \({\varvec{r}}{\varvec{a}}{\varvec{n}}{\varvec{d}}{\varvec{o}}{\varvec{m}}\_{\varvec{s}}{\varvec{t}}{\varvec{a}}{\varvec{t}}{\varvec{e}}\) parameter ensures that the random splitting is reproducible.
The \({\varvec{D}}{\varvec{e}}{\varvec{c}}{\varvec{i}}{\varvec{s}}{\varvec{i}}{\varvec{o}}{\varvec{n}}{\varvec{T}}{\varvec{r}}{\varvec{e}}{\varvec{e}}{\varvec{R}}{\varvec{e}}{\varvec{g}}{\varvec{r}}{\varvec{e}}{\varvec{s}}{\varvec{s}}{\varvec{o}}{\varvec{r}}\) class is created to build a decision tree regression model. This model is then trained (fitted) using the training data (\({\varvec{X}}\_{\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}},\) \({\varvec{y}}\_{\varvec{t}}{\varvec{r}}{\varvec{a}}{\varvec{i}}{\varvec{n}})\).
Using the trained model, predictions are made on the testing data (\({\varvec{X}}\_{\varvec{t}}{\varvec{e}}{\varvec{s}}{\varvec{t}})\). The predicted values are stored in the \({\varvec{y}}\_{\varvec{p}}{\varvec{r}}{\varvec{e}}{\varvec{d}}\) variable.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Uzunoz Altan, M., Nabatov, E. Using machine learning to enhance agricultural productivity in Turkey: insights on the importance of soil moisture, temperature and precipitation patterns. Int. J. Environ. Sci. Technol. 21, 6981–6998 (2024). https://doi.org/10.1007/s13762-023-05439-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13762-023-05439-x