Modeling long-term dynamics of crop evapotranspiration using deep learning in a semi-arid environment

Ahmed Elbeltagi, Jinsong Deng, Ke Wang, Anurag Malik and Saman Maroufpoor

Agricultural Water Management, 2020, vol. 241, issue C

Abstract: Crop evapotranspiration (ETc) is one of the most basic components of the hydrologic cycle that is effective in irrigation system design and management, water resources planning and scheduling, and hydrologic water balance. Thus, precise estimation of ETc is valuable for various applications of agricultural water engineering, especially in developing countries such as Egypt, which has lack of meteorological data, high cost and time to calculate ETc, and lack of information on future ETc values to consider management scenarios and increase production potential. Also, due to the existence of different climates in Egypt, the estimate of ETc has become a challenge. To this end, the aim of this study was to estimate the ETc to eliminate the limitations mentioned, and analyze the long-term dynamics of ETc based on limited climate data and simple method. Three Egyptian governorates namely Ad Daqahliyah, Ash Sharqiyah, and Kafr ash Shaykh of Nile Delta, were selected as major wheat-producing sites. The required historical required climatic data were collected from open access data library while future data were from two extreme scenarios of the Representative Concentration Pathways (RCP) i.e., RCP 4.5, and RCP 8.5. The available dataset was divided into three parts: (i) calibration from 1970−2000, (ii) validation from 2000−2017, and (iii) prediction from 2022−2035. The deep neural network (DNN) was employed for incorporating historical data and predicting future ETc. For the evaluation of generated DNN models, the research finding indicates that the correlation coefficients between actual versus predicted monthly ETc were found to be 0.95, 0.96, and 0.97 for calibration period, and 0.94, 0.95 and 0.95 for validation at Ad Daqahliyah, Kafr ash Shaykh, and Ash Sharqiyah regions, respectively. For the simulation of future climatic data, maximum temperature (Tmax) will increased by 5.19 %, 4.22 %, and 20.82 %, minimum temperature (Tmin) will increased by 1.62 %, 36.44 %, and 27.80 %, and solar radiation (SR) will increased by 6.53 %, 18.74 %, and 28.83 % for the study locations, respectively. Moreover, the DNN model exposed that the Kafr ash Shaykh attain the highest values of ETc with an increase of 11.31 %, slightly increased of 1.38 % for Ad Daqahliyah, and decreased by 15.09 % for Ash Sharqiyah in comparison to the historical data. Thus, the proposed model of crop water-use prediction effectively estimated ETc of wheat and make an efficient decision. The developed models produced satisfactory results for water managers to save water and achieve the sustainability of agricultural water.

Keywords: Crop evapotranspiration; DNN model; Climate change; Future projection (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (10)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0378377420306661
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:agiwat:v:241:y:2020:i:c:s0378377420306661

DOI: 10.1016/j.agwat.2020.106334

Access Statistics for this article

Agricultural Water Management is currently edited by B.E. Clothier, W. Dierickx, J. Oster and D. Wichelns

More articles in Agricultural Water Management from Elsevier
Bibliographic data for series maintained by Catherine Liu ().