Elsevier

Geoderma Regional

Volume 22, September 2020, e00317
Geoderma Regional

Soil salinity mapping by remote sensing south of Urmia Lake, Iran

https://doi.org/10.1016/j.geodrs.2020.e00317Get rights and content

Highlights

  • Urmia Lake is one of the largest permanent Lakes in the Middle East.

  • Remote sensing techniques are widely utilized for monitoring of salinity.

  • Neural network has high accuracy for predicting the salinity.

Abstract

Urmia Lake is a shallow terminal Lake located in northwest Iran and it is one of the largest permanent Lakes in the Middle East. In this study, the changes in soil salinity at Urmia Lake were investigated using satellite images and the oldest salinity map of the area over a period of 45 years from 1973 to 2018. The distribution of salinity in 2018 was estimated using the supervised classification by the nonlinear hybrid model of artificial multi-layered neural network-genetic algorithm model (ANN-GA) while the salinity map for the years of 1985, 1995, 2005 and 2015 was estimated by the unsupervised method. Further, the salinity data of surface soil in the region for the year 1973 was also digitized and utilized. For this purpose, 291 surface samples (258 samples for modeling and 33 samples for the re-evaluation of the model) of the studied region were collected and analyzed in 2018. The input neurons were selected by analyzing the satellite imagery bands, salinity indices, salinity ratio index and normalized difference vegetation index. The correlation coefficient and root-mean-square error of the training network model were equal to 0.94 and 0.04, respectively. The salinity map of the studied region was estimated using this model and classified into six classes (S0 to S5). The produced map of 2018 was used to re-evaluate the results. It showed that lower estimation accuracy was in classes S1 and S2. The obtained results in this study indicated that roughness, moisture, the density of halophyte plants and sodium slickspot were some of the sources for estimation of errors in lower salinity classes. The time-series changes in the salinity class of estimated maps showed that S3, S4 and S5 classes have expanded between 1973 and 2018. These are in agreement with the field observation and with the other scientific reports about the studied area.

Introduction

Soil salinity leads to land degradation, the decline in soil quality and lack of fertility in arid and semi-arid regions (Sidike et al., 2014; Makinde and Oyelade, 2019). Land degradation caused by soil salinity has been a global issue in dry regions especially in the Middle East (Qadir et al., 2006). Same as the other part of the arid environments, Lakes are exposed to the risk of severe degradation.

In this research the changes in soil salinity of the land surrounding Urmia Lake in Iran, which is the greatest salty Lake in the Middle East and the sixth-largest saltwater Lake on Earth, has been detailed studied. UNESCO Biosphere Reserve in 1976 and UNESCO's Man and the Biosphere (MAB) Program (UNESCO 2009) registered it as a protected region. Recently, the lake has been highly endangered and reduced by about 88% (AghaKouchak et al., 2015). Along with receding of the shoreline, salt accumulation (up to more than 300 g/l) is another prevalent problem (Ahmadzadeh Kokya et al., 2011). If the lake is dry, the region's temperate climate will become tropical and windblown salts will lead to agricultural land degradation, water resources poisoning and respiratory problems for millions of human beings.

Recently, remote sensing techniques have been widely utilized for monitoring studies of salinity due to the high reflectance of saline soil in the visible and infrared range (Farifteh et al., 2007; Alavipanah and Nezammahalleh, 2013, Yu et al., 2018). Updating maps using this technique is quicker and more straightforward than in the conventional methods (Farifteh et al., 2007; Bierwirth and Brodie, 2008; Mulder et al., 2011; Taghizadeh-Mehrjardi et al., 2014).

Remote sensing techniques using intelligent algorithms can predict surface salinity at various time intervals in large-scale regions (Metternicht and Zinck, 2003; Bouaziz et al., 2011; Sidike et al., 2014). Among the intelligent algorithms, the neural networks-genetic algorithms hybrid model (ANN-GA) has mainly been utilized to study the nonlinear relationship between earthly measurements and satellite data with higher accuracy compared to linear models (Seyam and Mogheir, 2011; Phonphan et al., 2014; Naderi et al., 2017; Ghimire et al., 2019; Li and Wang, 2019; Wang et al., 2018). The evolutionary algorithms like GA are coupled with ANN for optimizations of model parameters. Giving more details, the GA is applied to adjust the weights of the ANN properly and obtain the proper network topology (Icaga, 2005; Chang et al., 2010).

The present study first tries to use predict salinity distribution around Lake Urmia using satellite imagery and the ANN-GA hybrid model in 2018 with the real soil samples. Secondly, this study explores changes in the soil surface salinity in the study area from 1985 to 2018 through the processing of satellite images with applying the ANN-GA algorithm without soil samples data. Lastly, the soil salinity map of 1973 as the oldest salinity map of the study area, digitized and utilized in the evaluation of salinity changes. To the best of the authors' knowledge, this study is the first one of this kind conducted in this region.

Section snippets

Location of the studied region

Lake Urmia is a giant shallow hyper saline Lake with many islands in it, enclosed by dense slightly salty to saline marshes, mainly at its southern part. As a brief soil classification, fine textured alluvial soils, saline alluvial soils (Solonchak and Solonetz soil) and salt marsh soils are the dominant classes in the region. Drainage conditions vary from well-drained lands to waterlogged pasturelands or otherwise saline soils (halomorphic) in wastelands (Gholampour et al., 2015).

Urmia Lake

Soil analysis

The results of descriptive statistics for the soil chemical properties are presented in Table 2. The average ECe and pH of the samples are more than those for normal soil in which EC ≤ 2 and pH ≤ 7. Based on the maximum and minimum of ECe values it is evident that the salinity distribution remarkably varies from non-saline to extremely-saline soils. The average sodium absorption ratio (SAR) and exchangeable sodium percentage (ESP) illustrate the presence of sodium ions as a major cation in clay

Conclusion

In the present study, soil surface salinity values were estimated using the ANN-GA model based on the data of Landsat 7 ETM+ Satellite. The model obtained efficiency was evaluated for both supervised and unsupervised classifications. The main finding are listed below:

  • The ANN-GA model was calibrated (train and test) and had considerable efficiency in salinity estimation with real samples (supervised classification).

  • The distribution of the soil salinity surveyed using the calibrated model.

Declaration of Competing Interest

None.

Acknowledgment

The authors gratefully acknowledge the support of the University of Zanjan (ZNU). The authors are also grateful to Mr. Behzad Amiri and Mr. Porya Shahsavari for their aids in the soil sampling step.

References (41)

  • V. Mulder et al.

    The use of remote sensing in soil and terrain mapping—a review

    Geoderma

    (2011)
  • A. Sidike et al.

    Estimating soil salinity in Pingluo County of China using QuickBird data and soil reflectance spectra

    Int. J. Appl. Earth Obs. Geoinf.

    (2014)
  • R. Taghizadeh-Mehrjardi et al.

    Digital mapping of soil salinity in Ardakan region, central Iran

    Geoderma

    (2014)
  • X. Wang et al.

    Estimation of soil salt content (SSC) in the Ebinur Lake Wetland National Nature Reserve (ELWNNR), Northwest China, based on a Bootstrap-BP neural network model and optimal spectral indices

    Sci. Total Environ.

    (2018)
  • T. Ahmadzadeh Kokya et al.

    Evaluation of salt effects on some thermodynamic properties of Urmia Lake water

    Int. J. Environ. Res.

    (2011)
  • S.K. Alavipanah

    Study of desertification and land changes of playa Damghan by using multitemporal and multispectral satellites

    Desert J.

    (2004)
  • S.K. Alavipanah et al.

    The relationship of salt classification with distance to shoreline and elevation, case study Lake Urmia, Iran

    J. Environ. Treat. Tech.

    (2013)
  • A. Allbed et al.

    Mapping and modelling spatial variation in soil salinity in the Al Hassa Oasis based on remote sensing indicators and regression techniques

    Remote Sens.

    (2014)
  • E. Ben-Dor et al.

    The spectral reflectance properties of soil structural crusts in the 1.2-to 2.5-μm spectral region

    Soil Sci. Soc. Am. J.

    (2003)
  • A. Gholampour et al.

    Characterization of saline dust emission resulted from Urmia Lake drying

    J. Environ. Health Sci. Eng.

    (2015)
  • Cited by (0)

    View full text