Investigation of the physical-geographical characteristics of river delta with geophysical and satellite data. The case study of Pineios River, Greece

Abstract

This paper presents the methodology of an applied geophysical and remote sensing research at river deltas for their subsurface and surface structure and its application in the deltaic plain of Pineios River (Thessaly, Greece). The scope is to primarily calculate the thickness of deltaic post alpine deposits, with the contribution of Transient Electromagnetic Method (TEM), but also identify the surficial characteristics (biotic/ abiotic) of the area with the appropriate combination of spectral bands and spectral indices. Regarding the remote sensing approach, some geomorphological features were outlined. This was managed from six false color composites of the area, produced by combining spectral bands and additional five false color composites by combining several spectral indices.

• •The results of the TEM method along with their statistical analysis provided important information regarding the spatial distribution and thickness of the lithological units along the deltaic plain.
• •The combination of spectral bands 8, 7, 6 seems to adumbrate the hydrographic network of the area and even detects possible changes in the inflow.
• •The combination of spectral indices MSAVI2, WV-BI, WV-WI was helpful enough to geomorphological mapping of the deltaic plain.

Graphical abstarct

Specifications table

Subject area:	Earth and Planetary Sciences
More specific subject area:	Applied Geophysics & Remote Sensing Techniques for Environment.
Name of your method:	Combination of Transient Electromagnetic Method (TEM) and Satellite Image Processing for river delta structure.
Name and reference of original method:	Dilalos S., Alexopoulos J.D., Vassilakis Em., Poulos S.E. (2022). Investigation of the structural control of a deltaic valley with geophysical methods. The case study of Pineios river delta (Thessaly, Greece), Journal of Applied Geophysics, 202, 104,652, https://doi.org/10.1016/j.jappgeo.2022.104652 Alexopoulos, J.D., Dilalos, S., Mitsika, G.S., Poulos SE. (2019). Mapping of deltaic aquifers with the combined application of DC and TEM soundings. In: Near Surface Geoscience 2019–25th European Meeting of Environmental and Engineering Geophysics, We_25_P13, 5p. https://doi.org/10.3997/2214–4609.201902468
Resource availability:	Equipment: WalkTEM Software: ViewTEM, ERDAS Imagine and ArcGIS Pro (ESRI software) Software: ViewTEM, ERDAS Imagine and ArcGIS Pro (ESRI software) WorldView-2 satellite images

Method details

Introduction

More than half a billion people live on or close to deltas, often in large cities [1], [2]. There are numerous advantages for societal development established by the existence of Deltas, such as a smooth topography, available fresh resources, good transportation via waterways and the nearby coast, productive soils for agriculture, and a valuable biodiversity and recreational value through, such as beaches, wetland ecosystems and coastal forests [3]. Potential threats such as sea level rise and subsidence come from both environmental change but also human intervention [2]. Taking into account that sea level rise, sand depletion, groundwater pumping and human pressure on delta land are expected to increase, conventional approaches to control delta landscapes may become unsustainable [4]. The determination of the subsurface geological structure is crucial for modeling the paleo-relief where the river delta has been created on, but also allows the delineation of the river's flow paths [5].

Remote sensing techniques in combination with field data and Geographic Information Systems (GIS) have been used in various agricultural applications or research [6], [7], [8], [9]. With the accessibility of global remote sensing databases, the physical attributes of deltaic systems are becoming increasingly better quantified [10], [11], [12], [13], [14]. Geophysical surveys have been conducted in many deltaic regions and coastal zones around the world [15], [16], [17], [18], [19], [20], [21], [22].

Regarding the presented case study of Pineios river, the authors have already discussed the results and the interpretation from the application of several geophysical studies, including the Vertical Electrical Soundings (VES), the Transient Electromagnetic Soundings (TEM) and gravity measurements [23], [24], [25], [26], [27].

Study area

Pineios River is the largest Greek river with its drainage basin within the Greek borders. It is also the only river in Greece with reduced flow control (in only less than 10% of its catchment) [28]. Its delta has been recognized as an environmentally sensitive coastal zone protected by the NATURA 2000 network (RAMSAR convention). It is characterized by great fresh surface water availability supporting a valuable biodiversity of fauna and flora [29]. It includes an estuary with riparian woodland, riparian forests marshes, freshwater lakes, coastal zones and sand dunes. The delta of Pineios River is characterized by the expanding touristic exploitation of the coastline and the extensive agricultural cultivation [30], [31].

The relief of the deltaic plain is flat and incorporates meanders, radial riverbeds and isolated parts of old riverbeds, which function mainly during flood events, revealing the continuous geomorphological evolution of the plain over the years [32]. In Fig. 1, the coverage of the largest part of the deltaic plain by alluvial sediments is obvious, while sand dunes and coastal sediments exist along coastal areas. The northern (Olympus Mt.) and western (Pyrgetos alluvial fan) boundaries of the deltaic plain comprise Neogene deposits (conglomerates and marls) and Pleistocene formations (screes, debris cones and fluvial terraces). The southern margin (Ossa Mt.) is surrounded by metamorphic rocks, such as dolomites, crystalline limestones, marbles, shales, gneisses, blue schists and ophiolites [33]. Based on the authors of [34] the Neogene formations have undergone severe tectonic deformation leading to several block rotations with values of 10–30^ο, towards south. The NW dipping normal fault zone located south of Omolio settlement strikes WSW-ENE and appears to play a substantial role in the rotation of the Neogene formations.

Fig. 1.

Geological map of the study area, modified by [33], along with the location of the Transient Electromagnetic soundings and boreholes of [35].

From a hydrogeological aspect, groundwater occurs in the alluvial sediments of the deltaic plain of Pineios River, which based on [25], is not identified throughout the entire deltaic basin, but only at its central and northern parts. Below the alluvial sediments, a clay formation occurs, as revealed from wells logs, along with a deeper confined aquifer consisting mainly by conglomerates. Therefore, two major aquifer systems are defined: the phreatic and the confined deep aquifer.

Methodology

Our investigation along the deltaic plain of Pineios River is comprised of two parts, the subsurface investigation with geophysical methods and the surficial one, with the contribution of satellite image processing. The coordinate system used in all data is the Greek Geodetic Reference System 1987 (GGRS87/Greek Grid).

Geophysical survey for subsurface investigation

A geophysical survey has been conducted along the deltaic area of Pineios River for the delineation of the subsurface regime and hydrogeological characteristics. In this paper, we will deal with the results of the Transient Electromagnetic Method (TEM) that has been carried out in the area along with other geophysical methods in previous studies, such as Vertical Electrical Soundings (VES) and gravity measurements. The main objective of this geophysical survey was to calculate the thickness of the deltaic deposits followed by the deposition of the transgressive sediments.

The Transient Electromagnetic Method (TEM) has been used for hydrogeological applications [36], [37], [38] and for the determination of the subsurface geological structure as well [39], [40], [41]. Over the last decades the TEM method has become increasingly popular for environmental and groundwater investigation [42], [43]. TEM is a method highly sensitive to low-resistivity layers, successfully applied in the past by other researchers, especially for mapping coastal aquifers and seawater intrusion [44], [45].

A total of 127 TEM soundings (Fig.1) were acquired across the entire deltaic valley. The TEM measurements were acquired with ABEM's WalkTEM instrument. The field configuration for data acquisition was comprised of a 40 m x 40 m square transmitter loop with two in-loop antennas, the RC-200 (10 m x 10 m) and the RC-5 (0.5 m x 0.5 m) each with different receiver areas. In Fig. 2 the described configuration array deployed in the field is presented.

Fig. 2.

The set-up of in-loop field configuration (left) and a sounding location during data acquisition.

The processing of the TEM soundings and their inversion were carried out with ViewTEM software leading to their corresponding multilayer resistivity models. Each multi-layer resistivity model was transformed into a few-layer resistivity model. An example of this procedure is presented in Fig. 3, where the ‘Resistivity– Elevation’ chart is illustrated, presenting the same TEM sounding in both ways; as a multi-layer resistivity model and as a four-layer geophysical model. The layers’ boundaries and resistivity values are given in Table 1.

Fig. 3.

The resistivities calculated for the same TEM sounding as a multi-layer model (blue color) and as a four-layer model (red color).

Table 1.

Geophysical layers extracted from the four-layer resistivity model of the above TEM sounding.

Geophysical layer	Z-Top (m)	Z-Base (m)	ρ (Ohm.m)	Lithology
1	2.0	−14.0	34.0	Alluvial deposits
2	−14.0	−35.0	4.2	Clay formation
3	−35.0	−62.0	28.0	Transgressive deposits
4	−62.0		7.5	Blue Schists

The geological interpretation of the geophysical data was achieved calibrated with available drilling data (Fig. 1) from [27]. Beyond that, the interpretation of the TEM soundings, was also cross-checked based on the published results of the geoelectrical method carried out by the authors [23], [24], [25], [26] but also with the recently published results by [28] regarding the investigation of the deeper geological and tectonic structure of the deltaic plain by gravity measurements.

The surface lithology in the deltaic plain that is structured by alluvial deposits or sand dunes (at the coastal area) (Fig. 1) have been investigated with resistivity values ranging between 17 and 45 Ohm.m and 66–120 Ohm.m respectively. The underlaying lithological layer, observed throughout the study area, corresponds to an impermeable clay formation, with resistivity of 1–7 Ohm.m. These two lithological units comprise the deltaic deposits. The third lithological unit, investigated by almost all TEM soundings (114/127 soundings), is consisted mainly of sands and gravels with small amount of clay and represents the transgressive deposits of the last interglacial period and is characterized by resistivity values of 10–30 Ohm.m. In many TEM soundings, an additional layer was also investigated, corresponding either to the Neogene formations that appear west and northwest of the deltaic plain, or the roof of the alpine basement (Blue schists or Crystalline Limestones).

Satellite image processing for surficial investigation

A very high resolution WordlView-2 image acquired on 27/07/2013 has been processed. The 8-band multispectral imagery was at 2 m resolution and the panchromatic imagery was at 50 cm resolution. The high spatial resolution enables the discrimination of fine details and the high spectral resolution provides detailed information on such diverse areas. The processing of WorldView-2 satellite images was performed with the ERDAS Imagine 2014 software, as well as the calculation of ten spectral indices (Table 2).

Table 2.

The Spectral Indices calculated for the study area.

Spectral Index	Formula	Reference
Normalized Difference Vegetation Index	$\text{NDVI}={\displaystyle \frac{\left(\text{NIR}-\text{RED}\right)}{\left(\text{NIR}+\text{RED}\right)}}$	[46]
Iron Oxide	$\text{IRON}\text{OXIDE}={\displaystyle \frac{\text{RED}}{\text{BLUE}}}$	[47]
Improved Modified Chlorophyll Absorption Rate Index	$\text{MCARI}2={\displaystyle \frac{1.5\left[2.5\left({\rho }_{800}-{\rho }_{670}\right)-1.3\left({\rho }_{800}-{\rho }_{550}\right)\right]}{\sqrt{{\left(2*{\rho }_{800}+1\right)}^2-\left(6*{\rho }_{800}-5*\sqrt{{\rho }_{670}}\right)-0.5}}}$	[48]
Modified Soil Adjusted Vegetation Index 2	$\text{MSAVI}2={\displaystyle \frac{2*\text{NIR}+1-\sqrt{{\left(2*\text{NIR}+1\right)}^2-8\left(\text{NIR}-\text{Red}\right)}}{2}}$	[49]
Modified Red Edge Simple Ratio Index	$\text{MRESR}={\displaystyle \frac{{\rho }_{750}-{\rho }_{445}}{{\rho }_{705}-{\rho }_{445}}}$	[50]
WorldView Soil Index	$\text{WV}-\text{SI}={\displaystyle \frac{\left(\text{Green}-\text{Yellow}\right)}{\left(\text{Green}+\text{Yellow}\right)}}$	[51]
WorldView Water Index	$\text{WV}-\text{WI}={\displaystyle \frac{\left(\text{Coastal}-\text{NIR}2\right)}{\left(\text{Coastal}+\text{NIR}2\right)}}$
WorldView New Iron Index	$\text{WV}-\text{II}={\displaystyle \frac{\left(\text{Green}*\text{Yellow}\right)}{\left(\text{Blue}*1000\right)}}$
WorldView Built-Up Index	$\text{WV}-\text{BI}={\displaystyle \frac{\left(\text{Coastal}-\text{Red}\text{Edge}\right)}{\left(\text{Coastal}+\text{Red}\text{Edge}\right)}}$

The objective of the surface investigation was to identify the appropriate combination of spectral bands / spectral indices to highlight various biotic / abiotic characteristics of the delta. As a result, six false color composites of the area were produced combining spectral bands and additional five false color composites by combining several spectral indices. The spectral band and spectral indices combinations used in this study along with their benefits are given in Tables 3 and 4 respectively.

Table 3.

The benefits of the six false color composites of the area that were produced combining spectral bands.

Spectral band combination	Benefits
(7, 6, 5)	✓ conifers appear in brown-green shades ✓ grass and crops tend to appear in yellow and orange shades ✓ bare soils and rock outcrops show up in grayish shades
(8, 4, 1)	✓ quick recognition of land use and land cover patterns ✓ vegetation shows up in red, built-up areas appear blue with highly variable colors representing vegetation changes and structures
(3, 2, 1)	✓ water penetration and bathymetric information
(4, 8, 7)	✓ highlights the presence of sediment in the water
(8, 7, 6)	✓ vegetation in light colored yellows ✓ high degree of contrast between vegetation, water, and built-up features
(7, 3, 2)	✓ illustrate built-up features and roads ✓ high degree of contrast between shoreline and seawater

Table 4.

The benefits of the five false color composites of the area that were produced combining spectral indices.

Spectral indices combination	Benefits
(MSAVI2, WV-BI, WV-WI)	shoreline clearly distinguished from other bare soils and sand dunes
(MSR705, R31, WV-II)	highlights the drainage network due to the riparian vegetation
(WV-SI, IRON INDEX, MSR705)	water penetration
(WV-II, WV-BI, WV-WI)	high degree of contrast between crops and bare soils, rock outcrops and sands
(WV-VI, NDVI, MACARI2)	highlights water bodies and soil moisture

In Fig. 4 we can clearly distinguish the different land uses in this area of the delta where Pineios River exits Tempi valley. The combination used is spectral bands 8, 7, 6. The roads, railway, settlements, bare soils and rocky areas appear in blue and grayish shades, while yellow, orange and red colors represent different types of vegetation and crops. Particularly, we can note the contrast presented between the riparian vegetation and the surrounding area, an element that delineates the hydrographic network. The river appears dark blue and black.

Fig. 4.

False color composite by the combination of spectral bands 8, 7, 6. This combination highlights the drainage network due to the riparian vegetation (yellow) while built-up areas and bare soils appear blue.

The combination of spectral indices MSAVI2, WV-BI, and WV-WI is shown in Fig. 5. This is a very detailed imagery of the deltaic plain north of Stomio settlement. Vegetation appears in red color while built-up areas and roads in blue. The beach along with the sand dunes and the beach ridges with orientation NE-SW, are very well defined.

Fig. 5.

False color composite by the combination of spectral indices MSAVI2, WV-BI, and WV-WI. The beach is clearly distinguished from other bare soils. The beach ridges and sand dunes appear very detailed.

Results and discussion

Our main interest of the applied methodology and this paper is the deltaic deposits. Taking into consideration the elevation of each TEM station and its model, the absolute heights of the top and the base of the alluvial deposits were calculated individually, as well as the point thickness of the deposits. The same procedure was also followed for the clay formation.

The map in Fig. 6 shows the spatial distribution of the thickness for the alluvial deposits along the delta plain. The formation has an average thickness of 11.1 m, the mean value was calculated 9.5 and the standard deviation equal to 6.28. The minimum value is 1.3 m and the maximum value is 28 m. Based on the standard deviation and the mean value calculated for the alluvial deposits, we categorized their spatial distribution along the deltaic area in three zones/classes (yellow, orange and red). These 3 classes are the following: i) ≤ 4.86 m ii) 4.86 - 17.4 m iii) ≥ 17.4 m. The first (yellow areas) and third (red area) classes include values of thickness beyond a standard deviation from the mean value. The highest values of thickness (greater than 15 m) of the alluvial deposits are located in the areas west of Kouloura settlement and southwest of Palaiopyrgos settlement (red zone). Respectively, the smallest values of thickness (less than 5 m) can be observed north and southeast (yellow zone). The alluvial deposits covering the rest of the deltaic area have been identified with thickness ranging between 4.86 and 17.4 m (orange zone).

Fig. 6.

Map presenting the distribution of the thickness of the alluvial deposits in the investigation area. The classes’ limits were calculated based on the standard deviation and mean values of the formation.

In addition, a statistical analysis of these data was performed for the calculation of important characteristics, but also a short and clear presentation that will enhance the spatial presentation of the results. The Natural Neighbor spatial interpolation method was used to construct the raster data.

The Histogram Graph of Fig. 7 refers to the values calculated for the thickness of the Clay Formation throughout the study area. We observe that the formation has an average thickness of 31.8 m, the median was calculated 31 and the standard deviation equal to 9. The minimum value is 13 m and the maximum 52 m. Regarding the spatial distribution of the thickness of Clay Formation in the deltaic plain (Fig. 8), the highest values are observed throughout the central and eastern part of Pineios River deltaic plain. The formation exceeds 45 m of thickness in the area south and east of Kouloura settlement.

Fig. 7.

The results of the statistical analysis performed for the Clay Formation.

Fig. 8.

Map presenting the distribution of the thickness of the clay formations in the investigation area. The classes’ limits were calculated based on the standard deviation and mean values of the formation.

Based on the values of the standard deviation and the mean value for Clay Formation, TEM soundings were again classified in the following 3 classes of thickness: i) ≤ 22.8 m, ii) 22.8 - 40.8 m iii) ≥ 40.8 m. The map in Fig. 8 shows the result of this classification spatially distributed along the deltaic plain. The first (yellow areas) and third class (red areas) include values beyond a standard deviation from the mean value. Small values are limited mainly in the southern part of the deltaic plain, parallel to the slopes of Ossa Mt. but also in the western part of the deltaic plain, in the area north of Omolio settlement. A small appearance of values of the first class is found in the north, in Paliomana settlement. Values of the third class occupy the area of the present estuary and inland the settlement of Kouloura.

The ratio ‘Thickness of Clay Formation / Thickness of Alluvium Deposits’ was also calculated in order to compare the distribution of thickness of the two formations appearing in the study area. The minimum value of the ratio is 0.74 and the maximum 14.2. The average value of the ratio is 4.04, the median 3.12 and the standard deviation 2.71. The map in Fig. 9 presents the spatial distribution of the ‘Thickness of Clay Formation / Thickness of Alluvial Deposits’ ratio along the deltaic plain. We observe that the ratio gets its maximum values in the areas of the present and past estuary (north of Stomio settlement) as well as further north in the area of Paliomana settlement. We can also notice one sounding location with high ratio value north of Omolio settlement.

Fig. 9.

Distribution of the values of the ratio ‘Thickness of Clay Formation / Thickness of Alluvial Deposits’.

Conclusions

In general, the applied methodology provided valuable data for the subsurface structure of a deltaic plain along with important information regarding its surficial characteristics. The statistical analysis of the geophysical interpretation can provide even better understanding of the river delta structure and can contribute to the understanding of the physical-geographical characteristics of a river delta area.

Regarding the results and the interpretation of the geophysical survey, new information regarding the subsurface structure has been obtained. The average thickness of the Alluvial deposits’ formation has been estimated to 11 m and is hosting the shallow unconfined aquifer. These deposits were formed during the growth of Pineios River Delta that began 4–5.000 years BP. Regarding the Clay formation, which is an impermeable formation, its average thickness has been calculated equal to 32 m. Basically, it represents the pro-deltaic fine-grained deposits of the growing delta during the Upper Holocene. Finally, the transgressive deposits which are consisted mainly of sands and gravels with small amount of clay, host an aquifer, artesian in places, and overlays the Neogene formations which can be considered as the paleosurface during the last glacial period (sea level was 120–130 m below present sea level).

The statistical analysis of the thickness of the lithological units of the deltaic area and their ratio thickness, also revealed some important results. More specifically, the greater values of thickness of the Alluvial deposits are limited to the central part of the deltaic plain and can be associated with the frequent flooding phenomena that are being observed in this area. On the other hand, the smallest thickness of this formation is located i) in the area north in the deltaic plain, where many meanders and isolated parts of old riverbeds are present, ii) in the area of the present estuary and iii) in the old estuary area, north of the Stomio settlement. Regarding the clay formation, small thickness values mainly appear i) in the southern part of the deltaic plain near Ossa Mt.’s slopes, ii) in the western part in the deltaic plain north of the settlement of Omolio and iii) in the north in Paliomana settlement. On the other hand, the high values of clay thickness appear i) in the eastern part of the Pineios Delta, ii) at the present estuary area and iii) inland, around Kouloura settlement. This seems to be consistent with the interpretation of the marine acoustic survey conducted in the area [52], which has estimated the thickness of the deltaic deposits in the central part of the Pineios Delta up to-35 m.

The Clay Formation Thickness/Alluvial deposits Thickness ratio gets its maximum values in the areas where the fluvial processes dominate through the mainstream or the past tributaries. These locations are: i) the area of the present estuary, ii) the old estuaries (Stomio settlement) and iii) further north in the area with the appearance of the large number of meanders. Finally, one TEM sounding, remote from the areas just indicated, that appears with high ratio value (north of Omolio settlement) coincides with the location where Pineios River changes abruptly two times its direction of flow.

Regarding the surface investigation of the area, we came up with some conclusions, based on the results of the combinations of the spectral bands and indices that we applied. More specifically, we observed that the combination of spectral bands 8, 7, 6 can be used to clearly distinguish the land uses with the high degree of contrast between vegetation, water, and built-up features. The vegetation is colored yellow and orange while the bare soils, rocky areas and built-up features appear blue and gray, something that greatly outlines the hydrographic network of the area as a result of the intense difference of the riparian vegetation. This can be very useful in monitoring the whole basin area and detecting possible changes in the inflow, like stream piracy that can change the routing of water and sediment and may have an effect on landscape evolution. Moreover, the combination of spectral indices MSAVI2, WV-BI, WV-WI can be very helpful to geomorphological mapping of the deltaic plain, given the fact that many features such as the beach ridges, sand dunes and the part of the sandy beach appear very detailed with this combination of spectral indices.

CRediT authorship contribution statement

Georgia S. Mitsika: Methodology, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization. John D. Alexopoulos: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Supervision, Project administration. Emmanuel Vassilakis: Conceptualization, Methodology, Resources, Data curation, Writing – review & editing, Supervision. Spyridon Dilalos: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing. Serafim E. Poulos: Conceptualization, Methodology, Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability

• Data will be made available on request.