Next Article in Journal
LiDAR-Derived Relief Typology of Loess Patches (East Poland)
Next Article in Special Issue
Woody Plant Encroachment in a Seasonal Tropical Savanna: Lessons about Classifiers and Accuracy from UAV Images
Previous Article in Journal
Masked Graph Convolutional Network for Small Sample Classification of Hyperspectral Images
Previous Article in Special Issue
Mapping Areas Invaded by Pinus sp. from Geographic Object-Based Image Analysis (GEOBIA) Applied on RPAS (Drone) Color Images
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 15
Issue 7
10.3390/rs15071870
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Satellite, UAV, and Geophysical Data to Identify Surface and Subsurface Hydrodynamics of Geographically Isolated Wetlands: Understanding an Undervalued Ecosystem at the Atlantic Forest-Cerrado Interface of Brazil

by
Lucas Moreira Furlan
1,*,
Manuel Eduardo Ferreira
2,
César Augusto Moreira
1,
Paulo Guilherme de Alencar
1,
Matheus Felipe Stanfoca Casagrande
1 and
Vânia Rosolen
1
1
Department of Geology, Institute of Geosciences and Exact Sciences, São Paulo State University (UNESP), Av. 24A, 1515, Bela Vista, Rio Claro 13506-900, Brazil
2
LAPIG—Image Processing and GIS Laboratory, Institute of Socio-Environmental Studies, Samambaia Campus, Federal University of Goiás, Av. Esperança s/n, Goiânia 74690-900, Brazil
*
Author to whom correspondence should be addressed.
Remote Sens. 2023, 15(7), 1870; https://doi.org/10.3390/rs15071870
Submission received: 9 February 2023 / Revised: 19 March 2023 / Accepted: 28 March 2023 / Published: 31 March 2023
(This article belongs to the Special Issue Structure and Trend Monitoring of Forest Vegetation and Savanna Based on UAS Platform)
Browse Figures
Versions Notes

Abstract

:
In two small and isolated wetlands located at the interface of the Atlantic Forest and Brazilian savanna (Cerrado) in São Paulo State, Brazil, we employed a pixel-based supervised classification approach using a combination of panchromatic and multispectral bands obtained from Landsat 2, 5, 7, and CBERS-04A satellites (ranging from 80 to 2 m/pixel). In addition, we acquired DJI Phantom 4 Pro UAV-RGB images in twelve different periods with a resolution of +5 cm/pixel. Furthermore, we utilized 2D and 3D Electrical Resistivity Tomography (ERT) to obtain data on the surroundings and center of the wetlands. Finally, we conducted a climatological data analysis. The results from the multisource data allowed us to classify the ecosystems as geographically isolated wetlands (GIWs), for which we documented a seasonal month-to-month (12 months) spatial variation of inundated area, vegetation pattern, soil water interaction, and a point of surface and deep-subsurface water interaction. These results are essential for high-accuracy characterization of small wetlands’ hydrodynamics and hydroperiods at the local scale. Our study contributes to optimizing GIWs understanding, monitoring, and reapplication of the methodology in other wetlands or small ecosystems.

Graphical Abstract

1. Introduction

The integration of remotely sensed high-resolution images obtained from different sensors and near-surface geophysics combines reliable information to reduce uncertainty related to the hydrological dynamics of natural wetlands. Moreover, investigating the connections among surface soil water, saprolite-bedrock infiltration, and aquifer recharge are necessary to interpret how the hydrology of wetlands is integrated with the landscapes. Yet, one of the biggest challenges concerning wetlands is to integrate images and data obtained from different sensors, which are necessary to provide a more consistent inference of Earth’s physical and chemical properties [1]. The development of a hydrological model demands a set of qualitative and quantitative data at broad spatial and temporal scales [2]. Hydrology determines wetlands types, processes, and their primary natural hydrological attributes, such as frequencies, durations, water depths, and flooding variations [3]. Conversely, anthropogenic interventions have been disrupting the wetlands’ water cycle, water supply, and water quality, indicating the immediate need for management at the local to global scale of water resources, including wetlands [4,5].
Wetlands play a crucial role in regulating the hydrology of catchments, serving as important components of local and global water cycles. They contribute to the storage of meteoric water, recharge of different aquifer levels, attenuation of floods, drainage flows, sediment retention, water purification, microclimate regulation, and supply of drinking water for humans and animals. Additionally, wetlands act as important carbon sinks, storing carbon and contributing to the mitigation of climate change impacts [6,7,8]. In this sense, remote sensing is a powerful tool for expanding capacity and agility in order to map their locations and associated attributes of water flows, water storage, and connectivity between other water bodies [9,10,11]. Remote sensing satellite data has been increasingly applied in wetlands studies using different sensors (optical, active microwave, optical, active microwave (SAR), passive microwave, thermal, and gravimetry) [12]. Satellite images show advantages to covering large spatial areas, although their spatiotemporal and spectral resolution are limiting factors to mapping of small ecosystems and to seasonal monitoring. Yet, remote sensing via satellite imagery is also a powerful tool to evaluate land use over time, as well as its impact on soil water dynamics and conservation [9,13].
More recently, unmanned aerial vehicles (UAVs or drones [14]) are advanced technological tools used for providing appropriate scale data in both time and space for catchment management (i.e., soil moisture) and they provide solutions to water problems related to quantity and quality [15]. UAVs are a supplementary tool to satellite systems for water assessment and monitoring because they provide a very-high-resolution (+5 cm/pixel) image acquisition and also have broader temporal coverages. Additionally, UAVs have the capacity to be flown in small, enclosed areas characterized by high seasonal dynamics [16,17]. Thus, using UAVs surveys can be a more cost-effective and detailed alternative compared with high-resolution satellite imagery. Although satellite imagery has recently increased in spatial resolution and temporal coverage, it still lacks the flexibility that UAVs surveys offer.
Considering the hydrology and anthropogenic pressure surrounding small and isolated wetlands [18,19], the images generated at pre-planned and regular intervals provide a robust time series of spatial variability and characteristics of flooding, vegetation, and land use. Through the digital elevation models (DEMs) and high-resolution orthophotomosaics (orthomosaics) obtained from UAV-digital aerial photogrammetry, the local hydrogeomorphological features typical of wetlands systems and the acquisition of eco-physiological information at microscale can be carried out with accuracy measured in centimeters. Such data resolutions are equivalent to products generated by terrestrial/airborne light detection and ranging (LiDAR) [20,21,22,23]. Additionally, an important advance in UAV-photogrammetry allows for the capture of detailed surface 3D information, using the structure-from-motion-multi-view-stereo (SfM-MVS) method [24,25,26].
If the hydrological status of small, isolated wetlands can be assessed by remote sensing, the development of a consistent model involving the interaction between surface water stored from rain, the water stored in the regolith, and the groundwater stored in the aquifer requires a combination of multiple sources of data (e.g., obtained by imaging). Reliable and high-quality images from shallow subsurface can be obtained by applying near-surface geophysics inversion data [1]. The Electrical Resistivity Tomography (ERT) method is suitable for hydrogeological studies since the high variability in electrical properties related to different materials and levels of wetness or water saturation in heterogeneous matrices of regolith provides subsurface imaging at the intermediate scales of hillslopes and small catchments [27,28]. A fundamental challenge is how to combine geoinformation from remote sensing and geophysical measurements into one consistent model representative of the Earth Critical Zone [29], which requires spatially distributed data of the surface and subsurface [30]. In the region, small, isolated wetlands have been destroyed since the 1970s by the intensification of land use by urban, mining, and agricultural industries. These wetlands are highly sensitive to land use [31,32,33] and climate changes [34,35], both altering the rate between precipitation, evapotranspiration, and hydroperiods [36,37].
The aim of this study is to propose a hydrodynamic model of small, isolated wetlands, based on the physical imaging response from multisource remote sensing technologies (satellites, UAV, and ERT). Regionally, there has been an increasing concern about the quality and availability of water for a portion of São Paulo State (Brazil), located at the interface between the Atlantic Forest and Cerrado (Brazilian savannah) biomes [38]. Multiple factors, such as the water crisis caused by prolonged drought, high population growth, water-dependent economic activities, and intensive conversion of land use toward sugarcane agriculture for biofuel production make this region one of the most critical regarding water resources [39,40].

2. Materials and Methods

2.1. Study Sites

The study sites are located in the Paulista Peripheral Depression (PPD), in the central-western of São Paulo State, Brazil. The PPD is a geological–geomorphological inland compartment between the basalt–sandstone cuesta and the Atlantic crystalline plateau [41] (Figure 1). The region has a subtropical climate (Cwa type, according to Koeppen [42]), with an average annual temperature of 21.6 °C and an average annual rainfall of 1366 mm [43]. Regarding the seasonal rainfall distribution, the dry season is from April to September, and the wet season is from October to March. The area is a transition zone between two Brazilian biomes, with different types of native vegetation cover: the Atlantic Forest and the Cerrado (Brazilian savanna). The climate is less humid than the domain of the Atlantic Forest itself due to the Atlantic Plateau (Serra do Mar Mountain), which prevents the advance of wet masses for the Paulista Peripheral Depression (Figure 1).
Digital elevation data 1:50,000 for the Corumbataí River Basin (Figure 1) was acquired free of charge from the São Paulo State database (https://www.infraestruturameioambiente.sp.gov.br, accessed on 7 February 2023). The studied upland wetlands are in the Corumbataí River Basin (CRB), where 24 wetlands were recently systematically identified [45,46]. Hydrogeomorphological features were used to identify small, isolated wetlands, which are represented by circular or oval shapes, depressed areas surrounded by uplands, absence of visible surface channels connecting to rivers, and vegetation (grasses) typical of hydromorphic soils.
Two wetlands were selected (as it follows, (i) and (ii)) to study the water regime (hydroperiod), water volume, and anthropogenic pressure based on UAV data and the interaction between surface water and groundwater using ERT. The wetlands are within a distance of 1.5 km, and both were chosen for this study, not only due to their proximity, but also to represent ecosystems in the same biome, geological background, and climate conditions. However, their different hydrological patterns make them an ideal comparison for studying hydrological variations at the Paulista Peripheral Depression.
(i)
Wetland A: latitude: 22°25′41″S; longitude: 47°37′52″W; altitude: 610 m.
(ii)
Wetland B: latitude: 22°26′24″S; longitude: 47°37′8″W; altitude: 610 m.

2.2. Satellite Images Acquisition and Bands Composition

The imaging methodology employed in this study utilizes a combination of satellite, UAV, 2D, and 3D ERT geophysical technologies (as shown in Figure 2) to enable a comprehensive and multiscale approach. This approach allows for the identification, classification, delineation, and acquisition of both qualitative and quantitative data, thereby facilitating a thorough understanding of the wetland ecosystems under study.
To locate the wetlands within the catchment and characterize their hydroperiod, a combination of satellite data from various sensors and bands of Landsat (USGS-NASA) and CBERS-04A (China-Brazil Earth Resources Satellite) were acquired from 1975 to 2020. The images were obtained free of charge from the National Institute for Space Research [47] and the USGS Earth Explorer platform [48]. The objective was to obtain a high-quality historical series of images with the best available resolutions for the selected region, spanning each decade. Datasets are provided with geometric corrections from control points and refined with digital elevation models.
The composition of raster bands (Composite Bands) (Table 1) was performed on a GIS platform (ArcGIS Pro 2.8.2). In landscapes with uneven surfaces, low concentrations of water in suspended sediments strongly absorb near-infrared light (NIR), resulting in a greater contrast between green vegetation and water [49]. The shortwave infrared electromagnetic region (SWIR) is highly sensitive to soil and vegetation moisture, making it useful for distinguishing wet vegetation from dry [50,51]. To effectively enhance the differences between wetlands and their surroundings, the band compositions were chosen based on the best available resolution with a preference for NIR and SWIR [13,52]. This choice enabled the production of false-color compositions (FCC) that efficiently highlighted the disparities between wetlands and their surrounding areas.
The increase of spatial resolution was made using the tool GIS Create Pan-sharpened Raster Dataset, applied to 2000 and 2020 datasets, which makes a fusion of bands adding to the initial composition in the spatial resolution of the Panchromatic Band [53].
The Landsat series band’s composition of near-infrared (NIR), short-wave infrared (SWIR 1), and red (R) generate a composition with clear land-water boundaries and highlights subtle details not apparent in the visible bands alone. Inland water bodies are precisely recognized when more infrared bands are applied. Based on the band’s composition, pixel-based supervised classification was carried out in two polygons, referring to wetland A and B. The low spatial resolution of the Landsat series data was resolved by using the pixel-based method as suggested by [54].
Supervised classification was performed through the non-parametric machine learning classifier Support Vector Machine and aimed to delimit exposed soil, tree vegetation, agriculture, and water/wetland. From the classifications, wetland area-polygons were calculated in each dataset, aiming to map the expansion and contraction of borders along the decades.

2.3. UAV Data Acquisition and Processing

The unmanned aerial vehicle (UAV) used was the DJI’s Phantom 4 Pro with dual GPS/Glonass receiver (dji.com) with a coupled sensor DJI 1″ CMOS Effective with 20 megapixels, which covers red-green-blue (RGB) bands [55]. The ground control station consists of a radio connection control connected to a smartphone that, from a DJI application, allows for the monitoring of flight, battery levels, wind conditions, and UAV speed. The UAV’s internal GPS allows for flights to occur autonomously from a pre-programmed flight plan.
Flight plans were created using PIX4D software [56] in which it is possible to set up different flight variables, such as height (120 m for A and B), images capture interval, number of images (250 images for A and 140 images for B), direction, camera angle (90°), and flight strips. The setting of the flight strips is an important parameter, as it influences the percentage of images overlap, and thus, the number of similar points found in different images (key point matches), which in turn, directly affects the quality and accuracy of processing results [57]. Thus, the flight strips were elaborated in order to obtain a lateral and longitudinal overlap greater than 80% for the center of the wetlands [58].
UAV-flights were carried out over a year (August 2019 to July 2020) for two wetlands. They were always taken in the last week of each month, with slight variations of a few days between the images due to inadequate weather conditions for the flight’s performance (i.e., rainy or windy days). In total, 12 orthomosaics (12 months) and 1 digital elevation model (October 2019) were generated for each wetland.
Image processing was based on SfM-MVS methodology (Structure from Motion and MultiView Stereo), which consists of building a dense point cloud from a series of overlapping 2D images obtained from a UAV (SfM) and building a 3D mesh from vertices and polygons (MVS) [59]. For this paper, Agisoft Metashape software [60] based on SfM-MVS was used. We performed planialtimectric correction using 5 ground control points (GCPs) and Differential GPS (DGPS) Trimble GEOXT equipment [61]. The processing steps in the software are photo alignment; creation of the dense point cloud; build of the polygon and texture mesh and; creation of the digital elevation models and orthomosaics. The processing report automatically generated by Agisoft Metashape indicated a ground resolution of 3.44 cm/pixel (October 2019) and 3.48 cm/pixel (February 2020), after GCPs’ correction.
To obtain accurate maps that reflect the spatial distributions of soil moisture, water levels, and vegetation structures within a topographic depression of both wetlands, we based our approach on previous wetland compartmentation at the Cerrado Biome made by Sales et al. [62]. This approach involved analyzing spectral compartments based on RGB pixel values of a True Color Composition (TCC) and generating an unsupervised classification to create polygons, which were named as compartments [62]. By identifying patterns or gradients in the spatial organization of wetland areas, this technique allowed for a more comprehensive understanding of the wetland ecosystem. Following this step, additional photogrammetric processing was conducted using the Agisoft Metashape [60] platform to evaluate and measure the biotic and abiotic parameters of the wetland system, considering the complex spatial variability of the ecosystem.

2.4. Water Balance for Hydrodynamic Analysis

Surplus and deficit of atmospheric humidity in the dry and wet seasons were calculated, aiming to validate the spectral signatures used to assess hydroperiod and calculate the wetland area in each period analyzed, which directly reflects the amount of water stored. The rates of precipitation and evapotranspiration data are fundamental to hydrological studies. The climatological data were obtained from a meteorological station (CEAPLA—São Paulo State University, coordinates 22°23′3″S; 47°32′45″W) [63]) for the same months of image acquisition, from August 2019 to July 2020. Water balance data is based on Equation (1) [64]:
W B = P − E T
where: WB is the water balance [mm] (positive or negative), P the precipitation [mm], and ET the reference evapotranspiration [mm].

2.5. ERT Geophysical Data Acquisition and Processing

Electrical Resistivity Tomography technique (ERT) is a suitable tool for shallow subsurface acquisitions and analysis, i.e., dozens of meters deep. The physical parameter of electrical resistivity obtained is based on the potential electrical fields generated by the transmission of an artificially induced electrical current through the soil [65,66]. The variability of an electric current propagation in a subsurface is due to distinct physical-chemical properties and moisture content of the geological material. The laterality of electrical resistivity data in the subsurface is obtained by interpolation of the discrete spatial data generating bidimensional sections that can be interpolated to create a tridimensional block of the subsurface.
It is important to consider that the preliminary field data is known as apparent resistivity because they configure weighted arithmetic means due to the electrical current flow through different geological materials in the subsurface [67]. The equipment used was the Swedish Terrameter LS from ABEM, which allows for automatic acquisitions based on previous configuration [68]. Its specifications are the following: 84 channels, 250 W, a maximum current of 2.5 A, and 1 mV resolution.
The data acquisition occurred during the wet and dry seasons at wetland A. The DC resistivity method comprehended 27 lines of 200 m each (5400 linear meters of ERT acquisition), installed as displayed at Figure 2, contemplating the center and the surroundings of wetland A. For the acquisition, the Schlumberger array was employed, which involves placing four electrodes in a straight line with equal spacing, with the outer electrodes used for current injection and the inner electrodes used for voltage measurement [28,69,70].
Data processing was performed using Res2Dinv software [53] with the generation of 2D inversion models. The interpolation and data inversion are based on the mathematical model of Ordinary Least Squares (OLS), which is responsible for the smoothing of extreme values and reducing the discrepancy between measured and modeled electrical resistivity values. To evaluate the quality of the final model, the Root-Mean-Squared (RMS) parameter quantifies the correspondence between the calculated inversion model and the field data, besides the presence of extreme values [71,72].
The 3D block model was processed in the Oasis Montaj platform [73] (Seequent, Christchurch, New Zealand). Each numerical product of the 2D inversion model was inserted into a single worksheet. The integration of the 27 lines generated a 3D model that can be visualized as slices, according to the chosen depth.

3. Results

3.1. Satellite Products

In the images obtained by the bands’ composition (Figure 3), the categories of the Land Use Land Cover (LULC) were grouped via the pixel contrast among them. Soil-vegetation type and condition vary in hues (browns, greens, and oranges in Landsat series; greens and purples in CBERS-04A) and tone. Distinct levels of soil moisture are strongly related to the dark color scale in satellite images due to the infrared absorption capabilities of water.
The pixel-based supervised classification was based on the high contrast of the different elements of the landscape captured by satellite sensors. LULC classes were extracted: (1) Soil (exposed arable land in preparation, roads, clay mining, or exposed soil for other land uses), (2) Tree vegetation (native or planted forest), (3) Agriculture (mainly sugarcane crops), and (4) Water (inundated areas). The Water class was used to define the limits of the wetlands surrounded by sugar cane. For each of the satellite datasets, LULC classification was made in two polygons-cut representing the wetlands A and B and their near-surroundings. Wetland A has permanent flooding over the time series and, therefore, can be effectively analyzed over the entire period 1975–2020 (45 years). In 1975, the data has low reliability due to the low spatial resolution of the images available (80 m), which could reflect a low accuracy of the results for this period.
The green areas around the edges and within Wetlands A and B are vegetation. In Wetland A, there are small green areas consisting of both endemic and exotic grass species (such as Brachiaria) that were introduced as forage plants in the former pasture. The vegetation in these areas has taken root in sediment deposits resulting from farm soil runoff and currently helps to retain soil particles detached from slopes by erosion. In Wetland B, on the other hand, the vegetation in the interior is non-resistant to flooding while woody trees have developed at the edges due to the drier conditions. In this sense, satellite images do not assist in differentiating between hydromorphic-tolerant and non-tolerant vegetation.
The size of wetland A reduced from 175,909 m2 in 1984 to 149,750 m2 in 2020, corresponding to 14.87% of the total flooded area (Figure 3). The graphs in Image 3 reveal discrepancies between the flooded areas of Wetlands A and B, with the two wetlands not following the same trends over the years. The calculation of the flooded area for the time series of Wetland B, based on surface water, was inadequate due to its much smaller size and significant seasonal hydro-variation, making it difficult to compare the areas based on low spatial resolution data. Additionally, in Wetland B, flooding greatly varies between wet and dry seasons, which further complicates the remote sensing capture of flooded areas during periods when the water level is low, or when the wetland is almost completely dry. Satellite imagery has proven to be a useful tool for assessing the spatial variations of hydrological patterns in small, geographically isolated wetlands that are highly disturbed by intensive land use, with the exception of cases like wetland B where the size is not representative in satellite datasets with low resolution. This limitation arises from the spatial resolution, with a minimum detectable size of wetlands in an image being 30 m per pixel, as demonstrated by the Sentinel-5 1985 dataset. This constraint can limit the detection of small wetlands, but other resources, such as machine learning, deep learning-based methods, and very-high-resolution images obtained through unmanned aerial vehicles (UAVs), can be used to overcome this issue. It is necessary to use these resources to properly identify these small ecosystems in a reliable manner over a historical series [74,75].

3.2. UAV Very-High-Resolution Orthomosaics and Digital Elevation Models

The UAV-orthomosaics (August 2019 to July 2020) of the two wetlands are displayed in Figure 4. The images demonstrate the great dynamism of the ecosystem over a year, mainly represented by two characteristics: the great seasonality of flooding and the quick change of LULC.
The orthomosaics depict the significant impact of internal flooding on the wetlands, which varies seasonally throughout the year and is largely influenced by climatic conditions and LULC patterns and is especially influenced by the crop’s management. Additionally, the orthomosaics provide pixel-based evidence that both wetlands experience the lowest water levels in October 2019 and the highest flooding in February 2020. These two months were selected for in-depth analysis to extract more information on the hydroperiods and surface hydrodynamics. The lower flooding is suggested by the greater presence of vegetation during the month, while the higher flooding is suggested by the greater presence of water.
Understanding wetlands hydrology is one step towards ecological restoration projects [76]. Despite their proximity of only 1500 m and similar geological and climatic conditions, wetlands A and B exhibit distinctive hydrological patterns. They have the border expansion during and following the period of large precipitation surplus and retract during the dry period in common. During the dry period, wetland A still waterlogged while wetland B diminished, although it did retain a small volume of water in its center. In both areas, soil moisture and water are closely related to the local terrain topography. The digital elevation models indicate that flooded areas were developed in interfluves positions of hills where shallow topographic depressions are surrounded by upland well-drained soils (Figure 5). Additionally, the wetlands have a lack of well-defined surface water connection to other wetlands or water bodies.
(i)
Dry-season images of wetland A. Dry and wet soil contrasts; Zoom-in, a water storage portion and native grasses tolerant to flooding;
(ii)
Soil-fissures observed in the border of wetland A; Zoom-in, where sediment deposition inside the wetland is observed (silting process).
Photogrammetric calculations were performed in order to document seasonal changes. The hydrological pattern of wetlands was determined by extracting pixel values from a true color composition that represented wetness characteristics (measured on RGB scale of darkness) and vegetation cover density made by a semi-automatic classification. The results of photogrammetry were used to identify different zones (represented as polygons) in Figure 6, which provided quantitative data (as shown in Table 2) for integrated analysis.
The imagery acquisition during the wet and dry periods shows differences between wetlands A and B reflecting their sizes, the amount of stored water, the flooding period and limits, and the vegetation pattern at the borders. In wetland A (Figure 6), during the dry period (October 2019), three compartments were distinguished: (A1dry) external border with lower soil wetness, sparse vegetation cover (grass and shrubs), high internal complexity in terms of microtopography, and small zones with distinct humidity; (A2dry) intermediate compartment with high soil wetness, and (A3dry) center, internal compartment covered by water and grass vegetation.
The soil presents a network of fissures created by swell-shrink resulting from continuous hydration-desiccation (Figure 5), organized in a dendritic pattern filled by water constituting paths of preferential flow [77]. These fissures develop in contact with compartment 1 (external border) and have no continuity toward compartment 3. During the rainy season (February 2020), two compartments were distinguished: (A1wet) has no water with an increase in vegetation density and (A2wet) entirely flooded. Thus, the flooding boundary is the contact between compartments A1wet and A2wet, different from the condition of the past decades, based on the satellite-supervised classifications (Figure 3), when the water probably occupied the entire extent of the wetland (i.e., all the compartments).
Comparing the areas of each compartment in the distinct hydroperiods, in October 2019 and February of 2020, compartment A1dry and A1wet have a negative difference of 14,943 m2, which resulted from wetland expansion during the rainy season. The total increase in flooding area inside the wetland was 109,032 m2 between October 2019 and February 2020—A3dry was completely flooded during the rainy months, comprising a spatial increase in the flooding. The total flooded area in the rainy season is 130,339 m2 (February 2020, compartment A2wet).
In wetland B, during the dry season (October 2019), two compartments are distinguishable: (B1dry) external border with low soil moisture associated with trees, shrubs, and grass vegetation; (B2dry) internal zones with high soil wetness and zones with low soil wetness, as well as low vegetation cover concentrated in zones with higher humidity. During the rainy season (February 2020), three compartments are distinguishable: (B1wet) the external border showing high similarities with that observed in the dry season; (B2wet) with low soil wetness covered by grasses, and (B3wet) flooded with great internal complexity related to increasing density and distribution of vegetation. The flooding boundary is compartment B2wet.
The protective vegetation (riparian forest) has an average of 45 m in length and a clear difference in relation to the border-vegetation of wetland A. Compartment B1wet/dry remains constant throughout the year and the existence of woody trees is possible because this portion does not flood at any time during the year. Although it is not flooding vegetation, it is a well-defined portion that takes advantage of the accumulated moisture inside the area.

3.3. Water-Balance Integrated to Orthomosaics Imagery Response

Hydrology of wetland is directly correlated, not only with surface hydrological aspects, but also with groundwater dynamics, with each influenced by precipitation (P), evapotranspiration (ET), soil water storage, and surface-runoff [78,79]. Several processes can occur in the hydrological cycle [80], so identifying the processes and their interconnections inside the wetland area are complex, especially concerning soil water interaction. However, to highlight seasonal impacts on wetlands dynamics and to predict possible changes in a LULC and climate change scenario, P is considered the only source of water (i.e., input), and ET is the only data quantified as water loss. Relatedly, in the area, no extraction of water stored in wetlands for irrigation or another use related to human activities has been reported or observed.
In general, if ET > P in a specific period (month), the period is classified as arid or dry (deficit or negative WB), whereas if P > ET, it is humid or wet (surplus or positive WB) [81]. Thus, the water balance calculations served as a basis for comparisons with the generated orthomosaics (Table 3).
For wetland A, it is possible to integrate water-balance and orthomosaics into two distinct periods:
  • August 2019–October 2019 and March 2019–July 2020: the decreased water level is associated with a retraction in the flooded borders (darkened portions) of orthomosaics, presenting ET > P. Although March shows positive values of water balance, the orthomosaics reveal a decrease in the amount of surface water in the wetlands;
  • November 2019–February 2020: the water level rose, and the edges of the wetland expanded. Another imaging factor that indirectly indicates the rise of water level is the greater reflection of sunlight perceptible in the images of the period. The highest WB values and the highest water reflection in orthomosaics occur in the same months (January, February 2020), with P < ET. February is the month with the greater cumulative positive WB.
    For wetland B, it is also possible to integrate water-balance and orthomosaics into two distinct periods:
  • August 2019–October 2019 and March 2019–July 2020: decrease in water level, reaching its minimum value in October 2019. The dominance of green color inside wetland area represents an increase of vegetation cover, ET > P;
  • November 2019–July 2020: increase in water level. However, from December 2019 to February 2020, the area reached a water level that remained relatively stable, not presenting a sharp rise as suggested by the remarkable increase between January and February 2020 precipitation. Between March and July 2020 (dry period), unlike wetland A, wetland B presents an increase in water levels, despite the negative values in the WB, indicating a unique pattern of the wetland. P < ET. Two phenomena can occur simultaneously and explain this increase in pixels classified as water. The first is that the vegetation that dominates the wetland during the dry season is not resistant to moisture, so it is gradually suppressed during the flooding months, making the water more visible. The second is that during periods of drought, the wetland experiences less disturbance, which causes greater sedimentation of suspended particles, making the color more similar to that used in the supervised classification of water [74].

3.4. ERT Subsurface Imaging: 2D Profiles and 3D Slices

Hydrogeological interpretation was conducted using ERT 2D inversion resistivity models’ 2D and 3D block-slices, which show the pattern of wetland A water flow into hydric soils (Figure 7 and Figure 8). The downward water flow is vertical, reaching the shallow groundwater, which is retained by a thick horizontal argillic layer (bedrock).
Of the 26 2D lines acquired in the field, two examples are shown in Figure 7: (w–x), which represents the acquisition and response from the hillslope to the wetland border, and (y–z), which was acquired from the center to the border in the flooding area. The inversion model (w–x) demonstrates the high electrical resistivity (>800 Ω·m) of the entire surface portion, forming an impermeable horizontal layer approximately 30 m thick. This horizontality response represents the geological stratigraphy substrate of the locality—the sedimentary Corumbataí Formation. The deep aquifer does not have direct interaction with surface water in the uplands of the wetland, and its recharge and maintenance occur horizontally from a depth of 30 m.
In the inversion model (y–z), it is possible to observe the response of the center of the wetland where electrical resistivity is low (<20 Ω·m). This region is where the interaction between surface water and groundwater occurs. After the creation of the 3D block, slices were cut at certain depths to observe how the electrical resistivity behaves, which provides an understanding of the water infiltration architecture. Slices were cut at a depth (representing the center of the wetland) of 5 m (615 m a.s.l), 15 m (605 m a.s.l), 25 m (595 m a.s.l), and 35 m (615 m a.s.l), as shown in Figure 8.
In the 5 m deep slice, the central region has medium values of electrical resistivity (40–90.9 Ω·m), indicating the interaction of surface water and soil water (or moisture). Analyzing the other slices (15 and 25 m), it is possible to observe that the central portion of the wetland becomes less and less resistive (20–5 Ω·m), indicating a preferential channel for water infiltration to the deeper aquifer. When reaching the slice of 35 m depth, there is a horizontal distribution of zones of low electrical resistivity, indicating the horizontal recharge of the aquifer, also represented in the 2D profile (w–x).

4. Discussion

Remote sensing played a crucial role in identifying the types of wetlands found in the Paulista Peripheral Depression. This technology provided valuable information on the wetlands’ hydrological patterns, temporal variations, and spatial changes resulting from land use. At a local scale, wetlands in the area are characterized as depressed areas surrounded by highlands with no visible surface connection to other bodies of water. These wetlands can be classified as geographically isolated wetlands (GIWs) (in the sense of Tiner [82]).
The image-response captured by both satellite and UAV datasets measures the level of inundation during dry and wet seasons and is directly related to the amount of water stored in the topographic depressions of the wetlands. The volume of stored water and expansion of the flooded area are dependent on precipitation levels. However, it is important to note that sediments deposited inside the wetlands from agricultural runoff can reduce the depth of the water basin, resulting in a higher proportion of water being located along the wetland border.
In Wetland A, the presence of shrubs and arboreal plants remained relatively constant between 1984 and 2000 but showed a significant decrease in endemic vegetation in 2010 and 2020, reflecting the faster expansion of cultivated areas and increased soil exposure during planting cycles. Additionally, the fine-detection of LULC changes was facilitated by the increased spatial resolution of the satellite data, as seen in Figure 3. The average distance between wetland A border and crops is 10 m, which lights up a red flag about the environmental sensitivity of this ecosystem. GIWs under upland land use intensification are often impacted by runoff, greater hydrologic variability, and loss of biodiversity [83]. In addition, wetlands are vulnerable to fire and climate changes, further affecting biodiversity and hydrology [84,85]. Recent studies carried out in this region indicated that the natural fire regime over Cerrado vegetation has been altered by an increased dry season induced by the unbalance between accumulated rainfall and dry/wet seasonality [86,87].
Satellite images using NIR, SWIR, and R bands were effective in mapping the border of flooding by discriminating water from dry soil over a historical time-series. The use of coarse spatial resolution satellite images was inadequate for discriminating changes in flooding and vegetation cover during dry and wet seasons, particularly in very small wetlands, such as Wetland B (68,692 m2). In contrast, images obtained from unmanned aerial vehicles (UAVs) have proven to be more effective in calibrating the wetlands’ hydrological patterns responses. They provide surface data and can be combined with climatic parameters, such as precipitation (P) and evapotranspiration (ET), to provide a more comprehensive understanding of wetland dynamics. The use of high spatial resolution images was critical in characterizing and comparing wetlands A and B. Despite being situated in the same lithology and climatic conditions, these wetlands exhibit different seasonal hydrological patterns. High spatial resolution images were necessary to accurately capture the nuances of the wetlands’ dynamics. UAV-high-resolution images enhanced gradient soil moisture and soil structure, microtopography, and vegetation cover closely related to the dynamics of the retraction-expansion of wetlands areas necessary to wetland monitoring. These visual tools can be used to trace hydrological connectivity within the landscape [88].
Recognizing the spatial and temporal ephemerality of wetlands provides critical information for monitoring and managing land use and land cover (LULC) to promote water conservation and ecological function. This is particularly important for wetlands located near agricultural areas, such as wetlands A and B, which are surrounded by a road used by agricultural machinery. The expansion of agriculture into these areas has led to the destruction of buffer zones and the invasion of the wetlands, making them more susceptible to degradation and environmental sensitivity. Therefore, effective management of LULC in these areas is crucial to ensure the preservation of wetlands and their ecological functions. The total flooded area for wetland A during the rainy season was estimated to be 130,339 m2 using UAV photogrammetry. However, the supervised classification of the CBERS-04A satellite image for the same period indicated a flooded area of 149,750 m2. This represents a difference of 19,411 m2, which corresponds to a percentage difference of 12.96%. This significant discrepancy can be attributed to the higher spatial resolution of UAV-orthomosaics when compared with satellite images. Therefore, the data provided by UAV-orthomosaics is the closest representation to reality.
The fine spatial resolution provided by UAV photogrammetry allowed for a more accurate assessment of the surface hydrodynamics of small ecosystems, and thus, a more reliable estimation of the flooded area. Additionally, remote sensing provided valuable data on wetland parameters that are difficult to obtain through field measurements. However, determining the precise wetland area and ecology using remote sensing data can be technically challenging. This is because water in wetlands can often be located beneath the land surface, boundaries may be undefined, especially when impacted by land use, and there may be spatial resolution limitations that hinder the detection of small wetlands. Despite these challenges, remote sensing remains a useful tool for providing insights into wetland ecology and dynamics [9,74]
The precise separation between wetland and upland, respecting the original boundaries in the catchment, was hampered by decades of human disturbance, visible with accuracy in the contact with wet and adjacent managed agricultural areas. Certainly, the current areas identified as wetlands are smaller than they were in the 1970s, the period when intensification and land conversion was initiated.
The border of the wetland is most vulnerable to erosion/sedimentation. Mineral sediments partially cover the organic matter accumulated on the surface of the soil. The orthomosaics’ internal-compartments show a black color typical of hydric soil, with the presence of organic matter preserved under more saturated conditions [89]. In climates with alternating dry and wet seasons, most fine-textured soils tend to develop cracks that close during the wet season [90]. If the water level in the wetland continues to drop as a response, for example, to the global climate changes (which alter the volume and precipitation regime), the wetland can lose some of its key hydroecological functions.
Regarding the climatological setting, the difference in the hydroperiods suggests that the stock of water is dependent not only on P, T, and ET, but also on physical characteristics of the wetland, i.e., the wetland’s soil attributes.
The infiltration architecture of the wetland can be visualized as a recharge funnel. The surface expression of the wetland behaves like a surface water storage region for meteoric water, exhibiting high dynamism and great seasonality. The water table line is parallel to the land surface and is restricted to depressed areas (wetlands) (Figure 7). Under this condition, the water table is rarely static [89] and the upward water movement occurs mainly through rainwater input, with the center of the wetland representing the greatest interaction point between surface water and groundwater. Figure 7 and Figure 8 show the importance of combining UAV and ERT images as key approaches to understanding the interactions between surface-subsurface water in wetlands. The monitoring of soil moisture and water volume on a fine-resolution spatial scale (UAV-orthomosaics) and the knowledge of subsurface water flow (ERT) provided the hydrological status of the wetland in the catchment.
The difference of flooding probably occurs because smaller GIWs have a substantially greater influence on quick-flow events relative to the larger wetlands [91] or because vegetation cover (trees and grass) mitigates excessive runoff controlling water table level [92]. Additionally, the intensive land use practices surrounding small wetlands can intensify erosion rates and bury upper hydric soil horizons. This can lead to unnatural fluctuations in water levels, alterations in vegetative communities, and physical disturbances during the dry season [93]. Finally, the permeability of aquifer, the wetness in the catchment, and the water stored in upland soil that supply wetlands control the base flow and water level [94,95]. At the individual scale, hydrologic processes of wetlands are controlled by the combination of climate (extrinsic factor) and hydrogeomorphology (intrinsic properties) [79,96]. The water levels in the studied GIWs could be regulated by the timescale of subsurface flow paths, which are slower and longer than surface flow paths [79].
According to Tiner [81], GIWs are characterized by surface isolation, but can be connected to the local hydrogeology in the subsurface. This study provides robust evidence through 2D and 3D ERT geophysical surveys that wetland A is indeed connected to the local hydrogeology, emphasizing the significance of understanding subsurface hydrogeology to fully comprehend the functioning and connectivity of GIWs. The importance of considering subsurface hydrogeology is critical when making decisions regarding wetland conservation and management.
The hydropatterns of the wetland are influenced by precipitation regime and the underlying geological structure (Corumbataí Formation), which both contribute to the accumulation of water. The water is mainly confined to topographic depressions where GIWs act as recharge zones and indirectly contribute to the flow rate of springs and rivers in the catchment. The relatively high elevation of the GIWs, as depicted by the SRTM DEM in Figure 1, is essential to catchment hydrology as they serve as recharge points and help maintain groundwater for discharge areas. Additionally, the satellite imagery of the wetlands reveals their widespread distribution and abundance, underscoring their critical role in maintaining the local and regional hydrological cycle.

5. Conclusions

Hydrodynamics study of wetlands located in the Paulista Peripheral Depression is extremely important, especially since they are located at the Atlantic Forest and Cerrado (savanna) interface, which is one of the most important Brazilian agricultural and environmental frontiers. Multisource remote sensing, geophysical, and climatological approaches in this paper are important techniques that can be used to begin the consolidation of data regarding these ecosystems. The classification of these areas as geographically isolated wetlands (GIWs) is an important contribution to the advancement of national legislation (federal, state, and municipal) to guide more appropriate management practices supported by good governance.
The application of remote sensing techniques in characterizing surface dynamics is crucial, and the use of multiple sources, such as satellite and UAV data, has provided significant contributions to understanding the hydrodynamics of wetlands. While satellite data has been useful in defining historical trends in wetland flooding, UAV data has allowed for more detailed seasonal analyses, revealing the complexities of wetland hydrology. However, it is important to note that there were observed variations of approximately 13% between the measurements obtained from satellite data and UAV datasets, which emphasizes the importance of using UAVs in monitoring and characterizing of small ecosystems. Moreover, the integration of climatological data and UAV imaging has further advanced our understanding of the hydropatterns of GIWs. In addition, the use of ERT has highlighted the interaction between surface water, soil water, and groundwater in wetland ecosystems, emphasizing their ecological and social importance. Overall, the integration of these techniques has provided a systematic and comprehensive approach to characterizing wetland hydrodynamics.
There are still gaps in the study of GIWs at the interface between the Cerrado and Atlantic Forest biomes, requiring ecosystem services and specialized systematic analyses of vegetation patterns using multi-sensor high-resolution data. It is necessary to identify and characterize the spectral signatures of the different LULC classes in this ecosystem and its surroundings. Although these gaps still exist, this study provides opportunities for methodological replication in wetland monitoring and classification, not only in the Paulista Peripheral Depression, but also in other regions of Brazil. Furthermore, since there is a lack of detailed studies on small ecosystems, this research is highly relevant for future investigations. By addressing these knowledge gaps and building upon the findings of this study, we can better understand and manage the ecological and social importance of GIWs ecosystems.

Author Contributions

Conceptualization, L.M.F., C.A.M. and V.R.; methodology, L.M.F., M.E.F., P.G.d.A., C.A.M. and V.R.; software, L.M.F., C.A.M. and P.G.d.A.; validation, M.E.F., C.A.M. and M.F.S.C.; formal analysis, M.E.F., C.A.M. and V.R.; investigation, L.M.F., C.A.M. and V.R.; resources, C.A.M. and V.R.; data curation, L.M.F. and M.F.S.C.; writing—original draft preparation, L.M.F. and V.R.; writing—review and editing, M.E.F., C.A.M., P.G.d.A. and M.F.S.C.; visualization, L.M.F., C.A.M. and V.R.; supervision, V.R.; project administration, V.R.; funding acquisition, V.R. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by the FAPESP—Fundação de Amparo à Pesquisa do Estado de São Paulo (Process n. 2020/03207-9).

Data Availability Statement

Data available in a publicly accessible repository [43,44,47,48]. All the UAV data are available on request from the corresponding author. The data are not publicly available due to the database size.

Acknowledgments

The authors are especially grateful to the Fundação de Amparo à Pesquisa do Estado de São Paulo (Process n. 2020/03207-9) for funding the project and to the Center for Environmental Analysis and Planning (CEAPLA, Unesp) for providing climatological data. We would also like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for financial support to the Ph.D. scholarships of L.M.F. and M.F.S.C.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Moorkamp, M.; Lelièvre, P.G.; Linde, N.; Khan, A. Integrated imaging of the Earth. Theory and Applications; American Geophysical Union: Hoboken, NJ, USA; John Wiley and Sons: Hoboken, NJ, USA, 2016. [Google Scholar]
  2. Mousa, Y.A.; Hasan, A.F.; Helmholz, P. Spatio-Temporal Analysis of Sawa Lake’s Physical Parameters between (1985–2020) and Drought Investigations Using Landsat Imageries. Remote Sens. 2022, 14, 1831. [Google Scholar] [CrossRef]
  3. Mitsch, W.J.; Gosselink, J.G. The value of wetlands: Importance of scale and landscape setting. Ecol. Econ. 2000, 35, 25–33. [Google Scholar] [CrossRef]
  4. Rebelo, L.M.; Finlayson, C.M.; Nagabhatla, N. Remote sensing and GIS for wetland inventory, mapping and change analysis. J. Environ. Manag. 2009, 90, 2144–2153. [Google Scholar] [CrossRef] [PubMed]
  5. Neto, S. Water governance in an urban age. Util. Policy 2016, 43, 32–41. [Google Scholar] [CrossRef]
  6. Euliss, N.H.; Mushet, D.M. Impacts of water development on aquatic macroinvertebrates, amphibians, and plants in wetlands of a semi-arid landscape. Aquat. Ecosyst. Health 2004, 7, 73–84. [Google Scholar] [CrossRef] [Green Version]
  7. Melton, J.R.; Wania, R.; Hodson, E.L.; Poulter, B.; Ringeval, B.; Spahni, R.; Kaplan, J.O. Present state of global wetland extent and wetland methane modelling: Conclusions from a model inter-comparison project (WETCHIMP). Biogeosciences 2013, 10, 753–788. [Google Scholar] [CrossRef] [Green Version]
  8. Russi, D.; Ten Brink, P.; Farmer, A.; Badura, T.; Coates, D.; Förster, J.; Kumar, R.; Davidson, N. The economics of ecosystems and biodiversity for water and wetlands. IEEP Lond. Bruss. 2013, 78, 118. [Google Scholar]
  9. Ozesmi, S.L.; Bauer, M.E. Satellite remote sensing of wetlands. Wetl. Ecol. Manag. 2002, 10, 381–402. [Google Scholar] [CrossRef]
  10. Guo, M.; Li, J.; Sheng, C.; Xu, J.; Wu, L. A Review of Wetland Remote Sensing. Sensors 2017, 17, 777. [Google Scholar] [CrossRef] [Green Version]
  11. Reif, M.; Frohn, R.C.; Lane, C.R.; Autrey, B. Mapping Isolated Wetlands in a Karst Landscape: GIS and Remote Sensing Methods. GIScience Remote Sens. 2009, 46, 187–211. [Google Scholar] [CrossRef]
  12. Kandus, P.; Minotti, P.G.; Morandeira, N.S.; Grimson, R.; Trilla, G.G.; González, E.B.; Martín, L.S.; Gayol, M.P. Remote sensing of wetlands in South America: Status and challenges. Int. J. Remote Sens. 2018, 39, 993–1016. [Google Scholar] [CrossRef]
  13. Hakimdavar, R.; Hubbard, A.; Policelli, F.; Pickens, A.; Hansen, M.; Fatoyinbo, T.; Lagomasino, D.; Pahlevan, N.; Unninayar, S.; Kavvada, A. Monitoring water-related ecosystems with earth observation data in support of Sustainable Development Goal (SDG) 6 reporting. Remote Sens. 2020, 12, 1634. [Google Scholar] [CrossRef]
  14. Chabot, D.; Hodgson, A.J.; Hodgson, J.C.; Anderson, K. ‘Drone’: Technically correct, popularly accepted, socially acceptable: Different fields use different terms, but by changing its title, this journal is advocating for the discontinuation of ‘unmanned’ and recognition of ‘drone’ as an umbrella term for all robotic vehicles. Drone Syst. Appl. 2022, 10, 399–405. [Google Scholar]
  15. Debell, L.; Anderson, K.; Brazier, R.; King, N.; Jones, L. Water resource management at catchment scales using lightweight UAVs: Current capabilities and future perspectives. J. Unmanned Veh. Syst. 2016, 4, 7–30. [Google Scholar] [CrossRef]
  16. Chamoso, P.; González-Briones, A.; Rivas, A.; De Mata, F.B.; Corchado, J.M. The Use of Drones in Spain: Towards a Platform for Controlling UAVs in Urban Environments. Sensors 2018, 18, 1416. [Google Scholar] [CrossRef] [Green Version]
  17. Ledź, S.; Ewertowski, M.; Piekarczyk, J. Applications of unmanned aerial vehicle (UAV) surveys and Structure from Motion photogrammetry in glacial and periglacial geomorphology. Geomorphology 2021, 378, 107620. [Google Scholar]
  18. Golden, H.E.; Creed, I.F.; Ali, G.; Basu, N.B.; Neff, B.P.; Rains, M.C.; McLaughlin, D.L.; Alexander, L.C.; Ameli, A.A.; Christensen, J.R.; et al. Integrating geographically isolated wetlands into land management decisions. Front. Ecol. Environ. 2017, 15, 319–327. [Google Scholar] [CrossRef] [PubMed]
  19. Rains, M.; Leibowitz, S.G.; Cohen, M.J.; Creed, I.F.; Golden, H.E.; Jawitz, J.W.; Kalla, P.; Lane, C.R.; Lang, M.W.; McLaughlin, D.L. Geographically isolated wetlands are part of the hydrological landscape. Hydrol. Process. 2016, 30, 153–160. [Google Scholar] [CrossRef]
  20. Li, N.; Zhou, D.; Duan, F.; Wang, S.; Cui, Y. Application of unmanned airship image system and processing techniques for identifying of fresh water wetlands at a community scale. In Proceedings of the 2010 18th International Conference on Geoinformatics, Beijing, China, 18–20 July 2010; IEEE: Piscataway, NJ, USA; pp. 1–5. [Google Scholar] [CrossRef]
  21. Ouédraogo, M.M.; Degré, A.; Debouche, C.; Lisein, J. The evaluation of unmanned aerial system-based photogrammetry and terrestrial laser scanning to generate DEMs of agricultural watersheds. Geomorphology 2014, 214, 339–355. [Google Scholar] [CrossRef]
  22. Ishiguro, S.; Yamano, H.; Oguma, H. Evaluation of DSMs generated from multi-temporal aerial photographs using emerging structure from motion–multi-view stereo technology. Geomorphology 2016, 268, 64–71. [Google Scholar] [CrossRef]
  23. Iizuka, K.; Itoh, M.; Shiodera, S.; Matsubara, T.; Dohar, M.; Watanabe, K. Advantages of unmanned aerial vehicle (UAV) photogrammetry for landscape analysis compared with satellite data: A case study of postmining sites in Indonesia. Cogent Geosci. 2018, 4, 1498180. [Google Scholar] [CrossRef]
  24. James, M.R.; Robson, S. Straightforward reconstruction of 3D surfaces and topography with a camera: Accuracy and geoscience application. J. Geophys. Res. Earth Surf. 2012, 117, 03017. [Google Scholar] [CrossRef] [Green Version]
  25. Westoby, M.; Brasington, J.; Glasser, N.F.; Hambrey, M.J.; Reynolds, J.M. ‘Structure-from-Motion’ photogrammetry: A low-cost, effective tool for geoscience applications. Geomorphology 2012, 179, 300–314. [Google Scholar] [CrossRef] [Green Version]
  26. Fonstad, M.A.; Dietrich, J.T.; Courville, B.C.; Jensen, J.L.; Carbonneau, P.E. Topographic structure from motion: A new development in photogrammetric measurement. Earth Surf. Process. Landf. 2013, 38, 421–430. [Google Scholar] [CrossRef] [Green Version]
  27. Rubin, Y.; Hubbard, S.J. Hydrogeophysics; Springer: Dordrecht, The Netherlands, 2006. [Google Scholar]
  28. Moreira, C.A.; Rosolen, V.; Furlan, L.M.; Bovi, R.C.; Masquelin, H. Hydraulic conductivity and geophysics (ERT) to assess the aquifer recharge capacity of an inland wetland in the Brazilian Savanna. Environ. Chall. 2021, 5, 100274. [Google Scholar] [CrossRef]
  29. Lin, H.; Hopmans, J.W.; Richter, D.D. Interdisciplinary Sciences in a Global Network of Critical Zone Observatories. Vadose Zone J. 2011, 10, 781–785. [Google Scholar] [CrossRef]
  30. Linde, N.; Doetsch, J. Joint Inversion in Hydrogeophysics and Near-Surface Geophysics. In Integrated Imaging of the Earth. Theory and Applications; Moorkamp, M., Lelièvre, P.G., Linde, N., Khan, A., Eds.; American Geophysical Union: Hoboken, NJ, USA, 2016; pp. 119–135. [Google Scholar]
  31. Papastergiadou, E.S.; Retalis, A.; Apostolakis, A.; Georgiadis, T. Environmental Monitoring of Spatio-temporal Changes Using Remote Sensing and GIS in a Mediterranean Wetland of Northern Greece. Water Resour. Manag. 2008, 22, 579–594. [Google Scholar] [CrossRef]
  32. Haidary, A.; Amiri, B.J.; Adamowski, J.; Fohrer, N.; Nakane, K. Assessing the Impacts of Four Land Use Types on the Water Quality of Wetlands in Japan. Water Resour. Manag. 2013, 27, 2217–2229. [Google Scholar] [CrossRef]
  33. Ricaurte, L.F.; Olaya-Rodríguez, M.H.; Cepeda-Valencia, J.; Lara, D.; Arroyave-Suárez, J.; Finlayson, C.M.; Palomo, I. Future impacts of drivers of change on wetland ecosystem services in Colombia. Glob. Environ. Chang. 2017, 44, 158–169. [Google Scholar] [CrossRef]
  34. Mitsch, W.J.; Nahlik, A.; Wolski, P.; Bernal, B.; Zhang, L.; Ramberg, L. Tropical wetlands: Seasonal hydrologic pulsing, carbon sequestration, and methane emissions. Wetl. Ecol. Manag. 2009, 18, 573–586. [Google Scholar] [CrossRef]
  35. Cowood, A.L.; Young, J.; Dowling, T.I.; Moore, C.L.; Muller, R.; MacKenzie, J.; Littleboy, M.; Nicholson, A.T. Assessing wetland climate change vulnerability for wetland management decision support using the hydrogeological landscape framework: Application in the Australian Capital Territory. Mar. Freshw. Res. 2010, 70, 225. [Google Scholar] [CrossRef]
  36. Krasnostein, A.L.; Oldham, C. Predicting wetland water storage. Water Resour. Res. 2004, 40. [Google Scholar] [CrossRef]
  37. Zhang, G.; Bai, J.; Xi, M.; Zhao, Q.; Lu, Q.; Jia, J. Soil quality assessment of coastal wetlands in the Yellow River Delta of China based on the minimum data set. Ecol. Indic. 2016, 66, 458–466. [Google Scholar] [CrossRef] [Green Version]
  38. Sano, E.E.; Rosa, R.; Brito, J.L.S.; Ferreira, L.G. Land cover mapping of the tropical savanna region in Brazil. Environ. Monit. Assess. 2010, 166, 113–124. [Google Scholar] [CrossRef] [PubMed]
  39. Martinelli, L.A.; Filoso, S. Expansion of Sugarcane Ethanol Production in Brazil: Environmental and Social Challenges. Ecol. Appl. 2008, 18, 885–898. [Google Scholar] [CrossRef] [Green Version]
  40. Ussami, K.A.; Guilhoto, J.J.M. Economic and water dependence among regions: The case of Alto Tiete, Sao Paulo State, Brazil. Economia 2018, 19, 350–376. [Google Scholar] [CrossRef]
  41. Almeida, F.F.M. Geological foundations of São Paulo relief. São Paulo: Geographic and Geological Institute. Geol. State São Paulo 1964, 41, 167–263. [Google Scholar]
  42. Koeppen, W. Climatologia: Con un Estudio de Los Climas de la Tierra. Fondo de Cultura Econômica: Ciudad de Mexico, Mexic, 1948; p. 479. [Google Scholar]
  43. Cepagri. Clima dos Municípios Paulistas. Available online: http://www.cpa.unicamp.br (accessed on 28 July 2020).
  44. Modelo Digital de Elevação. Estado de São Paulo. Available online: Infraestruturameioambiente.sp.gov.br (accessed on 15 June 2020).
  45. Junqueira, D.A. Principles of the Environmental Governance in Water Resource Management Focusing on Wetlands at Depressão Periférica Paulista. Ph.D. Thesis, São Paulo State Universtiy, São Paulo, Brazil, 2020. [Google Scholar]
  46. Furlan, L.M.; Moreira, C.A.; de Alencar, P.G.; Rosolen, V. Environmental monitoring and hydrological simulations of a natural wetland based on high-resolution unmanned aerial vehicle data (Paulista Peripheral Depression, Brazil). Environ. Chall. 2021, 4, 100146. [Google Scholar] [CrossRef]
  47. Instituto Nacional de Pesquisas Espaciais. Available online: http://www.dgi.inpe.br/catalogo/ (accessed on 17 June 2020).
  48. USGS Earth Explorer. Available online: https://earthexplorer.usgs.gov (accessed on 20 June 2020).
  49. Lang, M.W.; Purkis, S.; Klemas, V.V.; Tiner, R.W. 25 Promising Developments and Future Challenges for Remote Sensing of Wetlands; Taylor and Francis: Boca Raton, FL, USA, 2015. [Google Scholar]
  50. Jensen, J.R. Remote Sensing of the Environment: An Earth Resource Perspective 2/e; Pearson Education: New Delhi, India, 2009. [Google Scholar]
  51. Li, Z.; Guo, X. Non-photosynthetic vegetation biomass estimation in semiarid Canadian mixed grasslands using ground hyperspectral data, Landsat 8 OLI, and Sentinel-2 images. Int. J. Remote Sens. 2018, 39, 6893–6913. [Google Scholar] [CrossRef]
  52. Taddeo, S.; Dronova, I.; Depsky, N. Spectral vegetation indices of wetland greenness: Responses to vegetation structure, composition, and spatial distribution. Remote Sens. Environ. 2019, 234, 111467. [Google Scholar] [CrossRef]
  53. Xu, H.; Gao, P. Custom image processing capabilities in ARCGIS. The International Archives of the Photogrammetry. Remote Sens. Spat. Inf. Sci. 2008, 37, 263–266. [Google Scholar]
  54. Hu, T.; Liu, J.; Zheng, G.; Zhang, D.; Huang, K. Evaluation of historical and future wetland degradation using remote sensing imagery and land use modeling. Land Degrad. Dev. 2020, 31, 65–80. [Google Scholar] [CrossRef]
  55. DJI. Phantom 4PRO. Available online: https://www.dji.com/br/phantom-4-pro (accessed on 30 November 2022).
  56. PIX4D. Available online: https://www.pix4d.com/ (accessed on 30 November 2022).
  57. Dandois, J.P.; Olano, M.; Ellis, E.C. Optimal Altitude, Overlap, and Weather Conditions for Computer Vision UAV Estimates of Forest Structure. Remote Sens. 2015, 7, 13895–13920. [Google Scholar] [CrossRef] [Green Version]
  58. D’Oleire-Oltmanns, S.; Marzolff, I.; Peter, K.D.; Ries, J.B. Unmanned Aerial Vehicle (UAV) for Monitoring Soil Erosion in Morocco. Remote Sens. 2012, 4, 3390–3416. [Google Scholar] [CrossRef] [Green Version]
  59. Gomez, C.; Hayakawa, Y.; Obanawa, H. A study of Japanese landscapes using structure from motion derived DSMs and DEMs based on historical aerial photographs: New opportunities for vegetation monitoring and diachronic geomorphology. Geomorphology 2015, 242, 11–20. [Google Scholar] [CrossRef] [Green Version]
  60. Agisoft Metashape Professional Software, Version 1.6.3. Available online: https://www.agisoft.com (accessed on 30 November 2022).
  61. Collischonn, C.; de Lima, E.M.; Ruiz, L.F.C.; Araujo, T.D.; Matsuoka, M.T. Posicionamento GNSS: Comparação entre coordenadas oficiais de estações da RBMC e as obtidas por PPP e posicionamento relativo e ajustamento. In Simpósio Brasileiro de Sensoriamento Remoto—SBSR.; INPE: João Pessoa, Brazil, 2015. [Google Scholar]
  62. Sales, J.C.; Bueno, G.T.; Rosolen, V.; Ferreira, M.E.; Furlan, L.M. The structure of an earth-mound field of the Brazilian Savanna. Geomorphology 2021, 386, 107752. [Google Scholar] [CrossRef]
  63. CEAPLA. Available online: https://igce.rc.unesp.br/#!/ceapla (accessed on 15 October 2022).
  64. Fries, A.; Silva, K.; Pucha-Cofrep, F.; Oñate-Valdivieso, F.; Ochoa-Cueva, P. Water Balance and Soil Moisture Deficit of Different Vegetation Units under Semiarid Conditions in the Andes of Southern Ecuador. Climate 2020, 8, 30. [Google Scholar] [CrossRef] [Green Version]
  65. Kearey, P.; Brooks, M.; Hill, I. An introduction to geophysical exploration. In Tradução de Maria Cristina Moreira Coelho, 1st ed; Oficina de Textos: São Paulo, Brazil, 2002. (In Portuguese) [Google Scholar]
  66. Milson, J. Field Geophysics; John Wiley & Sons: Sussex, UK, 2003. [Google Scholar]
  67. Mussett, A.E.; Khan, M.A. Looking into the Earth: An Introduction to Geological Geophysics. Cambridge University Press: London, England, 2000. [Google Scholar]
  68. Terrameter, L.S. Instruction Manual; ABEM Instrument: Sundbyberg, Sweden, 2012. [Google Scholar]
  69. Casagrande, M.F.S.; Furlan, L.M.; Moreira, C.A.; Rosa, F.T.G.; Rosolen, V. Non-invasive methods in the identification of hydrological ecosystem services of a tropical isolated wetland (Brazilian study case). Environ. Chall. 2021, 5, 100233. [Google Scholar] [CrossRef]
  70. Rosa, F.T.G.; Moreira, C.A.; Rosolen, V.; Casagrande, M.; Bovi, R.C.; Furlan, L.M.; dos Santos, S.F. Detection of Aquifer Recharge Zones in Isolated Wetlands: Comparative Analysis Among Electrical Resistivity Tomography Arrays. Pure Appl. Geophys. 2022, 179, 1275–1294. [Google Scholar] [CrossRef]
  71. Geotomo Software. Geoelectrical Imaging 2-D & 3D; Geotomo: Penang, Malaysia, 2003; 129p. [Google Scholar]
  72. Chulès, J.P.; Delfiner, P. Geostatistics—Modeling Spatial Uncertainty, 2nd ed.; Wiley: Hoboken, NJ, USA, 2012. [Google Scholar]
  73. Oasis Montaj. Available online: https://www.seequent.com/ (accessed on 20 November 2022).
  74. Mahdavi, S.; Salehi, B.; Granger, J.; Amani, M.; Brisco, B.; Huang, W. Remote sensing for wetland classification: A comprehensive review. GIScience Remote Sens. 2018, 55, 623–658. [Google Scholar] [CrossRef]
  75. Peng, H.; Xia, H.; Shi, Q.; Chen, H.; Chu, N.; Liang, J.; Gao, Z. Monitoring spatial and temporal dynamics of wetland vegetation and their response to hydrological conditions in a large seasonal lake with time series Landsat data. Ecol. Indic. 2022, 142, 109283. [Google Scholar] [CrossRef]
  76. Walton, C.R.; Zak, D.; Audet, J.; Petersen, R.J.; Lange, J.; Oehmke, C.; Wichtmann, W.; Kreyling, J.; Grygoruk, M.; Jabłońska, E.; et al. Wetland buffer zones for nitrogen and phosphorus retention: Impacts of soil type, hydrology and vegetation. Sci. Total Environ. 2020, 727, 138709. [Google Scholar] [CrossRef] [PubMed]
  77. Jarvis, N.J.; Moeys, J.; Koestel, J.; Hollis, J.M.; Lin, H. Preferential Flow in a Pedological Perspective; Hydropedology; Academic Press: Waltham, MA, USA, 2012; pp. 75–120. [Google Scholar]
  78. Frei, S.; Lischeid, G.; Fleckenstein, J. Effects of micro-topography on surface–subsurface exchange and runoff generation in a virtual riparian wetland—A modeling study. Adv. Water Resour. 2010, 33, 1388–1401. [Google Scholar] [CrossRef]
  79. Lee, S.; McCarty, G.W.; Moglen, G.E.; Lang, M.W.; Jones, C.N.; Palmer, M.; Yeo, I.-Y.; Anderson, M.; Sadeghi, A.M.; Rabenhorst, M.C. Seasonal drivers of geographically isolated wetland hydrology in a low-gradient, Coastal Plain landscape. J. Hydrol. 2020, 583, 124608. [Google Scholar] [CrossRef]
  80. Kite, G. Land Surface Parameterizations of Gcms and Macroscale Hydrological Models. JAWRA J. Am. Water Resour. Assoc. 1998, 34, 1247–1254. [Google Scholar] [CrossRef]
  81. Coral, A.C.; Tommaselli, J.T.G.; Leal, A.C. Cálculo de balance hídrico usando modelamiento de datos espaciales: Estudio aplicado a la cuenca del río Buena Vista, Ecuador. Formação 2015, 1. [Google Scholar]
  82. Tiner, R.W. Geographically isolated wetlands of the United States. Wetlands 2003, 23, 494–516. [Google Scholar] [CrossRef]
  83. Guo, Y.; Liao, H.-L.; Boughton, E.H.; Martens-Habbena, W.; Qiu, J. Effects of land-use intensity, grazing and fire disturbances on soil bacterial and fungal communities in subtropical wetlands. Agric. Ecosyst. Environ. 2023, 345, 108314. [Google Scholar] [CrossRef]
  84. Boughton, E.H.; Quintana-Ascencio, P.F.; Bohlen, P.J.; Fauth, J.E.; Jenkins, D.G. Interactive effects of pasture management intensity, release from grazing and prescribed fire on forty subtropical wetland plant assemblages. J. Appl. Ecol. 2016, 53, 159–170. [Google Scholar] [CrossRef] [Green Version]
  85. Yan, W.; Wang, Y.; Chaudhary, P.; Ju, P.; Zhu, Q.; Kang, X.; Chen, H.; He, Y. Effects of climate change and human activities on net primary production of wetlands on the Zoige Plateau from 1990 to 2015. Glob. Ecol. Conserv. 2020, 35, e02052. [Google Scholar] [CrossRef]
  86. Conciani, D.E.; dos Santos, L.P.; Silva, T.S.F.; Durigan, G.; Alvarado, S.T. Human-climate interactions shape fire regimes in the Cerrado of São Paulo state, Brazil. J. Nat. Conserv. 2021, 61, 126006. [Google Scholar] [CrossRef]
  87. da Silva, G.J.; Berg, E.C.; Calijuri, M.L.; dos Santos, V.J.; Lorentz, J.F.; Alves, S.D.C. Aptitude of areas planned for sugarcane cultivation expansion in the state of São Paulo, Brazil: A study based on climate change effects. Agric. Ecosyst. Environ. 2021, 305, 107164. [Google Scholar] [CrossRef]
  88. Blake, W.H.; Kelly, C.; Wynants, M.; Patrick, A.; Lewin, S.; Lawson, J.; Nasolwa, E.; Page, A.; Nasseri, M.; Marks, C.; et al. Integrating land-water-people connectivity concepts across disciplines for co-design of soil erosion solutions. Land Degrad. Dev. 2021, 32, 3415–3430. [Google Scholar] [CrossRef]
  89. Vepraskas, M.; Lindbo, D.; Lin, H. Redoximorphic Features as Related to Soil Hydrology and Hydric Soils; Hydropedology; Academic Press: Waltham, MA, USA, 2012; pp. 143–172. [Google Scholar] [CrossRef]
  90. Espinar, J.L.; Clemente, L. The impact of vertic soil cracks on submerged macrophyte diaspore bank depth distribution in Mediterranean temporary wetlands. Aquat. Bot. 2007, 87, 325–328. [Google Scholar] [CrossRef] [Green Version]
  91. Evenson, G.R.; Golden, H.E.; Lane, C.R.; McLaughlin, D.L.; D'Amico, E. Depressional wetlands affect watershed hydrological, biogeochemical, and ecological functions. Ecol. Appl. 2018, 28, 953–966. [Google Scholar] [CrossRef] [PubMed]
  92. Euliss, N.H.; Mushet, D.M. Water-level fluctuation in wetlands as a function of landscape condition in the prairie pothole region. Wetlands 1996, 16, 587–593. [Google Scholar] [CrossRef]
  93. Trettin, C.C.; Kolka, R.K.; Marsh, A.S.; Bansal, S.; Lilleskov, E.A.; Megonigal, P.; Stelk, M.J.; Lockaby, G.; D’Amore, D.V.; Mac-Kenzie, R.A.; et al. Wetland and hydric soil. In Forest and Rangeland Soils of the United States Under Changing Conditions; Pouyat, R.V., Page-Dumroese, D.S., Patel-Weynand, T., Geiser, L.H., Eds.; Springer: Cham, Switzerland, 2020; pp. 99–126. [Google Scholar]
  94. Chen, L.; Jin, Z.; Michishita, R.; Cai, J.; Yue, T.; Chen, B.; Xu, B. Dynamic monitoring of wetland cover changes using time-series remote sensing imagery. Ecol. Inform. 2014, 24, 17–26. [Google Scholar] [CrossRef]
  95. Cohen, M.J.; Creed, I.F.; Alexander, L.; Basu, N.B.; Calhoun, A.J.K.; Craft, C.; D’Amico, E.; DeKeyser, E.; Fowler, L.; Golden, H.E.; et al. Do geographically isolated wetlands influence landscape functions? Proc. Natl. Acad. Sci. USA 2016, 113, 1978–1986. [Google Scholar] [CrossRef] [Green Version]
  96. Winter, T.C. Hydrologic studies of wetlands in the Northern Prairies. In Northern Prairie Wetlands; van der Valk, A., Ed.; Iowa State University Press: Ames, IA, USA, 1989; pp. 16–54. [Google Scholar]
Remotesensing 15 01870 g001 550
Figure 1. Localization of the geomorphological compartments in the São Paulo State, Brazil, overlayed with the distribution of the Cerrado and Atlantic Forest ecoregions (i.e., biomes). The zoom-in of the study area (Corumbataí River Basin DEM [44]) with wetlands identified by Junqueira [45] Furlan et al. [46]. A and B are the location of the two studied wetlands.
Figure 1. Localization of the geomorphological compartments in the São Paulo State, Brazil, overlayed with the distribution of the Cerrado and Atlantic Forest ecoregions (i.e., biomes). The zoom-in of the study area (Corumbataí River Basin DEM [44]) with wetlands identified by Junqueira [45] Furlan et al. [46]. A and B are the location of the two studied wetlands.
Remotesensing 15 01870 g001
Remotesensing 15 01870 g002 550
Figure 2. Multi-scale process used during the development of the different stages of the study, incorporating satellites, UAV, and Electrical Resistivity Tomography workflows and applications during the study.
Figure 2. Multi-scale process used during the development of the different stages of the study, incorporating satellites, UAV, and Electrical Resistivity Tomography workflows and applications during the study.
Remotesensing 15 01870 g002
Remotesensing 15 01870 g003 550
Figure 3. 1975–2020 band composition images of the Landsat series and CBERS-04A satellite. Pixel-based supervised classification in wetland A and B (zoom-in). The graphs display the result of calculations of the inundated area (water class) for each year-dataset, based on the supervised classifications.
Figure 3. 1975–2020 band composition images of the Landsat series and CBERS-04A satellite. Pixel-based supervised classification in wetland A and B (zoom-in). The graphs display the result of calculations of the inundated area (water class) for each year-dataset, based on the supervised classifications.
Remotesensing 15 01870 g003
Remotesensing 15 01870 g004 550
Figure 4. One-year (August 2019–July 2020) systematic acquisition of Wetland A (A) and Wetland B (B) UAV-orthomosaics.
Figure 4. One-year (August 2019–July 2020) systematic acquisition of Wetland A (A) and Wetland B (B) UAV-orthomosaics.
Remotesensing 15 01870 g004
Remotesensing 15 01870 g005 550
Figure 5. October-2019 orthomosaics, digital elevation models, and elevation profiles (ab) of the two wetlands, A and B. (A) Wetland with perennial flooding and no tree vegetation, perimeter of 1485.1 m; (B) Wetland with temporary flooding and tree vegetation on its limits, perimeter of 972.3 m. Wetland A is approximately 34% bigger than B.
Figure 5. October-2019 orthomosaics, digital elevation models, and elevation profiles (ab) of the two wetlands, A and B. (A) Wetland with perennial flooding and no tree vegetation, perimeter of 1485.1 m; (B) Wetland with temporary flooding and tree vegetation on its limits, perimeter of 972.3 m. Wetland A is approximately 34% bigger than B.
Remotesensing 15 01870 g005
Remotesensing 15 01870 g006 550
Figure 6. (A) The polygons extracted from Wetland A delimit the spatial pattern that reflects the wetland’s characteristics with different RGB pixel responses related to seasonal hydro-periods. The images show a zoom-in view of a border portion of Wetland A for October 2019 and February 2020; (B) The polygons extracted from Wetland B delimit the spatial pattern that reflects the wetland’s characteristics with different RGB pixel responses related to seasonal hydro-periods. The images show a zoom-in view of a border portion of Wetland B for October 2019 and February 2020.
Figure 6. (A) The polygons extracted from Wetland A delimit the spatial pattern that reflects the wetland’s characteristics with different RGB pixel responses related to seasonal hydro-periods. The images show a zoom-in view of a border portion of Wetland A for October 2019 and February 2020; (B) The polygons extracted from Wetland B delimit the spatial pattern that reflects the wetland’s characteristics with different RGB pixel responses related to seasonal hydro-periods. The images show a zoom-in view of a border portion of Wetland B for October 2019 and February 2020.
Remotesensing 15 01870 g006
Remotesensing 15 01870 g007 550
Figure 7. (w–x; y–z) 2D Inversion resistivity models, acquired based on Schlumberger array.
Figure 7. (w–x; y–z) 2D Inversion resistivity models, acquired based on Schlumberger array.
Remotesensing 15 01870 g007
Remotesensing 15 01870 g008 550
Figure 8. 3D subsurface slices visualization models, slices of the 3D blocks, resulting from lateral interpolation of 2D inversion models. Depth of 5, 15, 25, and 35 m.
Figure 8. 3D subsurface slices visualization models, slices of the 3D blocks, resulting from lateral interpolation of 2D inversion models. Depth of 5, 15, 25, and 35 m.
Remotesensing 15 01870 g008
Table
Table 1. Technical characteristics of acquisition and processing of the satellite images dataset used in the analysis.
Table 1. Technical characteristics of acquisition and processing of the satellite images dataset used in the analysis.
YearSeasonSatelliteSensorBand CompositionResolution
1975DryLandsat 2MSS6(NIR)-7(NIR)-5(R)80 m
1984DryLandsat 5TM4(NIR)-5(SWIR 1)-3(R)30 m
1992DryLandsat 5TM4(NIR)-5(SWIR 1)-3(R)30 m
2000DryLandsat 7ETM+4(NIR)-5(SWIR 1)-3(R) + 8 (PAN)30 m (15 m)
2010DryLandsat 5TM4(NIR)-5(SWIR 1)-3(R)30 m
2020WetCBERS-04AWPM3(R)-4(NIR)-2(G) + 0 (PAN)8 m (2 m)
Table
Table 2. Wetland A and B photogrammetric calculations performed.
Table 2. Wetland A and B photogrammetric calculations performed.
October 2019February 2020
CompartmentArea (m2)CompartmentArea (m2)
Wetland AA1dry57,259A1wet42,316
A2dry21,306A2wet130,339
A3dry94,089
Wetland BB1dry34,134B1wet34,134
B2dry31,558B2wet10,200
B3wet21,358
Table
Table 3. Precipitation (P), Evapotranspiration (ET), and Water Balance (WB).
Table 3. Precipitation (P), Evapotranspiration (ET), and Water Balance (WB).
MonthP (mm)ET (mm)WB (mm)
Period 1August-1913.895.181.3
September-1956112.0856.8
October-19126.4138.7412.34
Period 2November-19171.1116.554.6
December-19151.5112.1139.39
January-20175114.2860.72
February-2031583.19231.81
Period 1March-20128.4114.1914.21
April-208.4102.9594.55
May-2023.680.7457.14
June-2079.268.3610.84
July-208.280.5573.35
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

Furlan, L.M.; Ferreira, M.E.; Moreira, C.A.; de Alencar, P.G.; Casagrande, M.F.S.; Rosolen, V. Satellite, UAV, and Geophysical Data to Identify Surface and Subsurface Hydrodynamics of Geographically Isolated Wetlands: Understanding an Undervalued Ecosystem at the Atlantic Forest-Cerrado Interface of Brazil. Remote Sens. 2023, 15, 1870. https://doi.org/10.3390/rs15071870

AMA Style

Furlan LM, Ferreira ME, Moreira CA, de Alencar PG, Casagrande MFS, Rosolen V. Satellite, UAV, and Geophysical Data to Identify Surface and Subsurface Hydrodynamics of Geographically Isolated Wetlands: Understanding an Undervalued Ecosystem at the Atlantic Forest-Cerrado Interface of Brazil. Remote Sensing. 2023; 15(7):1870. https://doi.org/10.3390/rs15071870

Chicago/Turabian Style

Furlan, Lucas Moreira, Manuel Eduardo Ferreira, César Augusto Moreira, Paulo Guilherme de Alencar, Matheus Felipe Stanfoca Casagrande, and Vânia Rosolen. 2023. "Satellite, UAV, and Geophysical Data to Identify Surface and Subsurface Hydrodynamics of Geographically Isolated Wetlands: Understanding an Undervalued Ecosystem at the Atlantic Forest-Cerrado Interface of Brazil" Remote Sensing 15, no. 7: 1870. https://doi.org/10.3390/rs15071870

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