Next Article in Journal
Citizen Science for Environmental Monitoring in the Eastern Region of Bolivia
Previous Article in Journal
Preparation and Electrochemical Performance of Activated Composite Carbon Nanofibers Using Extraction Residue from Direct Coal Liquefaction Residue
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 6
10.3390/su16062332
sustainability-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

Sustainable Watershed Management: Assessing the Impact of Data Precision on Sediment Transport in the Oostanaula Creek Watershed, Tennessee

by
Miroslav Bauer
1,*,
Barbora Jáchymová
1,
Josef Krása
1,
Karina Bynum
2,
John S. Schwartz
3 and
Tomáš Dostál
1
1
The Department of Landscape Water Conservation, Faculty of Civil Engineering, Czech Technical University in Prague, Thakurova 7, 16629 Prague, Czech Republic
2
Tennessee Department of Environment and Conservation, Division of Water Resources, Cookeville, TN 38506, USA
3
Department of Civil and Environmental Engineering, The University of Tennessee, Knoxville, TN 37996, USA
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(6), 2332; https://doi.org/10.3390/su16062332
Submission received: 1 February 2024 / Revised: 23 February 2024 / Accepted: 27 February 2024 / Published: 12 March 2024
(This article belongs to the Section Social Ecology and Sustainability)
Browse Figures
Versions Notes

Abstract

:
This study applies the WaTEM/SEDEM model, a watershed-scale model based on the Universal Soil Loss Equation (USLE), to enhance sustainable watershed management by identifying high-erosion-risk areas for targeted mitigation at various scales. It focuses on identifying potential errors in using statewide or worldwide land use data layers derived from remote-sensing algorithms and inaccuracies in the spatial distribution of cropland and soil types resulting in misinterpretation of sediment yields. These model limitations emphasize the need for field validation and precise input data, particularly stream data, to improve the reliability of these models. The study examines hydrologic rainfall-runoff processes in Tennessee’s 182 km2 Oostanaula catchment using various data sources, including the National Land Cover Database (NLCD), the European Space Agency WorldCover dataset (ESA), and manual field surveys. Three modeling scenarios were evaluated, with and without stream topology corrections, using the WaTEM/SEDEM model. It details the global data used, the methodology of the field survey, the simulation and validation of data, and the critical point identification. Significant discrepancies in long-term sediment transport predictions were found, depending on the land use data source. This study addresses watershed model validity and potential errors and recommendations for the use of globally available data.

1. Introduction

Intensive precipitation events and the resulting soil erosion and sediment transport are a global problem that leads to disruption of landscape processes and degradation of arable land and water resources [1]. Although soil erosion and sediment transport are mainly natural processes, human use of the landscape, especially through unsustainable agricultural activities, accelerates the whole process, including the severity of the consequences [2]. Excessive erosion can also cause extensive damage to community infrastructure and even directly endanger the lives of residents [3]. More than 20% of agricultural land in over a third of OECD member countries is at risk of moderate to severe erosion [4]. Worldwide, up to 30% of agricultural land is threatened by erosion due to the intensification of agriculture, and, among other consequences, erosion intensively threatens food production [5,6,7].
Excessive soil erosion on the land and in-stream channel erosion are global problems impacting the sustainability of the world’s water resources [8,9]. Central Europe experiences impacts from sediments transported from agricultural land, leading to damage in water courses and water bodies and damage to infrastructure in urban areas [10]. Furthermore, in the central and southern parts of the United States (including the Oostanaula Creek watershed), bank erosion is thought to contribute significantly (if not critically) to the overall transport of erosional sediments [11]. As can be seen, it is no longer sustainable to exploit the landscape and overlook the consequences of human activities [12]. Thanks to research activities, the causes and consequences of soil loss have been known for decades, but due to the rapid development of industrial society and technologies, the consequences are often ignored and underestimated because agricultural profit is the priority [13].
Another environmental factor further worsening the situation and deepening the conflicts of interest between the environment, people, the economy, and society is global climate change [14]. One of the studies predicts worsening soil erosion rates by up to 10% in the next 50 years due to climate change [15]. The global climate is changing dramatically and even faster than expected. It influences the distribution of precipitation (in some places) and the amount of precipitation and also changes the kinetic energy. One of the changing factors that is often neglected is the change in crop rotation in response to climate change or even the development of new crop varieties that affect the economic profit of the current market depending on the political situation [16].
Researchers observing and predicting climate change, coupled with further land development impacting our society, need to take action by effectively communicating, explaining, and disseminating information. An appropriate avenue for this is to outline the ongoing search for sustainable, socially acceptable solutions (measures) to reduce soil loss, sediments, and water pollution caused by factors such as bound nutrients and pesticides. This effort involves compiling information from the research literature and translating it into a format suitable for diverse audiences, including farmers, associations, competent authorities and the general public. The process includes describing the current environmental condition, identifying high-risk factors and areas, and testing various scenarios to understand potential model outcomes.
To model and assess the risk of severe erosion for a specific location, researchers can employ erosion and transport models to determine the soil loss or erosion intensity on the land or in the watershed. In general, models can be divided into empirical-statistical, conceptual, and physical models [17]. Models differ, on the one hand, in the required quality and quantity of inputs and, on the other hand, in the resulting quality and representativeness of the delivered results. The interannual yield of sediments, applied in reservoirs, was solved by using probabilistic erosion models [18]. For ungauged catchments, a mathematical model can be used to estimate erosion, as in the work of Hrissanthou [19].
Empirical models based on the principle of the Universal Soil Loss Equation (USLE) are generally used to determine the erosion risk in large areas with an acceptable amount and detail of inputs. The USLE determines the long-term average soil loss in a catchment caused by sheet erosion. One such model is the spatially distributed WaTEM/SEDEM model [20,21,22], which provides a sufficiently accurate estimate of erosion intensity and transported eroded material based on a relatively small amount of input data [23]. Based on the global review of the most commonly used models for predicting soil erosion and sediment yields, the WaTEM/SEDEM model is within the top five with RUSLE, USLE, WEPP, and SWAT [24]. This model has been extensively tested and used to determine sediment transport and sedimentation in reservoirs under conditions in the Czech Republic [25,26,27] as well as worldwide [28,29].
Land use is one of the critical inputs for watershed models that significantly influences the accuracy of erosion and sediment transport [10]. In general, the degree of the ability to control the land use layer quality is proportional to the size of the modeled area. For extensive areas, it is very time-consuming to perform targeted surveys to verify the accuracy of the input data. Nowadays, there are various satellite products that offer commercial or freely available land use data suitable for modeling. The Landsat program has the longest history, and its products are available both worldwide and regionally. In the US, the Global Land Cover dataset or the NLCD (National Land Cover Database) have been widely used [30,31], publicly available in 30 m resolution.
In Europe, the Corine Land Cover database has mostly been used in recent research [32]. Based on the Sentinel-2 satellite family, the European Space Agency recently developed the ESA WorldCover dataset in a much better spatial resolution of 10 m [33]. These recent advances make it possible to improve the input data worldwide, especially in regions where no other detailed land use product is available. In Europe in particular, detailed vector-based land use data for individual agricultural parcels are available as Land Parcel Identification System (LPIS). It was developed by the EU countries to control the compliance of farmers for subsidies and is generally available for any type of public or research use. In the US, there is no such publicly available detailed source that could be used. A particular focus of the indicated measures can be placed on buffer strips along watercourses to enhance trapping efficiency, initiate sediment deposition, and reduce water pollution. Such measures can contribute significantly to water protection [34]. Since the output of sediment yields from watershed models is highly dependent on accurate spatial land cover data, there is a critical need to better understand how potential land cover inaccuracies affect model outputs.
This manuscript presents a case study targeting a catchment in the Oostanaula Creek watershed located in the state of Tennessee. Within the presented work, the importance of land use data input was assessed to answer the following questions:
1. How significant is the influence of the land use input dataset in terms of modeled sediment inputs into watercourses?
2. Are generally available datasets of land use of sufficient quality to define high-risk locations in terms of erosion? Can they be reliably utilized for assessment without in situ validation of land use datasets?
3. Is it possible to define the probable contribution of channel erosion to the total transport of sediment from the catchment by combining erosion modeling in the catchment and field survey of the hydrologic network?

2. Materials and Methods

The presented case study has been prepared for a catchment within the Oostanaula Creek watershed, where a mathematical model of soil erosion and sediment transport, the WaTEM/SEDEM model [20,21,22], was used to assess the impact of alternative input data sources and to determine critical profiles concerning sediment input and transport.

2.1. Oostanaula Creek Watershed Study Site

The Oostanaula Creek catchment (Figure 1) is located in southeast Tennessee (tributary of the Hiwassee River). The area covered by modeling is 182.1 km2 of the Oostanaula Creek watershed—OCW (HUC 06020021101 and 060200021102).
Oostanaula Creek is typical of streams within the Ridge-and-Valley region of the eastern United States. Annual precipitation ranges from 1100 to 1400 mm. Average summer temperatures range from 19 to 30 °C, and January temperatures range from −3 to 7 °C. Elevations in the OCW range from 240 to 340 m above sea level. The lowest point is in the southwestern part of the watershed at Calhoun, where Oostanaula Creek flows into the Hiwassee River and Chickamauga Lake. This location is about 210 m above sea level [35].
A total of 323 km of streams (280 km natural and 43 km artificially stabilized) was included in the model in the land area covered by the following land use: forest (46%), grassland (38%), urban areas (12%), and arable lands (4%). The part of the catchment above the city of Athens, Tennessee, is dominated by agricultural use (cultivated crops/pasture), compared with the lower part of the catchment, which is covered mostly by forest. The outlet profile of the entire Oostanaula Creek watershed is located near the town of Calhoun, where Oostanaula Creek flows into the Hiwassee River.
Termed the “Oostanaula-Top”, it is a sub-catchment (9.4 km2) that was later identified as the research focus, consisting of field surveys, detailed calculations, and model tests (Figure 1—purple outline). The outlet point of the sub-catchment was established under the County Road 350 bridge crossing Oostanaula Creek (5.3 km of river length). The land use of the Oostanaula-Top part of the catchment is mainly covered by forest (41%), grassland (44%), arable lands (10%), and urban areas (3%). All abovementioned land uses are based on the National Land Cover Database 2016 (NLCD) [30].

2.2. Model WaTEM/SEDEM Inputs and Parameters

The WaTEM/SEDEM model [20,21,22] is an erosion and transport model based on a simple empirical equation considering six factors based on the USLE method [36]. The six factors include the rainfall erosivity factor (R); the slope length and slope factors (LS); the vegetation cover and tillage factor (C); the soil erodibility factor (K); and the erosion protection factor (P). The contributing area for the LS calculation is defined by the multiple flow algorithm for splitting the runoff into several streams (directions) [37]. The full functionality of the model deals with sediment transport—firstly, calculating distributed soil loss; secondly, calculating sediment transport using the transport capacity equation for each element (pixel); and finally, estimating the volume of sediment reaching the river network segments and summarizing the volume leaving the catchment annually (final river segment). The original USLE approach only estimates the potential soil erosion, while the WaTEM/SEDEM model calculates the transport capacity potential in each pixel and then determines the erosion/deposition and potential of the sediment transported from each pixel. The WaTEM/SEDEM model 2.1.0. version was used. For successful calculations, at least these spatially distributed inputs are required: digital elevation model (DEM), land use, and USLE factors (R—rainfall, K—soils, C—vegetation). To determine the sediment load in rivers, the topology of the river (river section) must also be prepared with the correct orientation and linkage. The possibility of sediment deposition in ponds was not relevant for the study and was not considered.
The DEM is based on a tiled collection of the 3D Elevation Program (3DEP) and has a resolution of 1/3 arc second (approximately 10 m) [38]. The 3DEP dataset serves as the elevation layer of the national map and provides basic elevation information for Earth science studies and mapping applications in the United States. Scientists and resource managers use 3DEP data for hydrologic modeling, resource monitoring, mapping and visualization, and many other applications. The elevations in this DEM represent the topographic surface of the bare Earth.
The R-factor used in this study was set at 4000 MJ·mm·ha−1·h−1·yr−1, based on a revised R-factor study published [39] with values very close to those originally known from RUSLE [40]. Tennessee is the sixth rainiest state in the United States according to a recent analysis of monthly and annual precipitation in each state, with an average annual precipitation of 1310 mm.
The K-factor was derived from the “STATSGGO” soil database. This nationwide STATSGO soil database was provided by the National Resources Conservation Service (NRCS). K-factor values were estimated directly based on the known distribution of soil units. K-factor values vary from 0.21 to 0.58 with the mean value of 0.28 (±0.02) in the entire watershed. In the Oostanaula-Top catchment, the min and max K-factor is the same range as the entire watershed, 0.21 to 0.58, and the average value is nearly the same, 0.27 (±0.04).
Data acquisition for the C-, LS-, and P-factors is described next. The C-factor was determined based on information from local farmers. The crop rotation consists of soybeans and maize, mainly using shallow, minimum tillage agrotechnology. The C-factor was set at 0.3 for arable land, 0.01 for grassland, and 0 for all other (non-erosive) types of land use. Due to the focus of the study, which compares the influence of variable factors, the C-factor was used equally in all scenarios. The LS factor is calculated as a routine of the WaTEM/SEDEM model based on the set calculation parameters. The LS and Nearing slope exponent from McCool [41] were used. The P-factor remains at 1 as no soil protection measures/strategies are applied.
Final model parameters setup for all modeled scenarios are based on previous calibration and usage [25,26,27]: Ktc (Low) = 35, Ktc (High) = 55, Ktc (Limit) = 0.1; Ptef (Cropland) = 0, Ptef (Forest, Pasture) = 75; and Parcel Connectivity (Cropland) = 40, Parcel Connectivity (Forest, Pasture) = 75, model parameters LS and Nearing slope exponent used by McCool [41], where the Ktc values are transport capacity coefficients, Ptef is a parameter dealing with parcel trapping efficiency by reducing drainage area based on land use coverage, and the Parcel Connectivity parameter issues the boundary effect of the single parcels. More detailed information is provided in documentation of the model published [21] and available online on the model website.

2.3. Preliminary Investigation of Land Use Datasets for WaTEM/SEDEM Model Input

Two land use datasets were obtained for use in the WaTEM/SEDEM model; they were the National Land Cover Database (NLCD) and the ESA WorldCover dataset. The NLCD is an operational land cover monitoring program providing updated land cover and related information for the United States at five-year intervals. The NLCD2016 extends temporal coverage to 15 years (2001–2016) and was published in 2019. The NLCD is a widely recognized and widely utilized database [30]. The ESA WorldCover dataset is a newly available worldwide data source in better spatial and temporal resolution (10 m), published in late 2022 based on Sentinel-2 and Sentinel-1 satellite images, reaching overall accuracy of 77% and independently validated [33].
Initially, the main purpose of the study was to use the NLCD, a publicly available and commonly used dataset across the United States supplying land use input data for the WaTEM/SEDEM model, and use that data for identifying problematic areas. Data were also used to build model scenarios of conservation areas through the installation of grass buffer strips along watercourses and ultimately to evaluate the improvements from protective practices. During the initial model development, substantial discrepancies were found between the source of land use data (NLCD, 30 m resolution) and visible observations from orthophoto aerial images, which appeared to strongly influence the model results.
Fortunately, to improve the land cover dataset, the ESA WorldCover dataset was available as an alternative data source for the study’s model. At the same time, a manual dataset of land use was created directly onsite through field measurements using imagery from an unmanned aerial vehicle (UAV).
As can be seen from the land use comparison (Figure 2) and the chart (Figure 3), the two data sources differ considerably, particularly in the classification of the built-up areas. In terms of the total watershed area (182 km2), the difference in cropland does not appear to be too great, but, especially near the watercourse, cropland (vs. other use) can play an important role in protecting the watercourse from the sediment transport. Grassland and tree cover are also of high importance in the protection of the soil surface.
Differences in land use categorical descriptors are compared and summarized in Table 1. Within the ESA WorldCover dataset, there is the coverage category “built-up” area, but based on automatic classification, this covers main routes (higher categories) only. For reasons of comparability with the NLCD, only the external update function, which is based on the Open Street Map route definition [42], was rasterized to the same 10 m resolution and placed over the ESA land use to provide comparable land fragmentation.

2.4. Oostanaula-Top Catchment Field Survey

The spatial discrepancies between the two available land use databases necessitated a ground-truthing field study. The main objective of the field survey was to map land use in detail, with a particular focus on arable land as opposed to grassland, as these categories significantly influence the severity of the soil erosion process. The second important reason for the field survey was to validate the stream network, as this is crucial information affecting sediment routing through the catchment. The field survey was conducted in the late summer of 2022. The area covered was the limiting factor for the detailed survey and the subsequent processing time. The upper part of the catchment (9.4 km2) was pre-selected as suitable for a detailed survey. The area was photographed with a UAV DJI Mavic 2 Pro with RGB camera and later digitized manually on the orthophoto background using a field-validated photo dataset. Manual digitalization based on available UAV monitoring precision orthophoto data sources covers the best available and current real dataset verified for the area. Visual comparison can be clearly seen in Figure 4.
The northern part of Oostanaula watershed (Oostanaula-Top sub-catchment) shows a significant trend towards more arable land and less grassland when we compare the NLCD first, the ESA second, and FIELD manual land use last (Figure 5). It can be expected that a lower spatial resolution neglects part of the reality, especially the contribution of arable land to soil loss and sediment transport.
The stream channels are important inputs influencing the connectivity of sediments from the agricultural fields to the outlet of the sub-catchment. Based on the manual field survey, many inaccuracies were identified in the channel definitions. As can be seen in Figure 6, some of the stream channels present in the official database (National Hydrography Dataset—NHD) no longer exist (or never existed). Based on the model approach (the WaTEM/SEDEM model), all the sediment reaching the river channel is transported to the outlet point (no redistribution within the river channels is calculated). Especially for that reason, the correct stream network is a crucial input providing relevant data at the outlet point based on a specially distributed soil loss/sediment transport model.
The Oostanaula-Top sub-catchment survey focused also on defining channel geometries to be able to evaluate real and potential channel volumes, volumes of sediment within channels, indices of bank and streambed erosion and sedimentation, and channel cross-section changes in different parts of the Oostanaula-Top sub-catchment. Open channels on the agricultural fields were surveyed in full length using UAV photogrammetry methods to create high-detail digital surface models to be able to define channel cross-sections and connectivity of fields to initial stream parts. Photogrammetry could not be used under the trees and in the forested stream sections, where direct measurements were taken where necessary (Figure 7).
For direct measurement of the stream cross-sections, LIDAR technology was used as it is implemented in Apple iPhone 13 PRO. The accuracy of the technology proved to be high enough to reproduce streambed shapes in the Oostanaula-Top sub-catchment. The device was used to create 3D models of the streambed in 14 different locations describing representations of cross-sections for each monitored stream type. The data were processed with an open 3D scanner application (https://3dscannerapp.com/, accessed on 1 January 2024) and by manual profile definition for different water levels. The tiled models of the river bottom were also used for potential sediment volume estimation (Figure 8).

3. Results

Erosion and sediment transport in the entire Oostanaula Creek watershed were calculated using the WaTEM/SEDEM model to compare the available land use and stream drainage network data sources. As noted above for land use, two generally available data sources (NLCD and ESA) were used and subsequently compared. A field survey was carried out for the Oostanaula-Top catchment, and it focused on preparing a detailed land use dataset and observing the actual condition of the watercourses. Based on the field survey, a land use layer dataset was compiled. In the Oostanaula-Top sub-catchment, the results of the erosion model were compared for three land use datasets with varying degrees of accuracy (NLCD land use model, ESA land use model, FIELD land use model based on field survey).

3.1. Oostanaula Creek Wateshed Results

First, the NLCD land use model calculations were performed throughout the Oostanaula Creek watershed (182.1 km2), resulting in values for erosion of more than 56,400 tons/year, deposition of 35,700 tons/year within the full-size catchment area, and sediment export of 20,700 tons/year through the outlet segment of Oostanaula Creek. For the ESA dataset, overall results show erosion of 77,900 tons/year, deposition of 52,700 tons/year in the watershed, and sediment transport trough of the outlet of 24,400 tons/year. Although the total values for erosion are significantly higher for the ESA land use dataset due to the higher deposition within the watershed, the values for sediment input into the watercourses are similar for both land use datasets (see Table 2).
In accordance with the main aim of erosion modeling in the Oostanaula catchment, high-risk points of sediment inflow into watercourses were defined. Fifty watercourse sections (approx. 10%) with the highest sediment input were defined as risky. The total input of sediment into these risk sections is about 66% of the total sediment load of the hydrologic network (for both land use data sources). Risk points in the Oostanaula catchment (for the NLCD and ESA land use datasets) are shown in Figure 9. Both data sources (NLCD and ESA) overlap in 33 risk points (38.3 km of rivers and 11,000 tons of predicted sediment, respectively). This indicates that the ESA dataset covers most of the significant risk zones in the NLCD, but, unfortunately, it cannot be concluded that the NLCD covers most of the high-risk places in the ESA dataset.
Table 3 provides information about erosion and sediment transport analysis in more detail. According to the NLCD land use dataset, the watercourse network is less loaded with sediment from agricultural parcels. Also, the risk points (or watercourse sections) are significantly different for both land use datasets. The total length of the watercourse sections that pose a risk for excessive sediment transport is longer when the NLCD is used than when the ESA land use dataset is used in the model.

3.2. Oostanaula-Top Sub-Catchment WaTEM/SEDEM Model Results

Based on the results of the erosion model using two commonly available data sources (NLCD and ESA) in the Oostanaula Creek watershed, it was found that land use input significantly influences the overall results and especially the location of the risk points or vulnerable water body sections in terms of sediment input.
Based on a field survey of the Oostanaula-Top sub-catchment (9.4 km2), a new land use dataset (FIELD land use) was defined to reflect the actual condition in this part of the catchment. The field survey also included the revision and subsequent adjustments of the stream network in the upper Oostanaula-Top sub-catchment. In the Oostanaula-Top sub-catchment, the calculation was carried out with the WaTEM/SEDEM model in the three land use variants, the NLCD, ESA, and FIELD. The differences between the individual land use datasets (NLCD, ESA, and FIELD) for the Oostanaula-Top sub-catchment are illustrated in Figure 4 and results comparison in Table 4.
In Figure 10, the visualization of the four stream sections can be seen. They are defined as (1) streams that are present in the official dataset but not in the field based on the field study, (2) the main stream of Oostanaula Creek, (3) the natural drainage channel for the main agricultural fields, and (4) the natural drainage channel for the forest valley areas and natural areas. Table 5 below describes the results of the four main sections of the drainage network and the sediment transport from nearby agricultural fields.
As can be seen from the comparison of the results in Table 5, in the first section (1), both the NLCD and ESA dataset achieve almost the same sediment transport with the dataset of the existing streams. In contrast, the FIELD dataset excludes the non-existent streams that significantly affect the sediment connectivity. Most of the eroded material is deposited before it reaches the existing open channels. In the second (2) and third (3) sections, it can be seen that sediment volumes are underestimated for the ESA dataset (2×–2.5×) and even worse for the NLCD (3×–5×); both cases are compared to the FIELD survey based on land use. In the fourth (4) section, the NLCD slightly underestimates reality and the ESA dataset overestimates it, but the overall difference is small.
Stream sections with approx. 75–80% of the total transport in the sub-catchment were selected for the identification of the particularly vulnerable areas (Figure 11). The comparison of the results in Table 6 shows that the FIELD dataset identified the most risky parts of the drainage network in less than 60% of the stream length identified by either the NLCD or ESA and highlighted the most risky areas (Figure 11) with extreme sediment inputs.

3.3. Stream Channel Volumes

Another important piece of information that emerged from the field survey was the analysis of the volume of the stream channels based on 14 measured cross-sections. From the measurements, it can be estimated that the total geomorphic volume of the stream channels is almost 26,600 m3, which corresponds to the value of 32,000 tons of sediment, and that the lowest 20 cm of the channel cross-section represents a volume of 5000 m3. Focusing on the northern part of the stream network, segment (3), noted previously as an agricultural drainage channel (Figure 10), results in a total channel volume of 4120 m3 for 2.8 km. Compared to the modeled sediment output, the same stream section (3) receives more than half of the sediment that reaches the river channels in the Oostanaula-Top sub-catchment (2150 tons/year). Characteristic of stream channels cross-sections is documented in Figure 12.

4. Discussion

The highlight of this study was demonstrating the differences in sediment yields that can occur depending on the land use and drainage network datasets applied in a watershed model. However, without measured sediment yields, the study only provided a relative comparison of yields from modeling scenarios using different land use datasets. Validating the WaTEM/SEDEM model results with measured sediment data would have been informative for its accuracy. Essentially, the only way to verify sediment transport is to measure sediment volume in reservoirs, as reported in [27], or to monitor the suspended sediment in rivers. However, as erosion events are highly episodic, such validation requires detailed monitoring with special attention to extreme discharges.
According to previous studies, 37% of the sediment in the streams in Oostanaula Creek watershed is due to surface erosion from agricultural land [35]. Since the measured sediment amounts in the streams (recorded by long-term monitoring) are relatively low (600 t/year) and the modeled values are in the higher ranges (20,000–25,000 t/year depending on the data source), most of the sediment enters the streams during intense erosion events. These episodes are not covered by long-term monitoring. Eroded material can be an important source of total phosphorus [43]. Reducing the transport of sediment and associated bound phosphorus during erosion can help improve surface water quality in the watershed. Modeled data compared to measured sediment concentrations in streams indicate significant sediment input during intense rainfall events. Therefore, improved erosion control management can also be beneficial in terms of reducing total phosphorus transport.
For this reason, as part of a watershed restoration plan [44], it is appropriate to plan erosion control measures on selected agricultural plots. If we follow the results of the erosion model, it becomes clear that the NLDC land use model defines different risk locations than the ESA model (see Figure 6). Significant differences in both data sources are documented in Figure 13. We see that the resolution and spatial detail of the ESA land use is higher and, in most cases, closer to the real condition based on visible land use patterns from orthophoto images. These differences are reflected in the erosion model and significantly affect the reliability of their results.
Discrepancies in land use patterns on the other hand were identified in all parts of the catchment and can play a key role in spatial pattern errors leading to misinterpretations of the sediment yield results. The only approach to correctly estimate or validate existing stream topology data is the field survey which was conducted here. Especially here, partly sufficient validation and correction can be performed with the database with orthophoto image time series. Probably the most comprehensive and longest available datasets worldwide can be viewed in the Google Earth Pro app (Figure 14). With limited certainty, a sufficient degree of correction can be achieved even without direct field inspection, which was confirmed here in the Oostanaula catchment. All significant topological differences in the stream network in the Oostanaula-Top sub-catchment were also successfully identified in aerial photographs (Figure 14). Here, it is necessary to mention that the other discrepancies in the topology of the streams were barely recognized in other parts of the Oostanaula Creek catchment, except for the upper catchment, which was surveyed in detail.
As was mentioned in the results, the total agricultural sediment transport into the streams in the watershed ranged from 20,000 to 25,000 t/year depending on the data source, which can be interpreted as sufficient results for both data sources. With knowledge of the difference from the small part of the upper Oostanaula-Top catchment, the actual transport that can be expected can reach up to 40,000 tons/year for the entire Oostanaula Creek watershed. The overall fact is that although erosion processes are severe in Oostanaula, they are located in particular parts. The sediment delivery ratio comparing the entire watershed reached 37% (NLCD) and 31% (ESA), compared with the upper sub-catchment, with values of 49% (NLCD), 50% (ESA), and 36% (FIELD). Lowering the SDR by “disconnecting the landscape” by designing buffer strips or installing field measures in the proper locations can significantly improve water quality issues.
The land use of the Oostanaula Creek watershed is spatially variable. As already mentioned above, most of the eroded material is transported into watercourses during intense rainfall-runoff events. From the results of the erosion model presented in this article, it is clear that the transport of sediment to watercourses is mainly localized in the agricultural part of the Oostanaula Creek watershed, which is in the north of this territory. This part (9.4 km2) makes up approximately 5% of the area of the entire watershed (182.1 km sq). The transport of erosional material from this part amounts to 8–13% of the total input of sediment into watercourses in the Oostanaula Creek watershed and is localized in a small number of heavily loaded parts of the stream network.
Therefore, it is suitable to combine different management layers. It is advisable to use generally available data, including orthophoto images, for orientation in the area and the search for locations with the occurrence of intensive erosion. For the detailed measure planning in the watershed, these analyses must be supplemented by a detailed field investigation, including a survey of the stream network.
Already, an estimated 37% of the sediment in the streams [35] in the Oostanaula Creek watershed is due to surface erosion from agricultural land. The results of the combination of the erosion model with the field investigation in the agricultural upper part of the catchment show that the situation here is quite different. The total volume of the streambeds in this part is 26,600 m3, which corresponds to the value of 32,000 tons of sediment. The annual value of sediment transport into watercourses is 4120 tons (according to the FIELD land use dataset). With a surface erosion rate of 37%, streambeds in the upper watershed would form in 3 years. From the perspective of the Oostanaula-Top channel volume, it is clear that sediment transport dominates, especially in the agricultural parts, compared to the estimated erosion of the channels. The 3-year volume of sediment transported from the agricultural fields corresponds to the volume of the channels that have been developing here for decades.
The data utilized in this study are subject to the inherent limitations of the USLE-based methods, which include uncertainties associated with long-term averages. Consequently, the presented values are not intended to serve as precise predictions of sediment yields. The WaTEM/SEDEM model can be characterized as a semi-quantitative tool, exhibiting varying degrees of predictive effectiveness across different watersheds. Despite these limitations, it is crucial to understand the model’s purpose: to support policy decision-making processes. The primary goal was to accurately describe the spatial distribution of major sediment sources, and this objective appears to have been reliably and effectively achieved.

5. Conclusions

Sustainable management of natural resources is topical globally. Under the conditions of climate change, soil quality and water quality and quantity have become more sensitive and often require more intensive and effective protection. To ensure effective, sustainable, and, at the same time, economically viable protection of natural resources, sufficient attention must be paid to quality and careful investment planning. Part of this planning is also a detailed analysis of the erosion and sediment transport processes and the search for the most problematic places.
With regards to the US Federal Clean Water Act and the associated Oostanaula Creek Watershed Restoration Plan [44]—Goal 1: Prevent and Eliminate Pollution Generation, the task of “providing sufficient data to make better land use decisions” is mentioned. Goal 2 is about reducing the amount of nutrients, sediment, and pathogens; Goal 3 is about capturing pollution before leaving critical areas. All of these are directly related to the study presented and all the methods described contribute to sustainable development and conservation of the cultural landscape by using appropriate datasets and sufficient, easy-to-use methods based on available input data. All methods mentioned here are easily applicable in other regions if the limitations of the methods are known.
This manuscript presents a case study targeting the Oostanaula Creek watershed located in the state of Tennessee. Within the presented study, the quality of generally available data was assessed, as was the possibility of their use for erosion and transport analyses and the definition of critical watercourse segments in terms of sediment input. Using the WaTEM/SEDEM model, at-risk stream segments were searched for in the Oostanaula catchment for two generally available data sources (NLCD land use dataset and ESA land use dataset). The results of the comparison show that both models agree in the total volume of erosion in the entire Oostanaula Creek watershed. Differences in the identification of at-risk locations were found. In a detailed analysis of the land use datasets, significant differences between the two datasets were observed. The ESA land use dataset is more detailed and corresponds more to the actual land cover conditions in the catchment. Results of the model using the ESA land use dataset are of sufficient quality for the general localization of at-risk locations to target further detailed activities and analyses. For the detailed design of measures in the watershed with the aim of maximizing their effectiveness, it is always necessary to validate generally available data directly in the assessed area.
Based on the identified differences in land use datasets, a field investigation was conducted in the upper (agricultural) part of the Oostanaula-Top sub-catchment. Also conducted were a survey of the watercourses, refinement of the watercourse input dataset for the model, and targeting of the watercourse profile in different parts of the watershed to determine the volume of riverbeds in these parts of the watershed. The results of the erosion model using the refined land use dataset (FIELD land use dataset) indicated a significantly greater total soil erosion and thus a significantly greater input of this sediment into the drainage network’s streams. This fact is due both to inaccuracies in the land use dataset and above all to inaccuracies in the stream network, which significantly changes the connectivity of the watershed and thus the overall sediment load in the watercourses.
Based on the analysis of the shape and size of riverbeds in the upper part of the Oostanaula, it was found that, although the main source of sediment is riverbed erosion in the entire Oostanaula Creek watershed, in the upper part of Oostanaula-Top sub-catchment, the situation is completely different. The main source of sediment input in the watercourses is eroded material transported from the agricultural land. The share of erosion in the total sediment in the streams is significantly higher than 37% (average value in the Oostanaula watershed). This case study demonstrates the importance of applying detailed land use and channel patterns with the use of a UAV if that resource is available, because current land use databases with inaccuracies can markedly affect modeled sediment yields.

Author Contributions

Conceptualization, T.D., J.S.S. and K.B.; methodology, T.D., J.K. and M.B.; software, M.B.; validation, T.D. and J.K.; formal analysis, M.B. and J.K.; investigation, M.B., J.K., T.D., K.B. and J.S.S.; resources, M.B., K.B. and J.S.S.; data curation, M.B.; writing—original draft preparation, M.B., T.D., J.K. and B.J.; writing—review and editing, K.B. and J.S.S.; visualization, M.B. and B.J.; supervision, T.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Ministry of Education of the Czech Republic, grant number LTAUSA19019, and the Technology Agency of the Czech Republic, grant number SS05010180.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

All data can be provided for further research activities and cooperation—please contact the authors directly.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Weggel, J.R.; Rustom, R. Soil Erosion by Rainfall and Runoff—State of the Art. Geotext. Geomembr. 1992, 11, 551–572. [Google Scholar] [CrossRef]
  2. Hirshfield, F.; Sui, J.; Hirshfield, F.; Sui, J. Changes in Sediment Transport of the Yellow River in the Loess Plateau. In Sediment Transport; IntechOpen Limited: London, UK, 2011. [Google Scholar] [CrossRef]
  3. Toy, T.J.; Foster, G.R.; Renard, K.G. Soil Erosion: Processes, Prediction, Measurement, and Control; John Wiley & Sons: Hoboken, NJ, USA, 2002; ISBN 9780471383697. [Google Scholar]
  4. OECD. Environmental Performance of Agriculture at a Glance; OECD Publications: Paris, France, 2008; ISBN 978-92-64-04589-7. [Google Scholar]
  5. Pimentel, D. Soil Erosion: A Food and Environmental Threat. Environ. Dev. Sustain. 2006, 8, 119–137. [Google Scholar] [CrossRef]
  6. Pimentel, D.; Burgess, M. Soil Erosion Threatens Food Production. Agriculture 2013, 3, 443–463. [Google Scholar] [CrossRef]
  7. Nanda, B.K.; Sahoo, N.; Panigrahi, B. Assessment of Soil Loss from Agricultural Lands of South Andaman District in Tropical Islands. Int. J. Curr. Microbiol. Appl. Sci. 2019, 8, 2190–2198. [Google Scholar] [CrossRef]
  8. Walling, D.E.; UNESCO; International Hydrological Programme; World Water Assessment Programme (United Nations). The Impact of Global Change on Erosion and Sediment Transport by Rivers: Current Progress and Future Challenges; United Nations Educational, Scientific and Cultural Organization: Paris, France, 2009; ISBN 978-92-3-104135-8. [Google Scholar]
  9. Owens, P.N. Soil Erosion and Sediment Dynamics in the Anthropocene: A Review of Human Impacts during a Period of Rapid Global Environmental Change. J. Soils Sediments 2020, 20, 4115–4143. [Google Scholar] [CrossRef]
  10. Bauer, M.; Dostal, T.; Krasa, J.; Jachymova, B.; David, V.; Devaty, J.; Strouhal, L.; Rosendorf, P. Risk to Residents, Infrastructure, and Water Bodies from Flash Floods and Sediment Transport. Environ. Monit. Assess. 2019, 191, 85. [Google Scholar] [CrossRef] [PubMed]
  11. Woockman, R.R. Rethinking Channel Protection Efforts: An Integration of Fluvial Geomorphology, Engineering, and Economics; The University of Tennessee: Knoxville, TN, USA, 2018. [Google Scholar]
  12. Gebeltová, Z. Exploitation of Agricultural Land in the Czech Republic and EU Countries. Agris On-Line Pap. Econ. Inform. 2017, 9, 33–44. [Google Scholar] [CrossRef]
  13. Chaplin-Kramer, R.; Chappell, M.J.; Bennett, E.M. Un-Yielding: Evidence for the Agriculture Transformation We Need. Ann. N. Y. Acad. Sci. 2023, 1520, 89–104. [Google Scholar] [CrossRef]
  14. Ashok, S. Climate Change, Collective Action, and Peaceful Change. In The Oxford Handbook of Peaceful Change in International Relations; Oxford Academic: Oxford, UK, 2020; pp. 337–350. [Google Scholar] [CrossRef]
  15. Borrelli, P.; Robinson, D.A.; Panagos, P.; Lugato, E.; Yang, J.E.; Alewell, C.; Wuepper, D.; Montanarella, L.; Ballabio, C. Land Use and Climate Change Impacts on Global Soil Erosion by Water (2015–2070). Proc. Natl. Acad. Sci. USA 2020, 117, 21994–22001. [Google Scholar] [CrossRef]
  16. Arora, N.K. Impact of Climate Change on Agriculture Production and Its Sustainable Solutions. Environ. Sustain. 2019, 2, 95–96. [Google Scholar] [CrossRef]
  17. Merritt, W.S.; Letcher, R.A.; Jakeman, A.J. A Review of Erosion and Sediment Transport Models. Environ. Model. Softw. 2003, 18, 761–799. [Google Scholar] [CrossRef]
  18. Rodríguez González, C.A.; Rodríguez-Pérez, Á.M.; López, R.; Hernández-Torres, J.A.; Caparrós-Mancera, J.J. Sensitivity Analysis in Mean Annual Sediment Yield Modeling with Respect to Rainfall Probability Distribution Functions. Land 2022, 12, 35. [Google Scholar] [CrossRef]
  19. Hrissanthou, V. Estimate of Sediment Yield in a Basin without Sediment Data. Catena 2005, 64, 333–347. [Google Scholar] [CrossRef]
  20. Van Oost, K.; Govers, G.; Desmet, P. Evaluating the Effects of Changes in Landscape Structure on Soil Erosion by Water and Tillage. Landsc. Ecol. 2000, 15, 577–589. [Google Scholar] [CrossRef]
  21. Van Rompaey, A.; Verstraeten, G.; Van Oost, K.; Govers, G.; Poesen, J. Modelling Mean Annual Sediment Yield Using a Distributed Approach. Earth Surf. Process. Landf. 2001, 26, 1221–1236. [Google Scholar] [CrossRef]
  22. Verstraeten, G.; Van Oost, K.; Van Rompaey, A.; Poesen, J.; Govers, G. Evaluating an Integrated Approach to Catchment Management to Reduce Soil Loss and Sediment Pollution through Modelling. Soil Use Manag. 2002, 18, 386–394. [Google Scholar] [CrossRef]
  23. De Vente, J.; Poesen, J.; Verstraeten, G.; Govers, G.; Vanmaercke, M.; Van Rompaey, A.; Arabkhedri, M.; Boix-Fayos, C. Predicting Soil Erosion and Sediment Yield at Regional Scales: Where Do We Stand? Earth-Sci. Rev. 2013, 127, 16–29. [Google Scholar] [CrossRef]
  24. Borrelli, P.; Alewell, C.; Alvarez, P.; Anache, J.A.A.; Baartman, J.; Ballabio, C.; Bezak, N.; Biddoccu, M.; Cerdà, A.; Chalise, D.; et al. Soil Erosion Modelling: A Global Review and Statistical Analysis. Sci. Total Environ. 2021, 780, 146494. [Google Scholar] [CrossRef] [PubMed]
  25. Van Rompaey, A.; Krasa, J.; Dostal, T. Modelling the Impact of Land Cover Changes in the Czech Republic on Sediment Delivery. Land Use Policy 2007, 24, 576–583. [Google Scholar] [CrossRef]
  26. Jáchymová, B.; Krása, J.; Dostál, T.; Bauer, M. Can Lumped Characteristics of a Contributing Area Provide Risk Definition of Sediment Flux? Water 2020, 12, 1787. [Google Scholar] [CrossRef]
  27. Krasa, J.; Dostal, T.; Jachymova, B.; Bauer, M.; Devaty, J. Soil Erosion as a Source of Sediment and Phosphorus in Rivers and Reservoirs—Watershed Analyses Using WaTEM/SEDEM. Environ. Res. 2019, 171, 470–483. [Google Scholar] [CrossRef]
  28. Liu, Y.; Fu, B. Assessing Sedimentological Connectivity Using WATEM/SEDEM Model in a Hilly and Gully Watershed of the Loess Plateau, China. Ecol. Indic. 2016, 66, 259–268. [Google Scholar] [CrossRef]
  29. Alatorre, L.C.; Beguería, S.; García-Ruiz, J.M. Regional Scale Modeling of Hillslope Sediment Delivery: A Case Study in the Barasona Reservoir Watershed (Spain) Using WATEM/SEDEM. J. Hydrol. 2010, 391, 109–123. [Google Scholar] [CrossRef]
  30. Wickham, J.; Stehman, S.V.; Sorenson, D.G.; Gass, L.; Dewitz, J.A. Thematic Accuracy Assessment of the NLCD 2016 Land Cover for the Conterminous United States. Remote Sens. Environ. 2021, 257, 112357. [Google Scholar] [CrossRef]
  31. Wickham, J.; Stehman, S.V.; Sorenson, D.G.; Gass, L.; Dewitz, J.A. Thematic Accuracy Assessment of the NLCD 2019 Land Cover for the Conterminous United States. GISci. Remote Sens. 2023, 60, 2181143. [Google Scholar] [CrossRef]
  32. Büttner, G. CORINE Land Cover and Land Cover Change Products. Remote Sens. Digit. Image Process. 2014, 18, 55–74. [Google Scholar] [CrossRef]
  33. Zanaga, D.; Van De Kerchove, R.; Daems, D.; De Keersmaecker, W.; Brockmann, C.; Kirches, G.; Wevers, J.; Cartus, O.; Santoro, M.; Fritz, S.; et al. ESA WorldCover 10 m 2021 V200. 2022. Available online: https://zenodo.org/records/7254221 (accessed on 1 January 2024).
  34. Wallace, C.W.; McCarty, G.; Lee, S.; Brooks, R.P.; Veith, T.L.; Kleinman, P.J.A.; Sadeghi, A.M. Evaluating Concentrated Flowpaths in Riparian Forest Buffer Contributing Areas Using LiDAR Imagery and Topographic Metrics. Remote Sens. 2018, 10, 614. [Google Scholar] [CrossRef]
  35. Huangfu, Y.; Essington, M.E.; Hawkins, S.A.; Walker, F.R.; Schwartz, J.S.; Layton, A.C. Testing the Sediment Fingerprinting Technique Using the SIAR Model with Artificial Sediment Mixtures. J. Soils Sediments 2020, 20, 1771–1781. [Google Scholar] [CrossRef]
  36. Wischmeier, W.; Smith, D. Predicting Rainfall Erosion Losses: A Guide to Conservation Planning; US Department of Agriculture: Washington, DC, USA, 1978; ISBN 001-000-03903-2.
  37. Desmet, P.J.J.; Govers, G. A GIS Procedure for Automatically Calculating the USLE LS Factor on Topographically Complex Landscape Units. J. Soil Water Conserv. 1996, 51, 427–433. [Google Scholar]
  38. Stoker, J.; Miller, B. The Accuracy and Consistency of 3D Elevation Program Data: A Systematic Analysis. Remote Sens. 2022, 14, 940. [Google Scholar] [CrossRef]
  39. Hoomehr, S.; Schwartz, J.S.; Yoder, D.C. Potential Changes in Rainfall Erosivity under GCM Climate Change Scenarios for the Southern Appalachian Region, USA. Catena 2016, 136, 141–151. [Google Scholar] [CrossRef]
  40. Renard, K.G.; Foster, G.R.; Weesies, G.A.; McCool, D.K. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE); US Department of Agriculture: Washington, DC, USA, 1997; ISBN 0-16-048938-5.
  41. McCool, D.K.; Foster, G.R.; Mutchler, C.K.; Meyer, L.D. Revised Slope Length Factor for the Universal Soil Loss Equation. Trans. ASAE 1989, 32, 1571–1576. [Google Scholar] [CrossRef]
  42. OpenStreetMap Contributors Planet Dump. 2017. Available online: https://planet.osm.org (accessed on 1 January 2024).
  43. Williamson, T.N.; Dobrowolski, E.G.; Kreiling, R.M. Phosphorus Sources, Forms, and Abundance as a Function of Streamflow and Field Conditions in a Maumee River Tributary, 2016–2019. J. Environ. Qual. 2023, 52, 492–507. [Google Scholar] [CrossRef] [PubMed]
  44. Hagen, J.; Walker, F. Oostanaula Creek Watershed Restoration Plan; University of Tennessee Extension: Knoxville, TN, USA, 2007. [Google Scholar]
Sustainability 16 02332 g001
Figure 1. Map of the Oostanaula Creek watershed study site location in East Tennessee, USA.
Figure 1. Map of the Oostanaula Creek watershed study site location in East Tennessee, USA.
Sustainability 16 02332 g001
Sustainability 16 02332 g002
Figure 2. Land use spatial resolution differences NLCD (left), ESA (right).
Figure 2. Land use spatial resolution differences NLCD (left), ESA (right).
Sustainability 16 02332 g002
Sustainability 16 02332 g003
Figure 3. Land use cover data source comparison for the entire Oostanaula Creek watershed per NLCD and ESA datasets.
Figure 3. Land use cover data source comparison for the entire Oostanaula Creek watershed per NLCD and ESA datasets.
Sustainability 16 02332 g003
Sustainability 16 02332 g004
Figure 4. Visual comparison of land use differences—Oostanaula-Top, the study’s sub-catchment, located in the upper part of the Oostanaula Creek watershed.
Figure 4. Visual comparison of land use differences—Oostanaula-Top, the study’s sub-catchment, located in the upper part of the Oostanaula Creek watershed.
Sustainability 16 02332 g004
Sustainability 16 02332 g005
Figure 5. Land use differences based on data source for the Oostanaula-Top sub-catchment.
Figure 5. Land use differences based on data source for the Oostanaula-Top sub-catchment.
Sustainability 16 02332 g005
Sustainability 16 02332 g006
Figure 6. Stream network field correction.
Figure 6. Stream network field correction.
Sustainability 16 02332 g006
Sustainability 16 02332 g007
Figure 7. UAV-based digital surface model for the first river section of the Oostanaula Creek formed in arable field.
Figure 7. UAV-based digital surface model for the first river section of the Oostanaula Creek formed in arable field.
Sustainability 16 02332 g007
Sustainability 16 02332 g008
Figure 8. Defining the stream cross-section by LIDAR technology in 3 steps: (A) RGB 3D model of the profile; (B) tiled model from the processed point cloud; (C) final cross-section (cross-section 400 m above the outlet of Oostanaula-Top catchment).
Figure 8. Defining the stream cross-section by LIDAR technology in 3 steps: (A) RGB 3D model of the profile; (B) tiled model from the processed point cloud; (C) final cross-section (cross-section 400 m above the outlet of Oostanaula-Top catchment).
Sustainability 16 02332 g008
Sustainability 16 02332 g009
Figure 9. Comparison of NLCD and ESA dataset locations of critical drainage network points within the Oostanaula Creek watershed.
Figure 9. Comparison of NLCD and ESA dataset locations of critical drainage network points within the Oostanaula Creek watershed.
Sustainability 16 02332 g009
Sustainability 16 02332 g010
Figure 10. Identification of four stream sections for further study comparisons within the Oostanaula-Top sub-catchment.
Figure 10. Identification of four stream sections for further study comparisons within the Oostanaula-Top sub-catchment.
Sustainability 16 02332 g010
Sustainability 16 02332 g011
Figure 11. Comparison of NLCD, ESA, and FIELD locations of critical stream segments in the stream network in the Oostanaula-Top catchment.
Figure 11. Comparison of NLCD, ESA, and FIELD locations of critical stream segments in the stream network in the Oostanaula-Top catchment.
Sustainability 16 02332 g011
Sustainability 16 02332 g012
Figure 12. Characteristic cross-sections of stream channels among different reach sections.
Figure 12. Characteristic cross-sections of stream channels among different reach sections.
Sustainability 16 02332 g012
Sustainability 16 02332 g013
Figure 13. NLCD land use on orthophoto (left) and ESA dataset land use (right) influencing model results (lower left and right).
Figure 13. NLCD land use on orthophoto (left) and ESA dataset land use (right) influencing model results (lower left and right).
Sustainability 16 02332 g013
Sustainability 16 02332 g014
Figure 14. Time series of orthophoto images for watershed topology validation.
Figure 14. Time series of orthophoto images for watershed topology validation.
Sustainability 16 02332 g014
Table 1. Table of NLCD and ESA categories definition in study.
Table 1. Table of NLCD and ESA categories definition in study.
ESANLCDCase
Built-up1.44%Developed, Open Space/Developed, Low Intensity/Developed, Medium Intensity/Developed, High Intensity11.55%Urban
Permanent water bodies0.10%Open Water/Woody Wetlands/Emergent Herbaceous Wetlands0.26%Water
Cropland5.30%Cultivated Crops3.81%Arable
Bare or sparse vegetation0.60%Barren Land0.03%Arable
Tree Cover61.87%Deciduous Forest/Evergreen Forest/Mixed Forest45.97%Forest
Grassland30.69%Shrub/Scrub/Herbaceous/Hay/Pasture38.38%Grass
Table 2. Resulting sediment erosion/deposition and net transport compared between the NLCD and ESA land use datasets.
Table 2. Resulting sediment erosion/deposition and net transport compared between the NLCD and ESA land use datasets.
NLCDESA
t/year
Erosion56,40077,900
Deposition35,70053,500
River transport20,70024,400
Table 3. Summary of sediment erosion/deposition and net transport for different land use scenarios.
Table 3. Summary of sediment erosion/deposition and net transport for different land use scenarios.
Sediment TransportRivers
CountSumSum/kmMinMaxMeanStdLength
[t] [km]
NLCD5012,95621792123725920758.4
ESA5016,019314114126832024851.1
Match NLCD3310,62427798123732222938.3
Match ESA11,6133031251268352265
Mismatch NLCD172332116922891374820.1
Mismatch ESA17440634411480625920412.8
Total—NLCD53120,74364012373997323.4
Total—ESA53124,420760126846118
Table 4. Resulting sediment erosion/deposition and net transport in Oostanaula-Top sub-catchment for three different land use datasets (NLCD, ESA, and FIELD).
Table 4. Resulting sediment erosion/deposition and net transport in Oostanaula-Top sub-catchment for three different land use datasets (NLCD, ESA, and FIELD).
NLCD ESA FIELD
t/year
Erosion4600602011,540
Deposition234029807420
River transport226030404120
Table 5. Table with details of stream section calculation.
Table 5. Table with details of stream section calculation.
NLCDESAFIELDChannel Volume
t/yearm3
No. 1 (red *)1238130591145
No. 2 (blue *)528789166213,770
No. 3 (yellow *)40981421494120
No. 4 (green *)881321098550
* colors linked to Figure 10.
Table 6. Risky location sub-selection comparison in Oostanaula-Top sub-catchment.
Table 6. Risky location sub-selection comparison in Oostanaula-Top sub-catchment.
NLCDESAFIELD
Sediment transport (t/year)187323733084
Length chosen (km)5.895.833.34
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

Bauer, M.; Jáchymová, B.; Krása, J.; Bynum, K.; Schwartz, J.S.; Dostál, T. Sustainable Watershed Management: Assessing the Impact of Data Precision on Sediment Transport in the Oostanaula Creek Watershed, Tennessee. Sustainability 2024, 16, 2332. https://doi.org/10.3390/su16062332

AMA Style

Bauer M, Jáchymová B, Krása J, Bynum K, Schwartz JS, Dostál T. Sustainable Watershed Management: Assessing the Impact of Data Precision on Sediment Transport in the Oostanaula Creek Watershed, Tennessee. Sustainability. 2024; 16(6):2332. https://doi.org/10.3390/su16062332

Chicago/Turabian Style

Bauer, Miroslav, Barbora Jáchymová, Josef Krása, Karina Bynum, John S. Schwartz, and Tomáš Dostál. 2024. "Sustainable Watershed Management: Assessing the Impact of Data Precision on Sediment Transport in the Oostanaula Creek Watershed, Tennessee" Sustainability 16, no. 6: 2332. https://doi.org/10.3390/su16062332

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