Embedded Files
Imagine venturing into the narrow, winding corridors of a slot canyon, with towering sandstone walls that seem to touch the sky. These natural wonders, unique to the Southwestern United States, offer unparalleled beauty and adventure. However, beneath their serene and majestic exterior lies a hidden danger: the risk of sudden and devastating flash floods.
Slot canyons, with their steep, narrow passages, are particularly susceptible to flash floods. A storm miles away can send a torrent of water through these canyons with little warning, turning a calm adventure into a life-threatening situation in an instant. Understanding and predicting these flash floods is critical for ensuring the safety of canyoneers.
This project uses advanced spatial analysis tools within ArcGIS to better define the risk of flash floods in select slot canyons in Southeastern Utah. By analyzing watershed characteristics and environmental factors, I hope to provide a clearer picture of the flash flood risks associated with some of the most popular slot canyons in Utah.
My primary question is: Can ArcGIS spatial and imagery analyst tools be used to better define the risk of flash floods for canyoneering in select slot canyons in Southeastern Utah by analyzing watershed characteristics and environmental factors?
Overview of Slot Canyons
Slot canyons are narrow, deep gorges typically formed in soft rock like sandstone through the process of erosion by water. These unique geological formations are characterized by their stunning, narrow passageways, often just a few feet wide, with walls that can tower hundreds of feet above the canyon floor.
Characteristics of Slot Canyons
Narrow Passages: The hallmark of slot canyons is their narrow width, which can range from a few inches to several feet. This creates a dramatic, claustrophobic effect as the walls rise steeply on either side.
High Walls: The walls of slot canyons can reach impressive heights, sometimes over 300 feet, creating a beautiful interplay of light and shadow as sunlight filters down into the depths.
Smooth, Sculpted Walls: The walls of slot canyons are often smooth and sculpted, shaped by the flow of water over thousands or even millions of years. This erosion creates intricate patterns and textures that are unique to each canyon.
Seasonal Water Flow: Slot canyons are formed by the seasonal flow of water, which can range from small trickles to powerful torrents. These flows carve out the canyon over time, deepening and widening it.
Location and Significance in Southeastern Utah
Southeastern Utah is renowned for its concentration of slot canyons, making it a prime destination for outdoor enthusiasts, photographers, and adventurers. Some of the notable slot canyons in this region include:
Leprechaun Canyon: Known for its narrow, winding passages that make for a dark and claustrophobic journey. The tight passages make you feel as if you're stuck, adding to the thrill and challenge.
Smiling Cricket Canyon: Arguably the most difficult slot canyon on the Colorado Plateau. It takes two whole days to traverse and features one of the largest keeper potholes known, testing the limits of even the most experienced canyoneers.
Black Box Canyon: Recognizable for its massive canyon that hosts lots of water, requiring much swimming in cold and dark narrow canyons. The extensive water-filled sections and the darkness add to the canyon's mystique and challenge.
These slot canyons hold significant ecological and recreational value:
Ecological Significance: Slot canyons provide unique habitats for a variety of plant and animal species adapted to the harsh, arid environment.
Recreational Significance: They offer unparalleled opportunities for hiking, canyoneering, and photography, attracting hundreds of visitors each year.
Flash Flooding
Causes of Flash Flooding
Intense Rainfall: Heavy downpours, especially in areas with steep terrain or impervious surfaces, can lead to rapid water accumulation.
Rapid Snowmelt: Sudden warming can cause large amounts of snow to melt quickly, overwhelming rivers and streams.
Development: Increased development reduces the land's natural ability to absorb water, exacerbating flood risks.
Impact on Slot Canyons
Slot canyons, with their narrow and steep-walled passages, are particularly vulnerable to flash floods. The confined space can channel water with tremendous force, creating dangerous conditions for hikers and canyoneers. Even a small amount of rain upstream can lead to a life-threatening situation within these canyons, making awareness and preparedness crucial.
Purpose of the Study
Purpose of the Study
The primary objective of this study is to assess and map the flash flood risk for 39 slot canyons in Southeastern Utah. By developing a comprehensive flash flood risk rating, this study aims to provide valuable insights that can enhance public safety and support sustainable outdoor recreation planning.
Objectives
Assess Slot Canyon Characteristics:
Evaluate key variables such as slope, watershed size, soil runoff risk, and typical precipitation intensity for each slot canyon using spatial data and hydrological models.
Use models to estimate the magnitude of peak flows during a 'design storm' typical for the area.
Create Risk Maps:
Develop detailed maps that visually represent the flash flood risk levels for each slot canyon based on the assessed characteristics and model predictions.
Rank the canyons according to their flash flood risk..
Methodology
Methodology
The methodology for this study involves a comprehensive approach that integrates data collection, risk assessment criteria, spatial analysis, Data Normalization, and hydrological modeling to evaluate the flash flood risk in 39 slot canyons across Southeastern Utah. The process is outlined in the following steps, with a focus on four key variables associated with flash flood risk: Watershed Size, Precipitation Intensity, Runoff Potential, and Slope.
Data Collection
Watershed and slope data were both derived from high-resolution Digital Elevation Models (DEMs) provided by the USGS 3D Elevation Program (3DEP) using ArcGIS Pro. Four 10-meter resolution DEMs were downloaded to capture most of Southeastern Utah.
Precipitation intensity data were sourced from NOAA's Precipitation Frequency tables. These tables are part of NOAA Atlas 14 and provide estimates of the frequency of different precipitation intensities over specified durations, such as 5 minutes, 15 minutes, 1 hour, etc. The estimates are derived from a frequency analysis of historical precipitation data and are expressed with confidence intervals to account for variability and uncertainty.
For this assessment, a design storm equivalent to a 25-year recurrence interval was used. The design storm calculation was based on the total precipitation depth occurring within a 15-minute timeframe. This means that the intensity and volume of rain used in the model represents a storm that statistically occurs once every 25 years.
Runoff potential and soil data were collected from the USDA Natural Resources Conservation Service (NRCS). This information was important for assessing soil runoff risk and hydrological behavior. The hydrologic soil groups (HSGs) are classified into four categories (A, B, C, and D) based on the soil's infiltration rate when thoroughly wetted:
Group A: Soils with the highest infiltration rates, assigned a runoff rating of 0.
Group B: Soils with moderately high infiltration rates, assigned a runoff rating of 25.
Group C: Soils with moderately low infiltration rates, assigned a runoff rating of 50.
Group D: Soils with the lowest infiltration rates, assigned a runoff rating of 100.
If no HSG data were present, a manual determination was made using satellite imagery and soil texture/bedrock information to ensure accurate classification.
Spatial Analysis
Spatial Analysis
Spatial analysis was conducted to examine the physical characteristics of the slot canyons and their surrounding watersheds. This process involved the use of ArcGIS Pro for analyzing DEMs to extract important information such as slope, aspect, and watershed boundaries. The following tools and processes were utilized:
DEM Analysis
Mosaic to New Raster: Combined multiple 10-meter DEM raster datasets into a single, continuous raster for analysis.
Fill DEM: Removed any sinks in the DEM to ensure accurate flow direction and accumulation analysis.
Flow Direction: Determined the direction of water flow across the terrain.
Flow Accumulation: Identified areas where water accumulates, which is important for delineating watersheds and understanding runoff patterns.
Watershed: Delineated watershed boundaries to understand the contributing areas for each slot canyon.
Extract by Mask: Extracted flow accumulation and other data of interest from the larger raster datasets using a 'watershed mask'.
Stream Order: Classified stream segments based on their hierarchical position in the drainage network.
Slope: Calculated the percent slope of the terrain from the DEM.
Aspect: Determined the compass direction that each slope faces.
Zonal Statistics: Summarized raster data within the boundaries of each watershed to derive key statistics such as median slope and aspect.
Additional Analysis Tools
Reclassify: Standardized the scales of various raster datasets to make them comparable.
Clip: Clipped the soils data to match the watershed boundaries.
IDW (Inverse Distance Weighting): Used to interpolate precipitation data across all watersheds in the study area.
Spatial Join: Combined the clipped soils layer and precipitation data with the slot canyon layer based on spatial relationships.
Raster Calculator: Used the raster calculator to 1) normalize data among the four variables and 2) produced a composite risk map with normalized variable data and weighted values.
Risk Assessment Criteria
Risk Assessment Criteria
The assessment of flash flood risk in the slot canyons was based on four key variables: Watershed Size (Acres), Precipitation Intensity, Runoff Potential, and Slope. These variables were selected for their significant influence on flash flood dynamics and were incorporated into a composite risk model through a weighted approach.
To determine the appropriate weights for each variable, a Pairwise Comparison Matrix (PCM) was utilized. The PCM is a tool used in Multi-Criteria Decision Analysis (MCDA) to assign relative importance to each criterion based on expert judgment and empirical evidence. The weights were assigned as follows:
Watershed Size (Acres): 0.97
Precipitation Intensity: 0.01
Runoff Potential: 0.01
Slope: 0.01
The rationale for these weights is rooted in the physical characteristics of the study area and the expected behavior of flash floods. Watershed size was given the highest weight because larger watersheds tend to collect more water, significantly impacting the volume of runoff and the potential for flash flooding. The other three variables, although important, were assigned lower weights because the characteristics of the canyons do not vary significantly in these aspects. The primary factor that substantially influences flash flooding is the size of the watershed, which was therefore given the highest weight.
The image on the right shows the Pairwise Comparison Matrix and the associated formulas to develop weights for each of the four variables.
Normalization is an important step in the process of evaluating flash flood risk as it allows the four different variables, each with its own range of values, to be compared on a common scale from 0 to 1. By normalizing these data, we ensure that each criterion contributes proportionally to the final composite risk score, regardless of its original units or magnitude.
The formulas in the image on the left were used to normalize the four chosen variables, ensuring each criterion contributes proportionally to the final composite risk score, ranging uniformly between 0 and 1.
Composite Risk Calculation
The normalized values for each of the four variables within each of the 39 watersheds were then combined using the raster calculator in ArcGIS Pro, applying the formula:
Composite Risk = ("Runoff Potential" ∗ 0.01) + ("Acres" ∗ 0.97) + ("Slope" ∗ 0.01) +("Precipitation Intensity" ∗ 0.01)
The composite risk calculation provided an integrated assessment of flash flood risk for each of the 39 slot canyons, highlighting the areas most vulnerable to potential flash flooding based on the combined influence of watershed size, precipitation intensity, runoff potential, and slope.
Hydrological Modeling
Hydrological Modeling
Hydrological modeling was conducted using the FS WEPP (Watershed Erosion Prediction Project) model. This step was an additional process, separate from the composite risk mapping project, aimed at acquiring estimates of peak flows in cubic feet per second that could be encountered in each of the slot canyons during the specific design storm previously mentioned.
Workflow of the FS WEPP Model:
Channel Delineation: The process begins with defining the main channels within the watershed. This step is important for understanding how water will flow through the landscape during a storm event.
Pour Point Selection: Next, pour points are selected. These are the points where water is expected to exit the watershed or slot canyon.
Watershed Delineation: The watershed is divided into smaller subcatchments. This step helps in detailed analysis of how different parts of the watershed contribute to runoff and peak flow.
Land Use Options: Land use data is incorporated into the model. FS WEPP pulls from an internal database that precisely accounts for every different land use type across the watershed. This data helps in understanding how different land uses affect runoff and soil erosion.
Soil Options: Soil data, also from an internal database, includes detailed information on soil types and their hydrologic properties.
Climate Options: Finally, the same design storm used for initial assessments is input into the model. This ensures consistency in the analysis and helps in predicting the runoff and peak flows accurately.
These models have an accuracy of +/- 50%, therefore the model was only used for estimations of peak flow and not the final rankings.
Results
Results
Composite Risk Assessment
The composite risk assessment for flash floods in the 39 slot canyons was calculated using the weighted combination of the four normalized variables: Runoff Potential, Watershed Size (Acres), Slope, and Precipitation Intensity. The resulting composite risk scores provided a ranked list of canyons based on their potential vulnerability to flash floods.
The final composite risk values ranged from 0.003 to 0.985, with higher values indicating greater risk. Notably, the canyons with the highest composite risk scores included Black Box, Lower Eardley, and Buck, highlighting their significant vulnerability to flash flooding. On the other hand, canyons such as Angel Slot and Bow & Arrow exhibited lower composite risk scores, indicating relatively lower flash flood risk.
Peak Flow Estimates
The FS WEPP hydrological modeling provided estimates of peak flows for each slot canyon during the specified design storm. These estimates are important for understanding the potential volume of water that could flow through the canyons, offering insights for risk assessment and management.
Key findings from the peak flow estimates include:
Black Box: The highest estimated peak flow at 1500 cubic feet per second (cfs), indicating a significant risk due to the large volume of water.
Lower Eardley: An estimated peak flow of 122 cfs, highlighting its high vulnerability.
Buck: An estimated peak flow of 101 cfs, contributing to its high-risk ranking.
Summary of Key Findings for both Assessments
High-Risk Canyons: Black Box, Lower Eardley, and Buck are among the highest-risk canyons due to their high composite risk scores and significant peak flow estimates.
Lower-Risk Canyons: Angel Slot, Angel Cove, and Bow & Arrow are among the lower-risk canyons with relatively lower composite risk scores and peak flow estimates.
Uncertainties
In this study, several uncertainties could affect the accuracy and reliability of the results:
Data Accuracy: The accuracy of soil data can influence the outcomes of the spatial and hydrological analyses. There were gaps of missing information in the soil data, which required manual determinations of runoff potential. These manual estimates introduce a level of subjectivity and potential error into the analysis.
Model Assumptions: The FS WEPP model and other hydrological models used in this study rely on certain assumptions about soil properties, vegetation cover, and hydrological processes. These models are generally accurate within +/- 50%, but deviations from these assumptions in real-world scenarios could lead to discrepancies between predicted and actual peak flows.
Temporal Variability: Flash flood risks can vary significantly with changes in weather patterns and land use over time. This study provides a snapshot based on current conditions, which might not fully represent future risks due to climate change or other dynamic factors.
Human Factors: Human activities such as land development, deforestation, and fire management practices can alter watershed characteristics and influence flash flood risks. These factors are difficult to quantify and incorporate accurately into the model.
Page updated
Google Sites
Report abuse