Progress 09/02/14 to 06/30/17
Outputs
Target Audience:Over the course of the project, our target audience included the following: 1) Students, scientists, and professionals in attendance at conferences where we have presented our research findings, or who are likely to read our published journal articles; 2) landowners affected by the High Park Fire in our study areas; and 3) local and regional goverment agency employees and utility operators who attended a High Park Fire post-fire research and restoration workshop where our findings were presented. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?This project has provided significant training opportunities for graduate students. These students have had the opportunity to work closely with faculty on the collection and interpretation of data, and they have gained experience communicating their research findings through conference presentations and publications. The LiDAR differencing has been technically challenging and required interaction with remote sensing experts. This interaction -- primarily with researchers at NEON, the National Ecological Observatory Network, as well as academics whose primary research focus is on high-resolution topography and remote sensing -- has provided strong professional development opportunities for the students working on the project. How have the results been disseminated to communities of interest?Our results have primarily been disseminated through presentations at professional conferences such as the Fall Meeting of the American Geophysical Union, the European Geosciences Union General Assembly, the Geological Society of America Annual Meeting, and the CSU Hydrology Days meeting, as well as through several peer-reviewed publications. We also have presented our findings at workshops and community meetings. In 2015, we hosted a community meeting in the Poudre Park Community Center where we presented research on the High Park Fire and its effects to a group of landowners who were affected by the fire. In 2016 we presented our work at the High Park Fire Post Fire Science Workshop, hosted by the Coalition for the Poudre River Watershed, which was attended by federal and univeristy scientists, and utility managers from around the Front Range. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Major findings for this project were accomplished through 1) extensive field data collection over the course of several years, 2) collection and analysis of repeat airborne LiDAR topographic data of our study watersheds, and 3) multidimensional hydraulic modeling to estimate the magnitude and geomorphic effectiveness of large floods that occurred in 2012 and 2013. Most of the field data consisted of surveys (primarily collected with highly accurate RTK-GNSS equipment) of 10 cross sections in the Hill Gulch watershed and 11 cross sections in the Skin Gulch watershed, which both are tributaries to the Cache la Poudre River and were both burned at moderate to high severity during the 2012 High Park Fire. Each cross section was surveyed 16 to 23 times, generally between seasons and after summer thunderstorms, and with each cross sectional survey a longitudinal profile survey was also collected. This dataset is rich with information and without equal in post-fire geomorphic research. We have devised methods to normalize cross section and longitudinal profile data from different surveys, which allows us to compute local changes in elevation and cross-sectional area from survey to survey. This analysis has led us to split the time since the fire into two distinct periods: before the 2013 Front Range floods and after them. Before the 2013 floods, our study watersheds (especially Skin Gulch) had very active geomorphic changes occurring from storm to storm and from season to season. During this time there was significant variability within cross sections and longitudinal profiles. The 2013 flood, however, caused extensive degradation throughout the channel networks in our study watersheds. Our interpretation of this event and the subsequent period is that it washed out most of the post-fire hillslope sediment that had accumulated in the channels, armored the channel beds, and greatly reduced the overall geomorphic sensitivity of the channels. After the storm and flood, the cross section and longituindal profiles exhibited very little variabilitiy from one survey to the next, even after somewhat large storms. We have used these observations to develop a new conceptual framework for the complex effects of fires and floods on the landscape. A major component of this research has involved analyzing a series of high-resolution topographic datasets acquired over the High Park Fire burn area via airborne LiDAR. These are 1-meter resolution digital elevation models derived from point clouds collected from a laser scanner mounted to an airplane, and at this resolution the topographic data captures features relevant to important geomorphic processes. These datasets were collected in fall 2012, summer 2013, fall 2013, summer 2014, and summer 2015. We developed techniques to co-register all of these datasets together, and once registered we differenced consecutive datasets to determine spatial patterns and overall volumes of erosion and deposition. We found that there was substantial deposition from 2012 to summer 2013, reflecting a large increase in sediment delivered to channels from burned hillslopes due to decreased erosion thresholds after the fire. The flood in September 2013 produced very significant erosion throughout both watersheds, flushing most of the fire-generated sediment from the watersheds. LiDAR differences for datasets collected after the 2013 flood suggest that both watersheds are essentially in equilbrium, with similar volumes of erosion and deposition. One of our goals has been to use these spatial patterns of erosion and deposition, along with the LiDAR topography itself, to better understand geomorphic controls on sedimentation patterns and to make progress toward better predictions of sediment delivery from watersheds. To this end, we have developed algorithms to delineate the valley bottom from LiDAR topography, divide the valley bottom into 50-m segments, and for each segment compute a suite of morphometrics including valley slope, valley width, contributing area and channel confinement. We have done ordinary and multiple linear regression of these metrics against the erosion and deposition for each 50-m segment computed from the LiDAR differencing and found slope and valley width to have the strongest correlation with erosion/deposition. These relationships are noisy, so we are presently trying to compute streamwise gradients in the morphometrics for use in a similar analysis, but for now, these results are somewhat encouraging as they suggest that topography itself can provide some information on expected patterns of erosion and deposition. Because our study has been punctuated by two extreme floods -- one in Skin Gulch in 2012 shortly after the fire was contained, and the 2013 Front Range flood that affected an area from Boulder to Wyoming -- we have put significant effort toward understanding those two events, including estimating their peak flow magnitudes. This work involved collecting high water marks after each flood, and using those with LiDAR topography in a two-dimensional hydraulic model to simulate floods of different magnitudes. After extensive analysis, we concluded that the 2012 storm produced a flood in Skin Gulch much larger than the longer-duration 2013 flood; in fact, the 2012 flood plots on the envelope curve of the largest rainfall-generated floods per unit area ever recorded in the United States. Given that flood resulted from a fairly typical thunderstorm that happened to be located over a high-severity burn, this event provides a compelling example of the flooding dangers posed by wildfire.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2017
Citation:
Brogan DJ, PA Nelson, and LH MacDonald, 2017, Reconstructing extreme post-wildfire floods: a comparison of convective and mesoscale events, Earth Surface Processes and Landforms, doi: 10.1002/esp.4194.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2017
Citation:
Brogan, DJ, PA Nelson, LH MacDonald, and JA Morgan, 2017, Geomorphic complexity of sequential fire and floods in mountain watersheds, Abstract EP51E-06 presented at the 2017 Fall Meeting, AGU, New Orleans, LA, 11-15 Dec.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2017
Citation:
Kampf, S, L MacDonald, F Saavedra, C Wilson, S Schmeer, D Brogan, P Nelson, and B Gannon, 2017, Erosion and sediment delivery to streams following wildfire: Processes and predictions, UCOWR/NIWR Annual Conference, Fort Collins, CO, 13-15 June.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2017
Citation:
Cotrufo, MF, C Boot, S Kampf, L MacDonald, P Nelson, and E Hall, 2017, Pyrogenic carbon redistribution from hillslopes to stream corridors following a large montane wildfire, Geophysical Research Abstracts, 19, EGU2017-18270.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2017
Citation:
Brogan, DJ, PA Nelson, LH MacDonald, and JA Morgan, 2017, How disturbing: The complications of sequential fire and floods in mountain catchments, Hydrology Days 2017, Fort Collins, CO, 20-22 March.
|
Progress 10/01/15 to 09/30/16
Outputs
Target Audience:During this reporting period, our target audiences have included students, scientists, and professionals in attendance at presentations and conferences where our research findings have been presented, or who are likely to read our published journal articles (in the international journals Global Biogeochemical Cycles, and Geomorphology). We also conducted an outreach event attended by many of the landowners affected by the High Park Fire in our study areas where we discussed our research findings with them. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?A graduate student has been primarily responsible for field data collection, data analysis, and presentation of research results at conferences and as scientific articles. Working closely with faculty, thisstudent has gained valuable experience in all of these areas of scientific research. How have the results been disseminated to communities of interest?We hosted a community meeting on October 15, 2015 at the Poudre Park Community Center. This meeting was attended by 20-30 people, many of whom are residents whose property was affected by the High Park Fire. Several graduate students and faculty presented research findings from studies conducted in the High Park Fire burn area, and the researchers answered residents' questions about fire effects and water quality and flooding issues. Two peer-reviewed publications were published during the reporting period, and a third manuscript is in review. Additionally, results have been presented at major international scientific meetings: the European Geosciences Union General Assembly, the Geological Society of America Annual Meeting, and the American Geophysical Union Fall Meeting. What do you plan to do during the next reporting period to accomplish the goals?Over the next year, we will complete the analysis of the Airborne lidar data to compute spatial and temporal patterns of erosion and deposition. When that is complete, multivariate statistics will be used to determine to what extent the morphometric measurements, precipitation data, predicted sediment production, and other characteristics (e.g., percent bare soil) can be used to predict the changes in sediment deposition and erosion for each stream segment for each time period. The survey data will be analyzed to quantify topographic changes through the channel network following a large wildfire and major flood. While the cross-sectional suveys have been analyzed, each section has a longitudinal profile associated with it, and these data still need to be analyzed to investigate changes in slope, knickpoint migration, and changes in the location of riffles and pools. When that analysis is completed, the field data will be summarized and presented as a stand-alone journal article. The terrestrial laser scanner (TLS) data still need to be processed. These datasets will be co-registered using the iterative closest point algorithm, then differenced using the multiscale model-to-model cloud comparison algorithm to quantify topographic changes at the reach and sub-reach scale.
Impacts
What was accomplished under these goals?
During the past year we made significant progress in our efforts to estimate and predict post-fire sediment delivery at the watershed scale. More specifically, we have completed our field data collection, and this includes 16-23 repeat surveys at the ten cross-sections in Skin Gulch and 11 cross-sections in Hill Gulch, the associated longitudinal profiles, ten terrestrial laser scanning (TLS) surveys, and over 9,000 photographs. The analysis of the cross-section data is completed, and these data have been used to derive the changes in mean bed elevation, changes in the active channel area, and the thalweg elevations and locations. The longitudinal profiles have been plotted, but more analysis is needed to quantify and analyze the changes over time. The TLS survey data still have to be analyzed. Much of the effort over the reporting periodwas devoted to analyzing the five airborne lidar (ALS) datasets that span the period from October 2012 through summer 2015. We have experimented with many different point cloud and raster differencing methodologies, and have settled upon differencing high-resolution grids, which enables us to calculate volume differences along with uncertainties. One of our major goals is to relate the spatial and temporal patterns of these erosional and depositional volumes to watershed characteristics, so we have developed Python scripts which,in conjunction with the recently developed FluvialCorridor valley extraction tool, identify and compute valley bottom characteristics (valley width, valley slope, contributing area) for every 50-m segment of the drainage network. We also have obtained the Cheyenne, WY Doppler radar data from 11 October 2012 through 2 October 2015. These data have been used to calculate daily precipitation totals and maximum 30-minute intensities for 0.5 km2 pixels over the entire High Park Fire burn area. Taken together, we will use the ALS differencing volumes, precipitation data, burn severity data, and the morphometric data topredict segment- and watershed-scale sediment production and delivery.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2016
Citation:
Cotrufo, MF, CM Boot, S Kampf, PA Nelson, DJ Brogan, T Covino, ML Haddix, LH MacDonald, S Rathburn, S Ryan-Burkett, S Schmeer, and E Hall, 2016, Redistribution of pyrogenic carbon from hillslopes to stream corridors following a large montane wildfire, Global Biogeochemical Cycles, 30, 1348-1355, doi: 10.1002/2016GB005467.
- Type:
Journal Articles
Status:
Published
Year Published:
2016
Citation:
Kampf, SK, DJ Brogan, S Schmeer, LH MacDonald, and PA Nelson, 2016, How do geomorphic effects of rainfall vary with storm type and spatial scale in a post-fire landscape? Geomorphology, 273, 39-51, doi: 10.1016/j.geomorph.2016.08.001.
- Type:
Journal Articles
Status:
Under Review
Year Published:
2016
Citation:
Brogan, DJ, PA Nelson, and LH MacDonald, Reconstructing extreme post-wildfire floods: a comparison of convective and mesoscale events, submitted to Earth Surface Processes and Landforms.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2015
Citation:
Brogan, DJ, PA Nelson, and LH MacDonald, 2015, Quantifying erosion and deposition patterns using airborne LiDAR following the 2012 High Park Fire and 2013 Colorado Flood, Abstract EP51B-0911 presented at 2015 Fall Meeting, AGU, San Francisco, Calif., 14-18 Dec.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
MacDonald, LH, S Kampf, D Brogan, S Schmeer, and P Nelson, 2016, Comparing and linking post-fire hillslope erosion and channel change for different storm types, Geophysical Research Abstracts, 18, EGU2016.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
Brogan DJ, PA Nelson, and LH MacDonald, 2016, How do watershed characteristics influence post-fire sediment storage and delivery over time?, Paper No. 302-4, GSA Annual Meeting, Denver, Colorado.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
MacDonald LH, JW Wagenbrenner, PR Robichaud, PA Nelson, SK Kampf, and DJ Brogan, 2016, Fires: Pushing the reset button or a flash in the pan?, Abstract H42A-08 presented at 2016 Fall Meeting, AGU, San Francisco, Calif., 12-16 Dec.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2016
Citation:
Brogan DJ, PA Nelson, and LH MacDonald, 2016, How do watershed characteristics and precipitation influence post-wildfire valley sediment storage and delivery over time?, Abstract H43G-1551 presented at 2016 Fall Meeting, AGU, San Francisco, Calif., 12-16 Dec.
|
Progress 10/01/14 to 09/30/15
Outputs
Target Audience:During this reporting period, the primary target audiences have been students, scientists, and professionals in attendance at presentations and conferences where our research findings have been presented. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?This project has provided significant training opportunities forgraduate students. These students have had the opportunity to work closely with faculty on the collection and interpretation of data, and they have gained experience communicating theirresearch findings through conference presentations and publications. The LiDAR differencing has been technically challenging and required interaction with remote sensing experts. This interaction -- primarily with researchers at NEON, the National Ecological Observatory Network, as well as academics whose primary research focus is on high-resolution topography and remote sensing -- has provided strong professional development opportunities for the students working on the project. How have the results been disseminated to communities of interest?Over the past year, results have primarly been disseminated through conference presentations and peer-reviewed publications. Dan Brogan, a graduate student working on the project, presented some of his work at the AGU Hydrology Days meeting in Fort Collins, CO in March 2015, and he intends to present additional work at the AGU Fall Meeting in San Francisco, CA in December 2015. Dan Brogan, Peter Nelson, and Lee MacDonald also submitted a paper to the journal Earth Surface Processes and Landforms, which is currently in revision. What do you plan to do during the next reporting period to accomplish the goals?Over the next year, we intend to continue with the topographicdifferencing work. This will include differencing an additional airborne LiDARdatasetcollected in June 2015. We also will attempt to generate a pre-fire surface digital elevation model using photogrammetry on aerial photos collected in 2008, and use the resulting dataset to better quantify the erosion and deposition between the time of the fire (June 2012) and the first airborne LiDAR data collection (October 2012). Overall, the erosion and deposition patterns from the topographic differencing will be related to watershed characteristics such as patterns of slope and valley confinement, with the goal of producing a relatively simple model that can predictwatershed-scale sediment yield as a function of these topographic characteristics. Such a model would be quite useful in future fires as resource agencies will need to prioritize where to apply post-fire treatments. We also will continue to collect and analyze water level and electrical conductivity data at multiple locations throughout the Skin Gulch watershed, with the goal of better understandingrunoff generation mechanisms and surface/subsurface water exchange, and how geomorphology affects these exchange processes.
Impacts
What was accomplished under these goals?
The major accomplishment achieved during this reporting period involved quantifying patterns of erosion and deposition in our study watersheds for the 2-year period following the 2012 High Park Fire. An improved understanding of how wildfire affects hydrologic response, erosion, and sedimentation and the consequent flooding and downstream sedimentation requires detailed observations of erosion and deposition, which can then be related to landscape features to predict watershed-scale sediment yields after fire. Over the past year, we have made considerable progress in using several high-resolution remote sensing topographic datasets (airborne LiDAR) to quantify erosion and deposition. We have used four LiDAR datasets, collected in October 2012, July 2013, November 2013 and October 2014, along withGPS ground survey data that are regularly collected after every storm, to characterize erosion and deposition in Hill Gulch (HG) and Skin Gulch (SG). The survey data were used to ensure that all datasets are ground-truthed, and we have used advanced techniques to co-register the airborne LiDAR datasets to each other so that uncertainty when comparing sequential datasets is minimized. Differencing high-resolution topographic datasets has been an area of considerable research interest recently, and we are using a state-of-the-art algorithm (the M3C2: Multiscale Model to Model Cloud Comparison) to difference sequential datasets and quantify patterns of erosion and deposition over time.Differencing was performed for the following periods:October 2012 to July 2013, capturing the winter 2012 and spring/summer 2013; July 2013 to November 2013, which captures the effects of the large Colorado Front Range flood of October 2013; and November 2013 to October 2014, capturing the winter of 2013 and spring/summer of 2014. Field observations prior to the July 2012 lidar suggests that very little channel change occurred in Hill Gulch; however, Skin Gulch experienced substantial deposition due to a summer convective thunderstorm just a few days after the fire was fully contained. This variation is believed to help explain why Hill Gulch generally saw more net deposition in the differenced datasets. During the summer of 2013 both watersheds accumulated sediment from the hillslopes due to convective thunderstorms (net deposition of24000 m3 in SG and 50000 m3 in HG). The September 2013 flood resulted in substantial erosion, but the volume of erosion was twice as much in Skin Gulch (104000 m3)than Hill Gulch (50000 m3). Following the September 2013 flood channel change was less substantial, although still depositional, and primarily resulted in a reworking of the exposed sediment in the valley bottom. Although difficult to prove, it is hypothesized that this 'loading the gun' affect in Skin Gulch from the July 2012 flood aided in additional scour to the valley bottom than what would have occurred otherwise. For both watersheds if it were not for the exceptionally long duration September 2013 storm, much of the deposited material likely would have persisted for many years to come.
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2015
Citation:
Brogan, DJ, PA Nelson, and LH MacDonald, 2015, Estimating and comparing two extreme post-wildfire peak flows in the Colorado Front Range, Hydrology Days 2015, Fort Collins, CO, 23-25 March.
- Type:
Journal Articles
Status:
Under Review
Year Published:
2015
Citation:
Brogan, DJ, PA Nelson, and LH MacDonald. "Estimating and comparing two extreme post-wildfire peak flows in the Colorado Front Range." Submitted to Earth Surface Processes and Landforms.
|
Progress 09/02/14 to 09/30/14
Outputs
Target Audience:
Nothing Reported
Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? On 5 September 2014, Peter Nelson, Lee MacDonald, and Daniel Brogan met with the USGS Powell Center working group on "Expanding high resolution topography for advancing the understanding of mass and energy transfer across landscapes: Opportunities, challenges, and needs." This is a group with considerable expertise using high-resolution topography to answer questions about hydrology, sedimentation, and geomorphology. Because this project intends to use high-resolution LiDAR topography to characterize patterns of erosion and deposition, this presented an opportunity to interface with experts in this area, and provided exposure for the project and professional development for the participants. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals? During the next reporting period, we expect to make considerable progress in developing the methods to compare high-resolution topographic datasets collected at different times. We have acquired several sets of airborne LiDAR coverage of our study areas, and will work with these data to a) determine the optimal way to register datasets to each other (for instance, using the iterative closest point algorithm over individual flightlines), b) difference the datasets to produce quantitative estimates of erosion and deposition (either by differencing point clouds directly using cloud-to-cloud or M3C2 algorithms, or by differencing raster grids using the Geomorphic Change Detection tool developed by Joe Wheaton at Utah State University), and c) relate these patterns to watershed characteristics, which also will be computed from analysis of high-resolution topography. Analysis of remotely-sensed topographic data will be complemented during the summer and fall of 2015 with ongoing field work aimed at characterizing rainfall, runoff, hillslope erosion, and channel change. We will use a network of rain gages to characterize spatial and temporal patterns of precipitation. Water level sensors placed in channels at different locations throughout the drainage network will provide data on hydrologic response at different spatial scales. Some of these water level sensors will be complemented with turbidity meters, suspended sediment samplers, and conductivity meters, to provide information on suspended sediment transport and water quality. After snowmelt and after each summer thunderstorm, our set of monumented channel cross sections and longitudinal profiles will be surveyed with GPS to characterize event-scale erosion or deposition.
Impacts
What was accomplished under these goals?
Because this project began on 2 September 2014, the reporting period covers only the first four weeks of the project. We therefore have not yet had time to complete any major activities or meet specific objectives, but we have begun to lay the groundwork for progress in the coming years. The primary activity performed during the reporting period relates to the objective of determining spatial and temporal patterns of erosion and deposition at scales from hillslopes to watersheds. We have obtained airborne LiDAR topographic data for our study sites, which were collected in the summer of 2012 and the summer of 2013 by the National Ecological Observatory Network (NEON). These are large datasets, and we are just beginning to work with them. A primary task for the next reporting period will be to develop methods to register these datasets to each other, to difference them, and to determine where differences are statistically significant. These high-resolution topographic datasets, along with the continued collection of field data, will be used to address the project goals of quantifying post-wildfire runoff, erosion, deposition, and sediment delivery, relating these to watershed characteristics, and assessing temporal changes in hydrologic and geomorphic response after a wildfire.
Publications
|
|