Western Forest Fires and Long-term Impacts on Water Quality ASHLEY RUST, PHD

Transcription

1 Western Forest Fires and Long-term Impacts on Water Quality ASHLEY RUST, PHD 1

2 Currently 4 large fires burning throughout the West, 3 are in California. More acres have burned in California in 2018 than in the past decade, > 1.6 million acres burned 2018 year-to-date: 52,080 fires to date 8,517,245 acres burned 10- year average year-to-date: ,375 fires 6,319,455 acres burned (Picture of Carr and Ferguson fires as glimpsed by astronaut Alexander Gerst, the Atlantic 8/10/18; data from National Interagency Fire Center 2

3 Background Image from Union of Concerned Scientists Data from Federal Wildland Fire Occurrence Data; Westerling et al. 2006; and Stewart et al

4 Image from Union of Concerned Scientists Data from Federal Wildland Fire Occurrence Data; Westerling et al. 2006; and Stewart et al

5 Fire impact on water supply & Aquatic ecosystems 2/3 of municipalities in the U.S. and 1/3 of large cities worldwide get their drinking water from forested areas Post-fire erosion of sediments and ash impact water quality Wildfires release significant amounts of nutrients Wildfires cause heavy metals to be mobilized Sediments, nutrients, and metals alter water quality and harm the aquatic ecosystem Committee on Hydrologic Impacts of Forest Management, 2008; Bladon et al

6 Many studies have observed how a single fire impacts local water quality. We evaluated 159 fires to identify common response. What mitigates water quality response and aids recovery? Can we use models to anticipate water quality changes after fire? 6

7 Study area We created a Fire-Water Quality Database: 159 fires from water quality parameters N percent change in concentration and loading rate from 5 year baseline period to 5 year post-fire period Rust, A.J., T.S. Hogue, S. Saxe and J. McCray (2018) Post-fire water quality response in the western United States. International Journal of Wildland Fire. Doi.org/ /WF

8 Common Water Quality Response The first five years after fire Nitrogen: particulate forms increase in concentration after 30% of fires; dissolved forms decreased after 30-45% of fires all forms of N increased in loading rate (flux) Phosphorus: Dissolved and particulate forms increased in concentration and flux after 33% of fires 8

9 Common Water Quality Response The first five years after fire Metals: Dissolved forms decreased concentration, total forms increased concentration, and flux for both increased in 33% of fires 9

10 Statistical study Post-fire response and recovery: Some streams experience high nutrient concentrations and flux, others experience high total suspended solids, while other streams recover quickly after fire and show almost no detectable change in water quality Objective: Identify the determinants of water quality response after fire 10

11 Statistical study Climatological Variables Aridity Index (Precip./ PET) Solar Radiation Biological Variables NDVI pre-fire (a measure of veg. greenness) NDVI post-fire, 1 year, 2 years, 3-5 years Physical Variables Elevation Latitude Slope Proximity to Fire Proximity to Urban Area Percent of Watershed Burned Percent of Fire Moderate and High Severity Edaphic Variables Percent of Calcium Carbonate in Soil Percent Clay Percent Silt Percent Organic Matter Soil K Factor Water Content- Low, Field Capacity Water Content- High 11

12 Statistical study 12

13 Percent Change in Nitrogen species Results Conditional inference tree Median percent change in dissolved metals concentration (Al, Ba, Cd, Cu, Mn, Pb, and Zn) five years after fire 13

14 Conditional inference tree Median percent change in total metals concentration (Bo, Cd, Cu, Fe, Mn, Ni, Pb, Se, Zn) five years after fire. 14

15 C-Tree Results Summarized Calcium Carbonate in Soil Nitrate Concentration Organic Matter Dissolved Metals Concentration Greenness after fire Total Phosphorus Flux Greenness after fire Total Metals Concentration & Flux 15

16 West Fork Complex Fire, Rio Grande, Colorado USA 16

17 West Fork Complex Fire Case Study How did the fire impact water quality, insects and fish and how quickly do they recover? 17

18 West Fork Complex Fire Study Area Fires start June 5, 2013 in the Upper Rio Grande, near the Rio Grande Reservoir 3 rd largest fire by area in Colorado history 110,000 acres burned 60% moderate to severe burn 18

19 TSS mg/l Results 1200 Total Suspended Solids August 5, mg/l Trout 0 Control Fire Marshall impacted Park Fire Trout impacted Creek mainstem tributary 19

20 Results Total Suspended Solids and Turbidity Fire Impacted mainstem Control Control Fire impacted tributary For 3 years after the fire, concentrations of total suspended solids remained high in the mainstem of the Rio Grande and in the fire impacted tributary. 20

21 Results Turbidity in the Rio Grande above and below Fire Unburned Burned Hydro-hyetographs with turbidity data plotted in red. Turbidity over the rainy season in the control (unburned) and impact (burned) stream sampling sites Rust, A.J., T.S. Hogue, J. Randell, A.S. Todd (2018) Water quality impacts from fire: monitoring ecosystem health and recovery in the upper Rio Grande. In prep. 21

22 22

23 Fish Population: Resilience on the Rio Grande Data from Colorado Parks and Wildlife 23

24 Modeling study Rapid Response Erosion Database Uses GeoWEPP model To model erosion and hillslopes at risk of failure Case Study: Colorado s 2013 West Fork Complex Fire where erosion caused high total suspended solids and high turbidity 24

25 25

26 RRED Predictions Results Unburned Burned The tributary watershed before and after fire with RRED GeoWEPP s predicted sediment yields T = tolerable erosion threshold, 1T = 1Mg/ha/y 26

27 Results Model Results compared to Estimates from Ground Measurements While GeoWEPP accurately predicted the area that did experience a landslide, the model over-predicts the sediment yield. Through sampling total suspended solids and measuring flow at the watershed s outlet, our aggregate estimate: 2014 (1 year after fire): Mg/ha/y OR 1/500 T per year RRED aggregate estimate is 1.05 T per year 2015 (2 years after fire): Mg/ha/y OR 1/77 T per year 2016 (3 years after fire): Mg/ha/y OR 1/333 T per year During the landslide event, which caused the fishkill, we measured 1000mg/L TSS for a single day, this would be the equivalent of 0.33 Mg/ha/y OR 3T 27

28 Conclusions Conclusions Statistical analysis reveals common water quality responses after fire Conditional inference tree method provides insights as to which watershed properties influence the type and direction of the response Initial RRED-GeoWEPP modeling results over-estimate the degree of erosion 28

29 More Communities Impacted by Fire June 6, 2018 Colorado s 4 th largest fire, the 416 fire, burns Hermosa Creek above Durango and the Animas River (same watershed as Gold King Mine Spill) High metals and nutrients have been observed in first few months after fire Photos courtesy of Denver Post and the Durango Herald 29

30 Acknowledgements Dr. Terri Hogue Colorado School of Mines Thank you! Contact: Ashley Rust DiNatale Water Consultants (303) (720) cell Award # L14AC00165 Rust, A.J., T.S. Hogue, S. Saxe and J. McCray (2018) Post-fire water quality response in the western United States. International Journal of Wildland Fire. Doi.org/ /WF17115 Rust, A.J., T.S. Hogue, J. Randell, A.S. Todd (2018) Water quality impacts from fire: monitoring ecosystem health and recovery in the Upper Rio Grande. North American Journal of Fisheries Management. In Prep Rust, A.J, S. Saxe, J. McCray, C. C. Rhoades, T. S. Hogue. Evaluating the factors responsible for post-fire water quality response 30 in forests of the western USA. In Review