Grantee Research Project Results
Final Report: Forecasting and Evaluating Vulnerability of Watersheds to Climate Change, Extreme Events, and Algal Blooms
EPA Grant Number: R835203Title: Forecasting and Evaluating Vulnerability of Watersheds to Climate Change, Extreme Events, and Algal Blooms
Investigators: Stevenson, R. Jan , Hyndman, David , Qi, Jiaguo , Moore, Nathan
Institution: Michigan State University
EPA Project Officer: Packard, Benjamin H
Project Period: June 1, 2012 through May 31, 2017
Project Amount: $749,801
RFA: Extreme Event Impacts on Air Quality and Water Quality with a Changing Global Climate (2011) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Water Quality , Climate Change , Air , Water
Objective:
Climate change is projected to increase temperature and the intensity of extreme weather events. Increases in water temperature with greater frequency and intensity of floods and droughts are a perfect storm for exacerbating problems with algal blooms in lakes. Higher water temperatures stimulate algae to grow faster. Heavy rains cause runoff of nutrients that stimulate algal growth. Drought and calm weather create conditions for harmful algal blooms. The goals of this project were to advance our knowledge of relationships between extreme events and water quality across the diversity of climatic and geologic conditions of the US, as well as to develop tools for advancing that knowledge and informing strategies for water quality management under different climate change scenarios. We had three primary tasks:
1. Identify historical algal blooms and statistically investigate how these relate to extreme events across a range of hydrologic regimes.
2. Use process-based hydrology and algae models to explore causal linkages between extreme events and algal blooms, and predict the influence of projected climate changes on algal biomass, and
3. Develop statistical models that can be applied nationwide to examine vulnerabilities to extreme events under different management strategies.
Summary/Accomplishments (Outputs/Outcomes):
Task 1
Our project required using historic satellite images to characterize algal abundance in lakes and to provide sufficient records of past and year-round conditions so that we can relate changes in temperature and extreme hydrologic events that result from climate change to changes in algal abundance. Most algorithms for estimating lake chlorophyll a with satellite data are based on linear regression and were not been sufficiently accurate to provide measurements of chlorophyll in lakes across the broad spatial and temporal scales that we needed for this study. We spent considerable time evaluating MODIS and Landsat imagery for assessing in-land lake chlorophyll because these satellites provided imagery with resolution fine enough to observe inland lakes. MODIS had relatively course spatial resolution, which would prevent its application in many smaller lakes, but it had high spectral resolution with many images measured with many spectral bands that could help resolve chlorophyll a signals from other color producing substances in water. Landsat had high spatial resolution so we could have many image pixels in almost all lakes, but it had low spectral resolution than MODIS. We acquired a 28-year dataset of Missouri reservoir conditions from Jack Jones at the University of Missouri. This dataset included measurements of total suspended solids and colored dissolved organic matter as well as chlorophyll a, which enabled evaluating interactive effects of the three main color-producing substances on chlorophyll a measurement with satellite imagery. We also used the water chemistry data for the 2007 National Lakes Assessment by the United States Environmental Protection Agency, which had a broad spatial scale. We compared many classical regression and machine learning methods to develop an algorithm to measure chlorophyll a in lakes with these two primary datasets.
We were able to develop accurate algorithms for measuring chlorophyll a in lakes with Landsat imagery by using all available bands and band ratios, excluding the temperature band, and a machine learning statistical method called boosted regression tree (BRT). Different BRT algorithms were best for the Missouri reservoirs and lakes in the National Lakes Assessment, which may have been due to differences in lake sampling or regional differences in color producing substances in lakes. Analysis of the Missouri reservoir data showed suspended sediment and colored dissolved organic matter did not interfere with chlorophyll estimates with Landsat images with enough magnitude to mask expected changes in chlorophyll with climate change (Lin et al. accepted). Analysis of the National Lakes Assessment data showed statistically significant relationships between chlorophyll measure with Landsat imagery and total phosphorus concentration in lakes, and this relationship could be as precise as relationships chlorophyll measured with water samples and lake total phosphorus if more than one Landsat image was used to measure chlorophyll (Lin et al. in review).
We related historical algal blooms in Missouri reservoirs to extreme events and temperature changes over a 28-year period having Landsat TM images. We hypothesized more intense precipitation projected with climate change could bring more nutrients from watersheds to lakes, which create conditions that increase risk of algal blooms with projected warming. This hypothesis has not been tested with long-term and multi-season observations of algal biomass in lakes. With the increases in lake surface water temperature and precipitation intensity (mm/d), algal biomass more likely responded to temperature than precipitation. Rising water temperature affected mean annual chlorophyll-a more than summer chlorophyll-a, indicating that projected warming might result in the expansion of the algal growth season rather than increasing the peak biomass during summers. Summer algal biomass did increase with increasing spring precipitation in some study reservoirs. The lack of effect of precipitation on reservoirs could be related to comparatively small size of watersheds relative to lakes. In addition, some lakes had low human land use in watersheds and therefore, low nutrients and relatively little capacity to respond to lake warming.
During the development of algorithms to measure lake chlorophyll with Landsat imagery, a technique for satellite data storage and cloud calculation of chlorophyll a by using Google Earth Engine was developed, which automatically and rapidly produces long-term whole-lake characterizations of algal biomass using Landsat satellite imagery and our chlorophyll measurement algorithm. Application of this tool improves the accuracy of lake assessment and produces results that can be used for national scale assessments, evaluation of historic conditions in lakes to relate algae to climate and land use change, and forecast vulnerability of lakes to climate change. This data is available for public use.
Task 2
We developed Landscape Hydrology Model simulations for two regional systems, Michigan’s Lower Peninsula (MLP) and the Northern High Plains. These models provide hourly flux estimates that have been validated using measured streamflow data from USGS gaging stations in three of its subregions of the MLP model (Grand Traverse Bay, Muskegon River, and St Joseph/Maumee Rivers).
We then developed spatially explicit nutrient loading maps for nitrogen and phosphorous sources across Michigan’s Lower Peninsula (Luscz et al., 2015) and the US side of the Great Lakes Basin (Hamlin et al. in prep.), including atmospheric deposition, manure application, agricultural chemical fertilizer, septic tanks, non-agricultural fertilizer, and point sources. The approach disaggregates county estimates to 30 m pixels across the region based on relative land uses using a Geographic Information System. These source maps are critical inputs to improving estimates of nutrient loads from watersheds. This information fed into a new nutrient loading model for Michigan’s Lower Peninsula that predicts total nitrogen and phosphorus loads by combining spatially-explicit maps of nutrient sources with a statistical transport model to estimate nutrient attenuation along multiple pathways (Luscz et al., 2017). This approach estimates the relative contribution of each nutrient source to the overall loads to the Great Lakes, which provides valuable information to environmental managers and policy makers who are trying to reduce nutrient loads and improve water quality.
Finally, we quantified the long legacy of human activities and associated nutrient loading to the landscape, which may eventually become transported in groundwater that is flowing toward surface water bodies (Martin et al., 2017). This groundwater legacy effect significantly complicates efforts to remediate nonpoint nutrient contamination plumes through (typically) unexpected delays in observing changes due to management actions. Water chemistry observed in surface water features can be better predicted using statistical methods that account for the legacy land cover, as compared to methods that use only current land cover.
Task 3
Remotely sensed whole-lake chlorophyll in 585 lakes in continental United States showed that hydraulic conditions such as basin slope, soil, and land use/cover were more important than lake surface temperature and precipitation in explaining the spatial variation of lake chlorophyll during May-August. Lake surface temperature had a positive effect on algal biomass. Precipitation intensity could have a positive effect on lake algal biomass, presumably by increasing runoff and nutrient pollution. Total annual precipitation, however, was negatively related to algal biomass, which indicated nutrients were diluted in lakes, and perhaps watersheds, with higher precipitation. As a result, climate change was predicted to have a range of effects on algal biomass in lakes depending upon the change in lake temperature and relative changes in precipitation and total annual precipitation. Although some effect was observed on algal biomass during summers, climate effects were better related to changes in estimated annual algal biomass in lakes. When we predicted algal biomass in lakes in 2099, we found ranges in effects from 3 µg chlorophyll a/L reductions to 5 µg chlorophyll a/L increases in annual averages of algal biomass in lakes. The greater the increase in climate, the more likely algal biomass will increase in lakes. The lakes with highest vulnerability to algal increases with climate change were those at high latitudes and high elevations with colder waters and lakes in climates where precipitation intensity will increase relatively more than total annual precipitation.
Conclusions:
1. Remote sensing of chlorophyll a in lakes can be used to assess lake condition and relate lake condition to natural watershed and climate factors, land use and land cover change, and climate change.
2. Rising water temperature affected mean annual chlorophyll-a more than summer chlorophyll-a, indicating that projected warming with climate change will increase algal biomass more during non-summer periods than summer in many lakes of the US.
3. Summer algal biomass will increase in lakes at high latitudes and elevations that currently have summer temperatures that are lower than the 23-25° C temperature range that produces peak algal growth and biomass.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 7 publications | 7 publications in selected types | All 7 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Liu B, Stevenson RJ. Improving assessment accuracy for lake biological condition by classifying lakes with diatom typology, varying metrics and modeling multimetric indices. Science of the Total Environment 2017;609:263-271. |
R835203 (Final) |
Exit Exit Exit |
Supplemental Keywords:
global change, land use change, management, phytoplankton, watershedsRelevant Websites:
https://msu.edu/~rjstev/X-EVENTS.html
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.
Project Research Results
- 2016 Progress Report
- 2015 Progress Report
- 2014 Progress Report
- 2013 Progress Report
- 2012 Progress Report
- Original Abstract
7 journal articles for this project