Web Resources for Tracking Drought Conditions
By David Kroodsma
Last week, as flood waters crested along the lower Mississippi, setting record water levels, just a few hundred miles away Texas was enduring the opposite challenge: drought. April was the fifth driest such much on record for that state, following the driest March on record. The past six months have been the second driest November to April period in the state’s history. Since January, these dry conditions have helped spark wildfires that have burned 2.7 million acres — an area roughly the size of Connecticut.
At Climate Central, we’re interested in how best to understand these climate phenomenon. How do we track droughts and floods, and how do we measure them? In this blog, the first in a series on climate data, I’ll summarize web resources for visualizing and analyzing drought.
Droughts are difficult to measure and compare, mostly because there’s no single definition of what exactly a drought is. Several different measures of drought — expressing different aspects of drought impacts — are commonly used. A drought can be agricultural (low soil moisture, which affects crops), hydrological (low rivers and reservoirs), meteorological (below average rainfall), or societal (affecting socieity in a measurable way), although the most popular benchmarks are whether it affects crops or rivers and reservoirs.
For tracking current droughts, the U.S. Drought Portal, provides a good start. Drought.gov is operated by the National Integrated Drought Information System, which is a collaboration of a number of govenrment agencies and led by the National Oceanic and Atmospheric Administration (NOAA). On the portal, three different interfaces provide information on current droughts, drought impacts, and projections for the next few months.
1) Drought Conditions: U.S. Drought Monitor
This analysis defines drought intensity from D0 (abnormally dry) to D4 (a drought only experienced every 50 to 100 years). These indices are a mix of a number of different measures, including soil moisture models, measures of stream flow, rainfall, and the subjective opinion of the drought specialists (see the full explanation here). The map is updated weekly and according to the May 24th report, almost 90 percent of Texas was experiencing severe drought, with almost half of the state in the midst of an “exceptional drought.” On the map, the letters "A" and "H" refer to whether the drought is agricultural, hydrological, or both. Click on it for a larger version.
2) Drought Impacts
This tool allows the user to find out exactly how drought is affecting a specific county. Click on a state, then click on a county, and read about different events that have been attributed to the current drought. These impacts are compiled from a few official sources, but members of the public can submit a drought related impact on this page as well. This tool is provided by the National Drought Mitigation Center at the University of Nebraska.
For instance, if I click on Texas, I get a map of all Texas counties. If I choose Tom Green County, an Oklahoma-shaped county in the center of the state, 13 impacts of drought are listed, mostly news stories about wildfires or agricultural challenges. There is also a link to a story about how desperate the community is for precipitiation: some locals have even been praying for rain.
3) Drought Projections
This forecast projects the likelihood of drought over the next three months. Predicting the weather beyond 10 days is a major challenge. Yet drought projections can be useful because drought is cumulative — it can take a few months of rain to recharge soils — and because large-scale ocean conditions such as El Niño make some long term forecasting more reliable. In May, as shown by the map, the center predicted drought to persist and intensify across Texas and the Southwest, but improve in the Southeast. An in-depth description of this map also notes that reduced snowpack in the Southwest could worsen conditions over the next few months.
These three tools, as well as other statistics, can be viewed on drought.gov's interactive map viewer.
Other Drought Indices
Palmer Drought Index
The Palmer Drought Index, one of the most widely used drought indices, is a calculation of soil moisture based on rainfall and estimated evaporation. It usually ranges from -4 (extreme drought) to +4 (extremely wet). Calculating the index, though, is complicated, and a full explanation can be found in this scientific paper.
This index is calculated for a region, and the National Weather Service has divided the U.S. into a few hundred climate divisions, which can be seen on the map to the right. This map is the Palmer Drought Index for the most recent month, which shows exceptionally dry conditions in Texas and New Mexico as well as a wet Ohio River Valley. Numerical values for each climate division can be found at the National Weather Service's Climate Prediction Center. The index has also been calculated for the entire country for every month during the past century.
Crop Moisture Index (CMI)
The CMI is similar to the Palmer Drought Index, but is more useful over shorter time periods. It uses a similar calculation based on rainfall and evaporation, but it incorporates changing conditions more quickly and is thus better for spotting emerging droughts (as well as emerging drought relief). To the right is the most recent map of the CMI. You can also find tabular data for each climate division.
Most climatologists don’t like using precipitation data alone to measure drought because droughts also depend on temperature and humidity, which influence evaporation, and hence also soil moisture, runoff, and water availability. Nonetheless, drought indices based only are precipitation are used because they are easy to calculate and compare.
Rainfall Data From The National Weather Service
The National Weather Service provides precipitation data for the entire country, allowing users to analyze any day, month, or year on record. For any given month or year, you can view the total amount of rainfall, the average amount, the departure from average, or the percent of the average. This nationwide data can be downloaded as shapefiles or netcdf files for technical users. The map on the right shows rainfall for April 2011 as a percent of normal rainfall. The Ohio River Valley experienced more than 300 percent of the average rainfall in April, while much of Texas and New Mexico had less than five percent of their average.
Standardized Precipitation Index
"Percent of average rainfall," as shown above, is not often used to estimate droughts. It can be misleading because the median rainfall is different than the mean. For instance, a few years of extremely heavy rain can significantly increase the mean rainfall, meaning that most years will be less than "100 percent" of the mean, even though they won't signify droughts.
The Standardized Precipitation Index, or SPI, is often used instead. It is essentially the number of standard deviations (with some other complicated math because rainfall isn’t a “normal” distribution) that precipitation is from the median value, with negative values signifying below normal and positive signifying above. This index can be calculated over any period of time, from days to years, and therefore it can measure drought severity over any time period of interest. The map on the right, provided by the Western Regional Climate Center, shows the SPI for the past month across the country for the various climate divisions of the U.S. On this SPI website, provided by the Western Regional Climate Center, you can also display the SPI for the past one to 72 months.
Stream Flow and Hydrology
Stream Flow Data
The United States Geological Survey (USGS) measures streamflow on hundreds of rivers around the country and records drought and flood conditions. The map on the right shows the current water flow at these stations as a percent of normal flow for this time of year. Red signifies below average (hence Texas is red) and blue is far above average (the Ohio River Valley and Mississippi Rivers). If you click on the map, you will visit the USGS WaterWatch, which also records drought based on river flow conditions around the country.
In the West, snowpack is an important factor in droughts. A thick snowpack is like a huge reservoir, storing water for future months. A thin snowpack can mean water stress for much of the West come summer and fall. This interface, provided by the National Resource Conservation Service, allows users to compare the snowpack for any winter or spring month with that of an “average” year. The map on the right shows the state of snowpack as of April 1, 2011, note the abundant snowpack throughout much of the West.
|Drought.gov also contains information on early drought warning monitoring for California and the upper Colorado River Basin. The Apalachicola-Chattahoochee-Flint River Basin, in Georgia, Alabama, and Florida, also has such a system in place. Also, California’s government has a website devoted to droughts, where users can peruse the status of rainfall, groundwater, reservoirs, and snowpack, as well as learn how to adapt to drier conditions and conserve water.|
|The Southern Climate Impacts Planning Program, a new climate research initiative that helps communities plan for disasters related to weather and climate, has a tool to track rainfall in many southern and southeastern states. You can browse rainfall statistics for different climate regions or use their various climate tools to better understand the region.|
|Although many ways to measure drought exist — by one estimate, there are more than 150 drought indices — ultimately, it is the impacts on natural and human systems that matter, and any metric will be incomplete without accompanying news stories and analysis.|
What is the Link Between Climate Change and Droughts?
While some places on the globe will see increased rainfall, as the planet warms in response to rising concentrations of greenhouse gases in the atmosphere, many regions will likely get significantly drier. For instance, one research paper, published in Science, analyzed 19 different climate models and found that almost all predict substantial drying in the American Southwest, and that the “average” conditions later this century would be similar to a drought today. The Southwest won’t be alone. According to another study, depending on greenhouse gas emissions, at the end of the century as much as 30 percent of the world’s land could experience extreme droughts in any given year. By comparison, in any given year today, only one percent of the earth’s land area experiences extreme drought.
Below is a map from a study by the National Center for Atmospheric Research (from fig 11 of the study), which calculated the future average Palmer Index for every point on the world by averaging rainfall and temperature outputs from more than 20 different climate models. The study used the "A1B scenario," a scenario in which carbon dioxide levels increase from their current levels of 390 parts per million to over 500 parts per million in 2050 and 700 parts per million by 2100. Note from earlier in this post that the Palmer Index usually doesn't drop below -4 (extreme drought), yet on the map large swaths of the world show a Palmer Index of -10 or worse!
These drought conditions will be due largely to warmer temperatures, which increase evaporation, instead of just changes in precipitation. Compare the map above to the figure below, which is copied from the United Nation Intergovernmental Panel on Climate Change’s 2007 Synthesis and Assessment Report. The figure shows how precipitation, as predicted by the average of almost 20 climate models, is expected to change by the end of the century (the same emissions scenario as above, "A1B," is used). The image on the left shows the percent change in rainfall for the months winter months, and the image on the right is for the summer. Notice the large red and yellow swaths in the subtropics, representing a decrease in rainfall of ten percent or more. But these decreases are not quite as dramatic as the increases in drought as shown on the map above -- droughts will be caused by both decreases in rainfall and warmer temperatures, and droughts will increase even in some places where total rainfall increases.
Technically, though, these drier regions won’t be experiencing drought because drought is defined as a period that is “drier than normal”, and the new conditions will be the new normal. (The National Weather Service updates what it considers the “normal climate” every decade, and Climate Central’s Heidi Cullen discusses this in a short video). The “30 percent of the world experiencing drought” quoted above compares future conditions to present conditions. By today’s standards, drought will increase. But, in comparison to future climate, will drought also increase? In other words, will the climate be more or less variable around these new average conditions?
Some locations, at least, will experience more dry periods because heavy rainfall is expected to increase, even though average rainfall is predicted to stay the same or decrease. Heavy rainfall events have already increased all over the globe, and a recent paper in Science demonstrated that this increase is partially due to an increase in greenhouse gases.
A study of climate models analyzed the number of consecutive rain-free days, or “dry days,” that any location would experience in a year. This isn’t the best measure of drought, but it does give an idea of whether a region will experience drier-than-average conditions. This study analyzed nine climate models and showed that on average the globe would experience a modest increase in the number of consecutive dry days, even though average rainfall would increase. That suggests that not only will heavy rains increase, but so will conditions conducive to drought. But this trend was only significant in simulations that assumed unabated emissions of greenhouse gases.
In short, our best research says this: Some regions, particularly the subtropics, will become drier, and the character of rainfall will change, with most places seeing more heavy rainstorms. Relative to this new average climate, increases in dry spells should also be likely, but this increase is not as robustly predicted as the expected increase in heavy rainfall. In addition, soil moisture and river flow will decrease in many regions — even regions where rainfall slighlty increases — because warming will result in greater evaporation.