The University of Arizona

Hydro-climate Dashboard | CLIMAS

Hydro-climate Dashboard

HydroClimate Context

Hydro-Climate Context for November
Updated on Oct 28, 2016. Text valid for Oct 28 - Nov 22

A weak La Niña event is expected to develop during the fall season, but is likely to be short lived.

Climatologically, precipitation becomes more abundant in northern CA and the Sierra-Nevada mountains in November as the water year starts (Figure 1).  Flash flooding does happen in November, as shown in the historical count of local storm reports, but is mostly in portions of the southern part of CONUS FEMA Region IX (Figure 2).  Climatologically, mainstem river flooding can (Figure 3) occur throughout CA and portions of eastern AZ in November.  Very few FEMA disaster declarations due to flooding have occurred in the CONUS FEMA Region IX area in the month of November, and those that have, have been mostly in southern AZ (Figure 4).   

The water year has started out strong in the northern half of CA, and northern NV, with 200-400% of normal precipitation, while the southern areas are starting out below normal (Figure 5).  Snowpack conditions (Figure 6) are currently low, but streamflow conditions (Figure 7) in the northern areas are largely running normal to above normal, and below to much below normal in the southern areas.  One month into the water year, reservoir capacities (Figure 8) look better in the northern areas than in the southern areas.

Currently, equatorial sea surface temperatures are below average, where a weak La Niña event looks to be developing.  NOAA’s Climate Prediction Center (CPC) favors weak La Niña conditions to develop this fall, but the event looks like it could be short lived and weaken in the winter months (Figure 10).  Impacts from La Niña tend to be felt the most during the late fall and winter in CONUS FEMA Region IX.  A correlation does exist that would favor drier than normal conditions for parts of the southern half of CA, southern NV, and AZ during the Nov-Dec-Jan season, and wetter than normal conditions for the very northern portions of CA and NV (Figure 9).  The latest 3-month outlook for Nov-Dec-Jan from CPC slightly favors below average precipitation for AZ, and far southeastern CA, in CONUS FEMA Region IX, with equal chances for above, near, or below average precipitation for the rest of the region (Figure 11).

Historical Averages

Figure 01. Precipitation Climatology

Figure Updated:  
Last Thursday of month
Figure Source and Data:  
NOAA-Climate Prediction Center
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) at the University of Arizona produced this map using the NOAA Climate Prediction Center’s Daily U.S. Unified Precipitation Dataset1. In this dataset, daily precipitation values are interpolated from stations within and around each 0.25 degree latitude by 0.25 degree longitude grid box (or 28 km latitude by 21 km longitude), which equates approximately to an area of 600 km2. At this resolution, there are 300 by 120 grid boxes encapsulate the U.S. More than 8,000 stations contribute to this dataset, which is updated each day and includes data from 1948 through the present. The interpolation uses the Cressman Technique2, which is generally more accurate than other simple methods such as linear interpolation. However, it is sensitive to reporting errors and the number of neighboring stations and can have more errors in mountainous areas due to large precipitation changes with altitude.
Figure Display:  

The map presents the average monthly precipitation; average is calculated over 1948–2013.

Interpretation Recommendations:  
  • This map shows the spatial pattern of the historical average precipitation. While it is unlikely that the current month will reproduce the average spatial pattern shown here, it provides the “best guess” against which forecasts are compared.
  • In addition to showing average conditions, this map conveys relative precipitation amounts, or areas that are relatively wet or dry compared to other regions on the map or other times during the year.
  • All monthly climate information products that display precipitation anomalies (the amount above or below average) and percentages above or below average are based on a similar long-term average as represented in this map.
  • Generally, floods are more likely to occur in wet regions during historically wet months.
  • Consulting the plot of all 12 months shows the seasonal transition between wet and dry conditions.

Figure 02. NWS Flash Flood Climatology

Figure Updated:  
Last Thursday of month
Figure Source and Data:  
NWS Local Storm Reports
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) at the University of Arizona produced this map using data from the The National Oceanic and Atmospheric Administration’s (NOAA) National Climatic Data Center Storm Events Database1. This database contains information on more than 48 different types of high-impact weather events specified in the National Weather Service Directive 10-16052, including flash floods. The information is collected by all National Weather Service offices across the country and the data are archived at the NOAA National Climatic Data Center. The database is updated several times a year. Most event records extend back to 1996; tornado and severe thunderstorm records extend back to 1950. Flash flooding impacts recorded in the Storm Events Database are defined as “a rapid and extreme flow of high water into a normally dry area, or a rapid water level rise in a stream or creek above a predetermined flood level, beginning within six hours of the causative event (e.g., intense rainfall, dam failure, ice jam-related), on a widespread or localized basis (NWS Directive 10-1605).”
Figure Display:  

This map displays the average number of flash flooding impacts observed each month for each county in FEMA region IX during the 1996–2013 period (18 years). Values close to 1 indicate that, on average, a flash flood impact was observed in that county every year; a value of 0.25 indicates that, on average, a flooding impact was observed in that county once every four years.     

Interpretation Recommendations:  
  • This map helps visualize the historical risk of flash flood events and their geographic pattern across the region. Flash floods often happen in specific places at specific times during the year.
  • Flash flood risk shifts across the region during the year in concert with the dominant drivers of precipitation variability, which includes the position of the winter storm track and the onset of the summer monsoon system (see plot of all months).
  • The average number of impacts observed each month should only be considered a coarse estimate of monthly risk based on a relatively short period of record. Additional data, including current conditions and weather forecasts, provide other risk assessment information that should be used in conjunction with this map.
Additional Resources:  

Figure 03. NWS Flood Climatology

Figure Updated:  
Last Thursday of month
Figure Source and Data:  
NWS Local Storm Reports
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) at the University of Arizona produced this map using data from the The National Oceanic and Atmospheric Administration’s (NOAA) National Climatic Data Center Storm Events Database1. This database contains information on more than 48 different types of high-impact weather events specified in the National Weather Service Directive 10-16052, including floods. The information is collected by all National Weather Service offices across the country and the data are archived at the NOAA National Climatic Data Center. The database is updated several times a year. Most event records extend back to 1996; tornado and severe thunderstorm records extend back to 1950. Flooding impacts recorded in the Storm Events Database are defined as “any high flow, overflow, or inundation by water which causes or threatens damage. In general, this would mean the inundation of a normally dry area caused by an increased water level in an established watercourse, or ponding of water, generally occurring more than 6 hours after the causative event, and posing a threat to life or property (outline in NWS Directive 10-1605).”
Figure Display:  

This map shows the average number of flooding impacts observed each month for each county in FEMA region IX during the 1996–2013 period (18 years). Values close to 1 indicate that, on average, a flooding impact is observed in that county once every year; a value of 0.25 indicates that, on average, a flooding impact was observed in that county once every four years.     

Interpretation Recommendations:  
  • This map can help visualize the historical risk of flood events and their geographical pattern across the region. Floods often happen in specific places at specific times during the year.
  • Flood risk migrates around the region during the year in concert with the dominant drivers of precipitation variability, which include the position of the winter storm track and onset of the summer monsoon system (see plot of all months).
  • The average number of impacts observed each month should only be considered a coarse estimate of monthly risk based on a relatively short period of record. Other data, including current conditions and weather forecasts, provide additional risk assessment information that should be used in conjunction with this map.
Additional Resources:  

Figure 04. FEMA Flood Disaster Climatology

Figure Updated:  
Last Thursday of month
Figure Source and Data:  
FEMA
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) at the University of Arizona produced this map using data from the FEMA Disaster Declarations Summaries Dataset1. The dataset contains information on all presidentially declared disasters from 1964 to present, including major disasters, emergency declarations, and fire management assistance situations. The FEMA National Emergency Management Information System maintains the dataset. It contains raw, unedited data that may contain errors from manual entry.
Figure Display:  

This map shows the average monthly number of presidentially declared disasters by county for events classified as floods or severe storms. The average is calculated for the entire period of record from 1964 to the end of the last complete year in the database. Values close to 0.1 indicate that, on average, a disaster designated as flooding and/or a severe storm was observed once every 10 years in the historical record, or five times in a 50-year record.

Interpretation Recommendations:  
  • This map helps visualize the historical risk of past flooding and severe storm disasters and their geographic pattern across the region. Higher risk areas migrate around the region during the year in concert with the dominant drivers of precipitation variability, which include the position of the winter storm track and the onset of the summer monsoon system (see plot of all months).
  • While a value 0.1 indicates a disaster was observed on average once every 10 years, it does not mean that events were evenly spaced through time. Disasters and any extreme events can cluster in time.  
  • The average number of disasters observed each month should only be considered a very coarse estimate of monthly risk based on a relatively short period of record. Additional data, including current conditions and immediate forecasts, provide critical risk assessment information that should be used in conjunction with this map.
Additional Resources:  

Current Conditions

Figure 05. Precipitation in previous 30 days

Figure Updated:  
Daily
Figure and Data Source:  
The National Oceanic and Atmospheric Administration’s (NOAA) High Plains Regional Climate Center produced this map. Data are assimilated each day from numerous sources1, including weather stations at airports and from the National Weather Service Cooperative Network. The number of stations reporting each day, however, can change. The data have been vetted by automatic quality control measures but are considered provisional until NOAA has conducted more thorough quality control.
Figure Display:  

Total precipitation accumulated in the last 30 days is compared to the average of the same 30-day period between 1981 and 2010 to produce a percent of “normal.” The 30-year, 1981–2010 period is a common reference time frame in climate analyses, providing a sufficiently long period for statistical significance testing.

An interpolated map is presented here because it enables quick, intuitive visualization of precipitation across the region. Interpolation, however, introduces error because data are created for locations where it is not actually measured. Compare the two images below. The dot image (right) is where NOAA measuring stations are recording data. The interpolated image (left), the one shown on the dashboard, is using those stations to create data in the interstices. Interpolation is performed using the Cressman Technique2, which generally is more accurate than other simple methods such as linear interpolation. However, it is sensitive to reporting errors and the number of neighboring stations and can have more errors in mountainous areas due to large precipitation changes with altitude.

Interpretation Recommendations:  
  • The interpolation map is good for obtaining a quick, general understanding of the region. Consulting the dot map will provide more accurate precipitation amounts and locations.
  • In regions and months where precipitation is historically low, a small amount of rain and/or snow can cause very high percent of normal values. For example, a half-inch of rain in June in Yuma, Arizona, would produce well above-average rain totals but this amount would likely not produce flooding.
Additional Resources:  

Figure 06. Snow Pack Conditions

Figure Updated:  
Daily
Figure and Data Source:  
The U.S. Department of Agriculture’s Natural Resources Conservation Service (NRCS) produced this map. The NRCS monitors nearly 2,000 high-elevation stations and focal points around the West using SNOTEL and manual measurements called snow courses; this network is the nation’s principal source of climate and weather information at high elevations. SNOTEL sites automatically record measurements every 15 minutes and transmit the data to NRCS. These increments are then averaged in hourly and daily format. Figures 1–3 show the locations of the SNOTEL sites; there are 32 stations in California, 57 in Nevada, and 22 in Arizona. Most stations are located in meadows or open areas and near the crests of ridges separating watersheds. They are often on northern aspects to avoid the brunt of the winter sun.
Figure Display:  

This map shows percent of normal snow water equivalent, or SWE. SWE is the amount of water contained in the snowpack, a measurement that standardizes the snowpack to account for density differences in the snow. The total SWE for the current day is compared to the 1981–2010 median for that same day. The 30-year, 1981–2010 period is a common reference time frame in climatological analyses, providing a sufficiently long period for statistical significance testing.

Interpretation Recommendations:  
  • This map conveys the potential for flood risk in the spring when warm temperatures can rapidly melt snowpacks.
  • This map provides only a snapshot in time, and the SWE percentages can fluctuate from day to day. Early in the winter SWE percentages can fluctuate day-to-day more than later in the winter. Consistently viewing this image can help monitor spring flood risk.
  • Low SWE percentages do not necessarily equate to low flood risk. A warm winter storm, for example, can raise the snow line and cause rain to fall on snow, producing  flooding even when snowpacks are below median.
  • High SWE percentages suggest that spring streamflows will be higher. This image can be viewed in combination with reservoir levels to suggest locations where high streamflows may increase flood risk.
  • California snowpack conditions are not well represented in these maps; the NRCS only monitors parts of the Sierra Nevada Mountains near Lake Tahoe and has several sites in the northeast corner of the state (see Figure 1). The California Department of Water Resources has its own snowpack monitoring network for other regions1.
Additional Resources:  
  1. Snowpack monitoring, California Department of Water Resources: http://cdec.water.ca.gov/snow/

Figure 07. Streamflow Conditions

Figure Updated:  
Daily
Figure and Data Source:  
The U.S. Geologic Survey (USGS) WaterWatch program produces this map. Generally, the data are updated hourly.
Figure Display:  

This map shows the streamflows value at each stream gauge as a percentile. Percentile is on a scale of 100 and indicates the percent of a distribution that is equal to or below it. For example, a river discharge at the 90th percentile is equal to or greater than 90 percent of the discharge values recorded on this day of the year during all years that measurements have been made.  In general:

  • streamflows greater than the 75th percentile are considered above normal
  • streamflows between the 25th and 75th percentiles are considered normal
  • streamflow less than the 25th percentile are considered below normal

The colors on the map represent streamflow conditions relative to their historical 30-year average for the same point in time.

On the map’s key (Figure), the flow category “Low” indicates that the estimated streamflow is the lowest value ever measured for the day of the year. Similarly, the flow category “High” indicates that the estimated streamflow is the highest value ever measured for the day of the year.

Open circles represent gauges that are "not ranked," indicating a flow category that has not been computed. Common reasons for a "not ranked" category are insufficient historical data or a lack of current streamflow estimates.

Interpretation Recommendations:  
  • This streamflow map highlights flood and high-flow conditions.
  • When streamflows are in the upper percentiles, flood risk is generally higher; floods may already be occurring in some locations.

Figure 08. Reservoir Conditions

Figure Updated:  
Last Thursday of month
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) generates this map using publically available data. Data for California come from the California Department of Water Resources1. Data for Nevada come from the U.S. Department of Agriculture’s Natural Resources Conservation Service (NRCS)2. For both states, reservoir storage data are updated around the 10th day of each month and reflect values for the end of the previous month. Reporting lag times create the updating delay.
Figure Display:  

On the left image below, the circle corresponds to the location of the reservoir and its color denotes the current reserovir capacity (as of the image update date reported on the image). The right image below shows the reservoir total storage volumn. The size of the circle is proportional to the reservoirs’ total capacity. The location on both images and the size of the symbols on right image do not change. The color on the left images does change each update period and represents the storage capacity as a percent of its 1981–2010 average. Data from both the NRCS and CDWR is often updated within the first week of each month and corresponds to the reservoir capacity at the end of the previous month. Only major reservoirs are displayed on this map. There are other, smaller water impoundment structures in each state.

Interpretation Recommendations:  
  • This map provides a quick assessment of regions where reservoirs have a lot or a little excess capacity to store water. Reservoirs that are full may have to allow water in streams or rivers to pass through the reservoir.
  • Reservoirs that are near full capacity may not be able to mitigate flood risk as effectively as reservoirs that are less full. However, the specific flood mitigation ability of each reservoir cannot be gleaned from this image. It should be used as a general guide.
  • Reservoir storage increases and decreases seasonally. Storage in reservoirs in Arizona, for example, tend to increase in the spring as snowmelt increases.
Additional Resources:  

 

Climate Outlooks

Figure 09. ENSO-Precipitation Risk

Figure Updated:  
Last Thursday of month
Figure Source and Data:  
CLIMAS;
Figure and Data Source:  
The Climate Assessment for the Southwest (CLIMAS) at the University of Arizona produced this map using the NOAA Climate Prediction Center’s Daily U.S. Unified Precipitation Dataset1. In this dataset, daily precipitation values are interpolated from stations within and around each 0.25 degree latitude by 0.25 degree longitude grid (or 28 km latitude by 21 km longitude), which equates approximately to an area of 600 km2. At this resolution, there are 300 by 120 grid boxes encapsulate the U.S. More than 8,000 stations contribute to this dataset, which is updated each day and includes data from 1948 through the present. The interpolation uses the Cressman Technique2, which is generally more accurate than other simple methods such as linear interpolation. However, it is sensitive to reporting errors and the number of neighboring stations and can have more errors in mountainous areas due to large precipitation changes with altitude.
Figure Display:  

The map shows the influence of the El Niño-Southern Oscillation (ENSO) on seasonal precipitation; seasonal precipitation is tallied over three consecutive months. On the map, locations with blue colors represent regions where extreme wet was more common than extreme dry during a defined ENSO event and season. Red colors are the opposite. “Extreme” wet and dry are defined as being above and below the specific threshold percentiles noted on the image, respectively (see below for percentile definition). The blue regions, for example, are areas where precipitation was equal to or above the upper percentile (i.e., 66th percentile) more often than it was equal to or below the lower percentile (i.e., the 33rd percentile). “More often” is quantified in the scale. If the scale number equals 5, then precipitation was equal to or above the upper threshold five times more often than the lower threshold, and vice versa for red regions.

Percentiles indicate the value below which a given percentage of observations fall. For example, a value equal to the 33th percentile is above 32.9 percent of the observations. A value equal to the 66th percentile is greater than 65.9 percent of the observations.

Only the areas that are statistically significant are assigned a red or blue color. Statistical significance means that the displayed risk (i.e., five times greater) for extreme precipitation did not result by chance, which insinuates that the ENSO precipitation influence is stronger in these regions than white-colored regions.

Statistical significance for each ENSO phase and for each three-month season was found by using a nonparametric permutation approach that compared the observed ENSO risk to the ENSO risk compiled by 1,000 unique iterations. In this appraoch, each iteration randomly selected a specified number of seasons from the historical record. The specified number was determined by the actual number of ENSO phases that occurred in the historical record for the season analyzed. Chances of extreme wet versus extreme dry were then computed using only the specified number. A grid was statistically significant if the observed risk was greater than 95% of the risks generated by the 1,000 random iterations.

The approach used here to generate this image is similar to that used by Wolter et al., 1999^3. Some key differences include the datasets used (which here included the CPC precipitation^1 and the Ocean Niño Index^4) and the local significance testing. Here, we use the following procedures:

  1. Based on the entire period of record entire record, find the precipitation value at each grid that corresponds to the upper and lower threshold (33rd and 66th percentile, respectively) for that particular season;
  2. For each ENSO phase (ENSO Nuetral, El Niño, and La Niña) and for each grid point, calculate the average precipitation for each 3-month season in which that event occurred;
  3. For each ENSO phase and for each grid point, calculate the number of seasons in which precipitation was equal to or greater than the 66th percentile value and equal to or below the 33rd percentile value;  
  4. Calculated the relatively chances that precipitation was equal to or greater than the 66th percentile value and equal to or below the 33rd percentile value (if precipitation was above the 66th percentile 5 times and below the 33rd percentile 3 times, than it has been 1.66x more likley that precipiation has been above the upper percentile than the lower percentile;
  5. For each ENSO phase and for each grid point, use a nonparametric permutation approach to calculate local signficance (see above) 

 

Interpretation Recommendations:  
  • This map shows areas in red or blue colors where ENSO appears to have a stronger influence on extreme seasonal precipitation.
  • In the blue-colored grids, extreme wet precipitation is more likely than extreme dry precipitation and therefore indicates regions with a higher potential disaster risk.
  • The ENSO signal is variable, so a large number of neighbor grids with the same color is more robust than one single colored grid.
  • We chose to look at the odds of extreme percentiles because the extreme wet seasons are more likely to create conditions favorable to flooding. However, flooding has occurred within relatively dry seasons.
Additional Resources:  
  1. Description of data source: http://www.esrl.noaa.gov/psd/data/gridded/data.unified.html#detail
  2. Cressman Interpolation Technique: http://iridl.ldeo.columbia.edu/dochelp/StatTutorial/Interpolation/#Cressman
  3. Wolter, K., R. M. Dole, and C. A. Smith, 1999: Short-Term Climate Extremes over the Continental United States and ENSO. Part I: Seasonal Temperatures. J. Climate, 12, 3255-3272.
  4. The NOAA Climate Predictions Center's Ocean Niño Index ENSO index: http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/enso...

Figure 10. ENSO Forecasts

Figure Updated:  
1st and 3rd Thursdays of month
Figure and Data Source:  
The International Research Institute for Climate and Society1 (IRI) at Columbia University produces this image in collaboration with the NOAA Climate Prediction Center (CPC). Probabilities are generated by creating numerous simulations of the tropical sea surface temperature (SST) anomalies in a specified region in the east-central Pacific Ocean (referred to as the Nino3.4 region) from different dynamical and statistical models. Each model simulation produces a slightly different evolution in time of SSTs as a result of different configurations of the models. The IRI also produces a spaghetti plot of each model simulation, which shows the variability of the models2. Probabilities for three-month seasons are then calculated from the suite of model simulations. For example, if one of the 10 model simulations produces an SST anomaly of -1.0 degree C (La Niña events are characterized by a temperature anomaly of less than -0.5 degrees C), then there is a 10 chance that a La Niña will occur in that season. This is one of several ways IRI assesses ENSO probabilities, but it is the most commonly referred to method.
Figure Display:  

For each three-month season, the heights of the bars indicate the current probabilities for La Niña (blue), ENSO-neutral (green), and El Niño (red) events. The sum of the three probabilities equals 100%. The solid colored lines represent the average probabilities of each ENSO phase occurring in a three-month season. These averages are calculated based on the historical record. For example, 22 El Niño events, 22 La Niña events, and 20 neutral events occurred during the December-January-February (DJF) period between 1950 and 2014, according to the NOAA CPC3. This creates average probabilities of about 34 percent for El Niño (22/65*100). Probabilities for La Niña and neutral are calculated in the same way.

Interpretation Recommendations:  
  • It is very unlikely to have a 100 percent probability for an ENSO phase.
  • The probabilities are most precise for the upcoming three-month season. However, prior to the March-April period each year, ENSO forecast models have difficulty forecasting the ENSO phase for future seasons.
  • The ENSO forecast becomes less accurate as the forecasted season increases.
  • El Niño and La Niña events tend to materialize in the fall and winter, which is reflected in the higher average probabilities (solid, colored lines).

Figure 11. Precipitation Forecast

Figure Updated:  
3rd Thursday of month
Figure and Data Source:  
The NOAA Climate Prediction Center (CPC) produces this image from a combination of objective analytical tools and subjective, consensus-driven discussion. The objective tools include historical precipitation amounts during ENSO phases, dynamical and statistical climate models, and precipitation trends, among others. These forecast tools are discussed and areas are drawn on the map indicating where climate signals favor above-, below-, or near-normal precipitation in upcoming seasons.
Figure Display:  

This map show the probabilities that total precipitation in three-month seasons will fall into one of three percentile categories. The map does not show probabilities of precipitation falling above or below average. There are three categories and not two. The lowest tercile category is defined as the precipitation values that equal or are below the 33rd percentile value found between 1980 and 2010. In other words, in this 30-year period, the 10th driest season equals the 33rd percentile value. The highest tercile category is defined as the precipitation values that equal or exceed the 66rd percentile value found in the 1980–2010 period. The middle tercile represents values that fall in between.

The contours on the map convey the probabilities of total seasonal precipitation falling in the upper, lower, or middle terciles. At any point on the map, the sum of the probabilities of these three categories is 100 percent. Colors on the map are assigned when a category’s odds exceed 33 percent (e.g., green colors denote when probabilities for seasonal total precipitation in the upper tercile are above 33 percent). There are numerous combinations of the probabilities for each above, below, or equal chances1 probability.

The CPC also produces a corresponding map that shows the precipitation anomaly associated with the forecast2.

Interpretation Recommendations:  
  • An equal chances forecast means all the forecast tools consulted do not shift the odds in any of the categories. There is a 33 percent chance that seasonal total precipitation will fall in each category.
  • If a location on the map is covered by the “above” and is within a probability contour of 40 to 50 percent, the interpretation is the following: “there is a 40–50 percent chance that total seasonal precipitation in, for example, southeast Arizona will be similar to the precipitation during the wettest 10 years in the 1980–2010 period.”
  • This map does not provide information on the intensity of precipitation. It is possible to have numerous floods within a dry season.
  • The influence of ENSO on precipitation is often represented in this maps. If a La Niña is expected to be present during the December–February season, for example, southern Arizona and California often have increased chances for “below” precipitation.