TEXAS A&M UNIVERSITY

Office of the Texas State

Climatologist

ASSESSMENT of HISTORIC
and FUTURE TRENDS of

EXTREME WEATHER
IN TEXAS, 1900-2036
1
Cover photo: The San Bernard River became a tree-lined field during the 2011 drought.
Photo by John Nielsen-Gammon

REPORT AUTHORS
John Nielsen-Gammon, Department of Atmospheric Sciences, Texas A&M University
Jacob Escobedo, Department of Atmospheric Sciences, Texas A&M University
Catherine Ott, Environmental Programs, College of Geosciences, Texas A&M University
Jeramy Dedrick, Department of Atmospheric Sciences, Texas A&M University
Ali Van Fleet, Department of Atmospheric Sciences, Texas A&M University

MARCH 5, 2020, DOCUMENT OSC-202001

 INTRODUCTION

Texas is vulnerable to a wide range of natural accurate or consistent over time to yield reliable trends.
hazards, most of which are associated Second, natural climate variability and the randomness of
with weather and climate events. The extreme weather events can mask or even overwhelm any
natural environment has evolved partly in underlying long-term trends. Despite these limitations, there
response to these natural hazards. For example, are sound reasons to expect continued change in a variety
plant hardiness is largely determined by ability of aspects of extreme Texas weather, and knowledge of such
to survive extreme winter cold and drought. The built likely changes can be very useful in a variety of planning
environment, including for example homes, roads, and power contexts.
plants, is designed to a certain level of resiliency to natural
hazards. Human activities as fundamental to survival as food Trends are not the only potentially useful historic information.
production and water supply are tailored to the particular Some natural climate variations that occur on multidecadal
combination of weather and climate risks at play in a given time scales have a substantial effect on Texas weather.
location. The present-day scientific ability to accurately predict such
variations twenty years into the future is quite limited.
The future of Texas depends on its resilience to the natural Nonetheless, knowledge of these variations informs the
hazards of the future. It is up to Texans, both individually and understanding of past trends and suggests whether recent
collectively, to decide what level of resilience is appropriate, weather patterns are representative of typical future
and at what cost, compared to the costs of damage and conditions.
recovery on both an economic and societal level. Nobody
knows which specific weather and climate events will befall This report addresses historical and future trends in extreme
Texas over the next couple of decades. But a wide variety of temperatures, extreme precipitation, severe thunderstorms,
information can be used to estimate the risks of certain types and hurricanes. It also addresses trends in drought, floods,
of weather and climate events over that period. wildfire, and coastal erosion, to the extent that these natural
hazards are affected by changes in weather and climate. For
The standard practice for estimating the risk of natural each natural hazard, the report considers the quality of the
hazards has been to assume that future risk is equal to historical data, the historical risk and trends (data permitting),
historical risk. This practice works if the underlying climatic the causes of any observed or expected trends, and the
conditions are unchanging. However, Texas climate is projection of trends of future risk. For context, this report
affected by changing patterns of vegetation, irrigation, and also considers trends in annual average temperature and
urbanization. Texas climate is also embedded in the global precipitation.
climate system, which is itself changing. All of these factors
have influenced historical trends in weather and climate Expected typical conditions in 2036 are expressed as a change
extremes and will continue to influence trends in the future. compared to average conditions in 1950-1999 and 2000-2018.
Given a changing climate, historic trends may provide a better With all projections, there is considerable uncertainty as to
guide to future risk than mere historical averages. how things will actually turn out. This report does not attempt
to quantify that uncertainty; prudent planning recognizes that
The sponsors of these projections requested that projections we cannot know whether reality will end up higher or lower
be based primarily on existing trends. Doing so makes than the best available present-day estimates.
sense only if the causes of those trends are understood and
are expected to continue. The scientific understanding of This report was commissioned and sponsored by Texas 2036.
the causes of trends draws upon a large body of research, The report content is solely the responsibility of the authors.
utilizing both observations and experiments with global A previous version of this report has been peer-reviewed;
climate models. This report presents trends for a variety of reviewer comments and responses are available from the
historic periods, and the projections are based on historic Office of the State Climatologist. The analyses that are original
trends that are expected to continue according to currently to this report are based on data from publicly accessible data
available science. 1 sources; analysis spreadsheets and software are available
from the Office of the State Climatologist.
Several factors influence our ability to project historical trends
into the future. First, historical data may not be sufficiently

3
 AVERAGE TEMPERATURES low-emissions pathway RCP 4.56, is around 0.64 °F per
decade. Up to mid-century, climate projections are not
While average temperatures do not themselves very sensitive to the choice of emissions pathway.7
constitute weather and climate extremes, changes in
average temperatures, either locally or globally, affect On the whole, the agreement between models and
many aspects of extreme weather and climate trends. observations is decent. Factors that cause observed
In addition, all else being equal, a change in average trends to differ from simulated trends include
temperature would lead to a change in frequency of inadequacies in the models, inaccuracies in observations,
extreme temperatures, increasing hot extremes and natural variability, and local land surface changes such as
decreasing cold extremes. irrigation and afforestation.

The National Centers for Environmental Information Historical data and climate models lead to similar
maintain very good analyses of monthly averages of conclusions. If recent trends continue, as expected,
daily maximum and minimum temperatures from 1895 a middle-of-the-road estimate of the overall rate of
to present throughout the lower 48 states.2 The annual temperature increase in Texas would be about 0.6 °F per
average temperature in Texas over the period 1895-2018 decade. This means that average Texas temperatures
exhibits year-to-year variations of 2 °F or more. Broadly in 2036 should be expected to be about 1.6 °F warmer
speaking, Texas temperatures climbed gradually until than the 2000-2018 average and 3.0 °F warmer than
the mid-1950s, declined until the late-1970s, and rose the 1950-1999 average. This would make a typical
thereafter. year around 2036 warmer than all but the absolute
warmest year experienced in Texas during 1895-2018.8
The rate of temperature increase since 1895 has
averaged 0.11 °F per decade, less than the global average Temperature trends since 1975 have been similar in all
of 0.14 °F per decade.3 Indeed, the southeastern United seasons except for summertime temperatures, which
States, including eastern Texas, is almost the only land have been increasing at a slower rate than those in other
area on the globe whose temperature increase over the seasons. The 1950-2018 July-August trend is 0.10 °F per
20th century was nearly zero.4 More recently, the Texas decade, while the 1975-2018 July-August trend is 0.45 °F
temperature trend has been larger. Since 1950, the trend per decade. It is not known whether this slower trend is
is 0.26 °F per decade, and since 1975, 0.63 °F per decade. due to historical factors that will continue or that have
Recent temperatures have increased in all seasons and in already played out. Possibly the transition from open
all regions of Texas. grassland to shrubland and forest across much of Texas
over the past century has led to a reduced summertime
The historic Texas temperature trend simulated by warming trend.9 If so, since that transition seems largely
global climate models since 1950 is 0.29 °F per decade, complete, an assumed future summer trend similar to
and since 1975, 0.51 °F per decade.5 The simulated that of the other three seasons seems reasonable.
current rate of increase in Texas, based on climate
model simulations and projections for 2000-2040 for the

4
 Temperature Trends, 1975-2018

Temperature (°F Increase/Decade)

0.80 - 0.85
Right: Temperature trends
0.75 - 0.80
since 1975 according to
NCEI nClimDiv data. 0.70 - 0.75
0.65 - 0.70
Bottom left: Texas
annual average statewide 0.60 - 0.65
temperatures (°F). 0.55 - 0.60
Bottom right: Comparison 0.50 - 0.55
of observed and simulated 0.45 - 0.50
temperature trends in Texas.

ANN Trend since 1950 Trend since 1975 Trend since 1895
Simulations Observations
Texas Average Temperature Trends
68 1.2

67
0.9
Degrees Fahrenheit per Decade

66
0.6

65

0.3

64
3
0
63

-0.3
62 1895-2018 1950-2018 1975-2018 2000-2040
1895 1915 1935 1955 1975 1995 2015
Texas annual average statewide temperatures (°F)

5
 EXTREME TEMPERATURES 60
The projected changes in average temperature imply
changes in unusually high or low temperatures as well.
This assessment of extreme temperatures relies on two 45
Average Number of 100 °F Days

aggregated sets of temperature data: a set of stable

urban, semi-urban, and rural stations (hereafter referred
to as index stations)10 and a set with one composite 30
station per county.11

In Texas, a good benchmark for extreme heat is the

15
number of 100+ °F days in a given year. The number
of 100-degree days is closely related to the average
summertime temperature. At rural and semi-urban
0
index stations, where 2000-2018 July-August average 78 80 82 84 86 88

temperatures average around 83 °F, there are typically Average July-August Texas Temperature

about 12 days per year that reach or exceed 100 °F. If

summertime temperatures rise at a similar rate Historical relationship between summertime average temperature
as the projected annual Texas average, the typical and number of 100°F days at selected long-term non-urban
number of 100-degree days would nearly double, to stations in Texas
about 21 per year, by 2036.

To analyze actual trends in 100 °F days in Texas, the all types of stations, although the longer trend since 1950
index stations in each station class were grouped into generally has less than a doubling. Thus, a near-doubling
four regions and conditions were averaged across of the number of 100-degree days between 2000-2018
stations within each class and region, then the values and 2036 appears to be a reasonable projection.
from each region were averaged together to obtain
statewide averages by class.12 Because the above graphic Urban stations, which are in major metropolitan areas,
shows that 100 °F day counts are a nonlinear function have seen a more dramatic increase in 100-degree days.
of temperature and because counts cannot be negative, The frequency has approximately tripled, whether one
trends were analyzed on the basis of a linear fit to the considers the shorter 1975-2018 period or the longer
5
time series of the logarithm of annual 100 °F day counts. 1950-2018 period. There are presently more 100-degree
A plot of number of days reaching 100 °F at index stations days in major metropolitan areas than elsewhere. This
shows some memorably hot summers: 1980, 1998, and suggests that the urban heat island effect has enhanced
2011. Over the past 40 years, the linear trend shows a the risk of extreme heat in major cities in Texas, above
more than doubling of the number of triple-digit days at and beyond the trend due to larger-scale climate change.

6
 Days Over 100°F 57
44
37 33 38
30

25

20

15

10

0
1950 1960 1970 1980 1990 2000 2010 2020
Urban Stations Semi-Urban Stations Rural Stations
Trend Since 1950 Trend Since 1950 Trend Since 1950
Trend Since 1975 Trend Since 1975 Trend Since 1975

Trends and historic variability in index temperature stations in Texas, grouped by amount of urbanization. Arrows at top
indicate data points that are above the margin of the graph. Linear trend lines are fit to the logarithm of 100 °F day counts
to ensure non-negative values.

Since 1975, 100 °F days have increased at least as rapidly contrast to mean and extreme temperatures in all other
at rural stations as at urban stations. The excess heat seasons, may be related to the land cover trends such as
at rural stations has primarily occurred during the most were mentioned earlier.16 Whatever the cause of the long-
extreme summers (1998, 2000, 2009, 2011), while urban term slight downward trend, it appears to have reversed,
stations tend to have the most 100 °F days during less with a slight increase in extreme monthly heat in recent
extreme summers. This is consistent with evidence decades.
suggesting that, as temperatures increase, the drying of
rural soils can lead to an excessive increase in daytime Extreme low temperatures during the winter months
temperatures,13 while urban moisture varies less due to exhibit a stronger and more robust trend.17 Despite year-
large expanses of impervious surfaces. to-year fluctuations that are much larger for extreme cold
than extreme heat, there is a long-term warming trend
An alternative way of examining extreme heat trends is in monthly extreme cold temperatures across Texas, and
to consider the average hottest day in each month during the trend is larger in recent decades. Extreme cold at
June-September in each county in Texas.14 This metric urban index stations has warmed at a greater rate than at
is more sensitive to exceptional heat events than to rural index stations.
sustained heat. It also enables us to use longer historical
records, as this metric is not sensitive to changes in At all time periods, extreme cold is warming fastest,
observing time. while extreme heat is warming slowest. This is broadly
consistent with expectations: extreme cold air comes
Looked at this way, there is a very slight downward trend from the Arctic, which in general is warming faster than
in extreme heat over the long term. This trend reflects other parts of the globe. There have been some studies
the general pattern over the continental United States, in recent years debating whether loss of Arctic sea ice
where the most exceptional heat occurred back in the and overall Arctic warming leads to changes in weather
1930s.15 The cause of this downward trend, which is in patterns that favor more intense incursions of Arctic
6
7
 air18, but this tendency, if present, has not been strong Extreme monthly summertime temperatures averaged
enough in Texas to stop the accelerating rise of extreme 97.8 °F during 1950-1999 and 98.3 °F during 2000-2018.
cold temperatures. This warming of extreme cold Based on the 1975-2018 trend, the expected value of
temperatures in Texas, much more rapid than warming extreme monthly summertime temperatures would be
of average temperatures, is consistent with what has 98.9 °F by 2036, an increase of 0.6 °F compared to the
been observed elsewhere in the Northern Hemisphere 2000-2018 average and 1.1 °F compared to the 1950-
and is inconsistent with the argument that loss of Arctic 1999 average. Meanwhile, extreme monthly wintertime
sea ice is enhancing extreme cold over the continents.19 temperatures averaged 18.5 °F during 1950-1999 and
Note that there has not been a winter with extreme cold 20.8 °F during 2000-2018. Based on the 1975-2018 trend,
temperatures above 24 °F in over two decades, despite extreme wintertime temperatures would increase 3.3 °F
the upward trends, which could be a statistical aberration by 2036 compared to the 2000-2018 average and 5.6 °F
or a consequence of the aforementioned changes in compared to the 1950-1999 average.
weather patterns.
Trends in extreme cold are much larger than trends in
Overall, extreme heat is becoming more frequent and extreme heat, which is leading to an overall decrease
more severe, while extreme cold is becoming less in the range of annual temperature extremes over
frequent and less severe. The temperature extreme time. That decrease is likely to continue. The extreme
projections here are based on the 1975-2018 trends, heat projections remain largely within the range of
since temperature trends themselves are well simulated historical variability, but the increase in the number
over that period by climate models and are projected by of hot days is such that the typical number of triple-
models to continue at a similar pace. digit days by 2036 is projected to be larger than all
but four years since 1950.20 Meanwhile, the expected
The number of 100-degree days at typical stations is warming of extreme wintertime temperatures would
expected to nearly double by 2036 compared to 2000- make typical wintertime extremes by 2036 milder
2018, with a typical year having 25 triple-digit days than all but five of the winters in the historic record.21
at urban and rural stations and 20 triple-digit days Also, trends in extreme temperatures in urban areas
at semi-urban stations. Triple-digit counts will tend to are generally as large or larger than those in rural
be larger toward the south and away from the coast, and areas, so that extreme heat is an increasing large risk
smaller elsewhere. in urban areas and extreme cold is an increasingly
remote one.

8
 Extreme JJAS Temps 1950-2018 1975-2018 1895-2018

110

106

102

98

94

Hottest day of the month, June-September, at

composite county stations across Texas, with linear
90
1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 trends beginning in 1895, 1950, and 1975.

Extreme DJF Temps 1950-2018 1975-2018 1895-2018

28

24

20

16

12

Average coldest day of the month, December-February,

8 at composite county stations across Texas, with linear
1890 1900 1910 1920 1930 1940 7
1950 1960 1970 1980 1990 2000 2010 trends beginning in 1895, 1950, and 1975.

Extreme Cold Average Extreme Heat

1.2

0.9
Degrees Fahrenheit per Decade

0.6

0.3

Comparison of the sizes of the trends in average

-0.3 8 temperatures and extreme heat and cold in the
1895-2018 1950-2018 1975-2018 county data

9
 PRECIPITATION models, with the average simulated trend being -2.6%
per century.23 Models and observations both tend to
Precipitation in Texas is quite variable, both in space and feature more positive (or less negative) trends toward
time. Much of the state has two rainy seasons, with the northeastern Texas than toward southwestern Texas.24
rainiest months on average being May, June, September,
and October. In far West Texas, the wettest months are Superimposed on the generally upward precipitation
July and August, while far East Texas averages similar trend is considerable variability. El Niño has a prominent
amounts in every month. Rainfall amounts increase from influence on cool-season rainfall in Texas: during El
west to east, with the southeast corner of the state near Niño years, Texas tends to be wet, while during La Niña
Beaumont averaging over eight times the annual rainfall years, Texas tends to be dry. Individual El Niño or La Niña
of some areas near El Paso. events tend to last only 1-3 years, so this relationship is
not useful for multidecadal predictions. However, both
The long-term trend of precipitation in Texas has been the Atlantic and Pacific Oceans feature multidecadal
positive. Over the past century, parts of central and variability: the Atlantic Multidecadal Oscillation (AMO)
eastern Texas have experienced precipitation increases and the Pacific Decadal Oscillation (PDO). Both have
of 15% or more, while in much of the western part of been shown to substantially influence precipitation in the
the state the long-term trend is flat or even downward.22 United States, including Texas.25
The tendency for increasing precipitation in Texas
is not consistent with the majority of global climate

Precipitation Trends, 1895-2018

Precipitation (% change/century)
15.0 - 21.0
10.0 - 15.0
5.0 - 10.0
0.0 - 5.0
-6.3 - 0.0

Precipitation trends since 1895 according to NCEI nClimDiv data.

10
 Texas Decadal Precipitation and Natural Drivers
25

20

15

10

0
1900 1920 1940 1960 1980 2000 2020
MAMJ JASO NDJF -AMO M-O PDO N-J
MAMJ JASO NDJF -AMO M-O PDO N-J

Solid: Average Texas precipitation, inches, nine-year running averages, for March-June (green), July-October (orange), and November-February
(gray), with dots showing values for last nine individual years. Dashed light green: Pacific Decadal Oscillation, November-June, nine-year
running average, with squares showing values for last nine individual years. Dashed dark red: Atlantic Multidecadal Oscillation, March-
October, sign reversed, nine-year running average, with squares showing values for last nine individual years. Both AMO and PDO have been
rescaled and offset for ease of comparison with Texas precipitation.26

The graph of decadal-scale variations of average Texas decades, this implies that variability is likely to have a far
precipitation illustrates the distinct seasons during which larger effect on precipitation changes than any long-term
the AMO and PDO influence Texas precipitation. The AMO climate trend. Unfortunately, neither the AMO nor PDO
was relatively high in the 1930s, 1950s, and 2000s and low can reliably be predicted two decades into the future.
in the 1900s, 1970s, 1980s, and early 1990s. The negative The last nine years illustrates that the PDO is much more
of the AMO is plotted here so that upticks in the graph erratic than the AMO. Also, given the extreme annual
correspond to upticks in Texas precipitation. The AMO precipitation in some recent years and the dry year of
pattern can be detected in the March-June precipitation 2011, the nine-year precipitation averages will almost
(correlation = -0.27) and July-October precipitation certainly exhibit a short-term upward blip.
(correlation = -0.47). Meanwhile, the PDO was relatively
low near 1920, 1950, and 2010, and relatively high near Both the AMO and PDO favored unusually dry conditions
1940, 1980, and 1990. This pattern can be detected in the in Texas during the first part of the 21st century. Since
November-February precipitation (correlation = 0.53) and that pattern is unlikely to recur given the slow variability
March-June precipitation (correlation = 0.54). So the AMO of the PDO and AMO, the expectation is that rainfall
influences Texas precipitation on a decadal scale from in Texas over the next two decades will tend to be
spring to fall, while the PDO influences Texas precipitation greater than what was experienced from 1998 to
from late fall through spring. 2012. The positive long-term precipitation trend
further supports that expectation. The climate model
The decadal-scale variability is so large that the eye has projections provide weak evidence for a precipitation
difficulty detecting the aforementioned long-term upward decline. Also projected is a change in the seasonality of
trend in Texas precipitation. Over the next couple of rainfall in North Texas, with less rain in the summer.27
11

11
 EXTREME RAINFALL
Many studies have documented an increase in extreme
rainfall in Texas and surrounding areas for a variety of
durations and thresholds.28 On average across the region,
extreme one-day precipitation has increased by 0.2” to
0.5” since the middle of the 20th century.29 Within Texas,
the local experience of extreme rainfall varies widely from
place to place, with some locations having experienced a
decrease in intensity of extreme rainfall over the period
of data availability while the majority of locations have
experienced an increase.30 The median change since 1960
is an increase in intensity of 7%.

All other things being equal, an increase in overall

precipitation amounts would be expected to lead to
an increase in extreme precipitation amounts, which
inevitably implies an increase in the frequency of extreme
precipitation above a given threshold. So the overall trend
in Texas precipitation, discussed previously, contributes
to the observed trend in extreme precipitation
probabilities.

In addition to the overall precipitation effect, extreme

rainfall is strongly affected by increased temperatures. A
column of air that is producing rainfall will produce about
4% more rainfall for every °F of warming.31 The extra
precipitation intensity can also affect storm structure,
and climate change can alter weather patterns and the
frequency of dangerous storms. The direct temperature
effect appears to be most important over the long haul
in most midlatitude locations such as Texas. Changes
in storm motion and frequency are expected to be
comparatively subtle.32

As noted earlier, computer model projections of overall

rainfall amounts in Texas are somewhat inconsistent,
but in general they show an overall leveling off or slight
decrease of precipitation amounts.33 This suggests that
trends of extreme precipitation in the future will be
dominated by the increasing temperature effect and
changes in storm structure. The strongest influence
should be temperatures near the moisture source for
Texas extreme precipitation, that is, the tropical oceans
and the Gulf of Mexico. Tropical ocean temperatures
are not expected to rise as quickly as temperatures over
land in Texas, but would still be sufficient to produce an
additional 2%-3% increase in extreme rainfall intensity
from the temperature effect alone.34 So extreme rainfall
intensity and frequency are projected to continue
increasing, though probably not as rapidly as they have
increased in the past.

Rainfall risk is often characterized in terms of the 100-

year rainfall event, which is an amount of rain over a
given duration that has a 1% chance of occurring in any
given year. If extreme rainfall amounts increase by just

12
20%, the 100-year rainfall event
threshold is exceeded twice as
often.35 So Texas’s 7% median
increase in the 100-year rainfall 100 Year Return Period, One-Day Rainfall
amount between 1960 and 2017
corresponds to a roughly 30%
increase in the frequency of

±
heavy rainfall exceeding the older
100-year threshold. Based on
projected temperatures and
the dominance of the direct
temperature effect on extreme
rainfall, we anticipate an
additional increase of 2%-3%
in expected extreme rainfall
intensity in 2036 compared to
2000-2018 and an overall increase
of 6%-10% compared to 1950-
1999. These changes in amount
correspond to increases in the
odds of extreme precipitation
of 10%-15% and 30%-50%,
respectively.

Note that the variations of extreme Percent Change, 1960-2017

rainfall trends across Texas imply
> 40 -10 - 0
that one should not assume that
recent extreme rainfall history 30 - 40 -20 - -10
in a given location is a suitable 20 - 30 < -20
baseline for projecting future 10 - 20 Insufficient Data
trends. While estimates of extreme
0 - 10
rainfall risk based on historical data
show a large uptick in Houston
but little change in Dallas-Fort
Worth36, climate change should
be acting to increase the risk
more uniformly across the entire Percentage change in the intensity of one-day rainfall with a 1% probability of occurring
state. The projected increase of in any given year, based on a time-dependent statistical fit of annual daily precipitation
extreme rainfall risk given above maxima at composite stations.
is relative to the expected past
risk of extreme rainfall in a given
location, not the actual occurrences
of extreme rainfall in that location.
Likewise, the actual extreme rainfall
over the next 20 years at any given
location may defy the odds, either
favorably or unfavorably.

12

13
 DROUGHT The corresponding drought severity based on
precipitation alone, as measured by the cumulative
At first glance, the long-term increase in average rainfall 9-month Standardized Precipitation Index, does indicate
should imply a decrease in drought as well. The linear declines in severity. The change in cumulative dryness
trend in total statewide precipitation since 1895 is +0.85% is -6.0% per decade since 1896, -14.4% per decade since
per decade, since 1950 is +2.34% per decade, and since 1950, and +15.7% per decade since 1975. As noted with
1975 is +0.87% per decade.37 The large trend since 1950 respect to precipitation, the change since 1950 is large
is due to the extended dry conditions beginning around in part because of the extended drought that occurred
1950, especially elevating the trend since 1950 because of early in the 1950s. The rise of cumulative dryness since
such a low starting point. 1975, however, has happened despite a slight increase of
precipitation over the same period.
Texas Statewide Precipitation
45 The explanation for this difference can be seen in the
statewide precipitation graph. Note that the period of
40
time from 1965 to 1985 featured very little precipitation
35 variability, with statewide average precipitation ranging
between 22” and 35”. Since then, precipitation has
30
become more variable, with 2011 precipitation below 14”
Precipitation (inches)

25
and 2015 precipitation above 41”. Greater precipitation
variability leads to more intense droughts even if the
20 overall precipitation amount doesn’t change.
15

10

0
1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005 2015

Statewide average precipitation, from NCEI’s nClimDiv dataset, with

linear trends since 1895, 1950, and 1975.

14
 9-Month SPI: Overall Yearly Drought Severity
2.5 1890s to 2018 1950 to 2018 1975 to 2018

40

2 30

20

1.5 10

1 -10

-20

0.5 -30

-40
3-Month SPI 9-Month SPI 24-Month SPI PDSI

0
1896 1906 1916 1926 1936 1946 1956 1966 1976 1986 1996 2006 2016

Comparison of short-duration (three months), medium-duration

Statewide annual cumulative drought severity, with linear trend (nine months), and long-duration (24 months) cumulative annual
lines beginning in 1896, 1950, and 1975. 38 drought severity based on precipitation, as quantified using the
Standardized Precipitation Index (SPI). Also shown are trends in the
cumulative drought severity based on the Palmer Drought Severity
Index (PDSI), whose variations are most strongly correlated with the
The historic record of precipitation variability is, itself, 9-month SPI.
quite variable, with no clear trend over the period of
record despite the recent uptick in variability. Climate
models generally simulate an increase in interannual However, while the SPI neglects the temperature effect
precipitation variability in Texas since 1950 but little or on drought entirely, the PDSI may overestimate it.43 This
no change in interannual precipitation variability going is because, in addition to precipitation and temperature,
forward.39 The lack of future variability is itself anomalous carbon dioxide also affects drought. Elevated carbon
compared to the rest of the Northern
14 Hemisphere land dioxide levels improve the water use efficiency by plants,
areas, which tend to show an increase in variability in so would lead to increased soil moisture and decreased
excess of the increase in average precipitation itself.40 drought. Elevated carbon dioxide
15 levels also increase
So the recent trend of increased interannual variability biomass if plants are not otherwise water- or nutrient-
is probably not going to continue apace through 2036, limited, which might increase water use and decrease
unless average precipitation itself continues to increase, soil moisture. While these two effects work in opposite
and the precipitation variability experienced over 2000- directions, the water use efficiency effect seems to be
2018 is probably representative of what should be dominant, thereby allowing increased CO2 to lead to
expected through 2036. improved plant growth despite meteorological factors
leading to increased drought, at least on a global scale.44
Temperature also affects drought directly, by increasing Nonetheless, improved plant health overall does not
the rate of evaporation from the soil and from water eliminate possible detrimental effects from drought
bodies. Many drought indices, such as the Palmer accompanied by increased temperatures, particularly
Drought Severity Index, attempt to include the effect of since decreased plant water use itself implies increased
temperature on dryness.41 Compared to rainfall-only SPI temperatures.45 Also, different plants respond to carbon
indices at various time scales, particularly the 9-month dioxide (and heat stress) differently.46 Lastly, global
scale that the PDSI is most responsive to, the PDSI shows climate model simulations that attempt to incorporate
a less negative trend over the entire period and a larger changes in plant physiology still indicate a substantial
positive trend recently.42 decrease in future soil moisture across the Great Plains.47

Increased carbon dioxide does not reduce the

temperature effect of evaporation from lakes and
reservoirs or from bare soil. Historically, there has been
an increase in the evaporative capacity of the atmosphere
across most of Texas, especially in West Texas and
the Panhandle, and that trend is expected to continue
due to robust projections of rising temperatures.48
A continuation of the observed trend would lead

15
 to a roughly 4% increase in expected
summertime evaporative losses from
reservoirs in 2036 compared to 2000-2018.49

For hydrological drought, runoff is important.

Runoff changes depend on changes in
soil moisture and changes in precipitation
amount and intensity. Across the
midlatitudes, there is on average a decrease
of topsoil moisture but little change in deep
soil moisture.50 This would favor a slight
decrease in runoff for a given amount of
precipitation. Countering that is the tendency
for greater rainfall intensity,51 which would
lead to a greater runoff fraction and in turn
would partially cause the decrease in soil
moisture, while a model-simulated change
in precipitation seasonality in some areas
could lead to greater drying in summer when
drought is most severe.52

Despite all the nuances discussed above, the

primary driver for changes in future drought
remains changes in overall precipitation
amount. In Texas, the projected slight
decrease mentioned in the previous section
would imply a small increase of drought
intensity by itself. For agricultural drought,
this small increase of meteorological drought
could easily be neutralized by improved
water use efficiency by plants, leading to
no agricultural drought trend or even a
reduction in drought susceptibility, though
several other factors will come into play,
not least the increasing temperatures. For
hydrologic drought, the increase of surface
water evaporation would worsen the impact
of the increased meteorological drought.

Because of all the factors at play, it

is impossible to make quantitative
statewide projections of drought trends.
The majority of factors point toward
increased drought severity. Nonetheless,
it seems highly likely that any such
underlying trend will be dwarfed by the
impact of multidecadal variability, which
historical records show is large for Texas.
Also, as indicated by paleoclimate records,
worse droughts have occurred in Texas
than the climate data record alone would
indicate.53

16
RIVER FLOODING
Texas has been impacted greatly by
river flooding in the past, causing
both fatalities and economic
damage. In fact, Texas ranked
highest amongst the United States
in flood related fatalities during
1959-2005.54 Throughout most of
the state, intense daily and weekly
precipitation events are the primary
meteorological drivers correlated with
the most extensive flooding.55 This
would imply that a single or a short
series of extreme precipitation events
plays a dominant role in causing
river flooding, and that antecedent
soil conditions have limited impact.
However, east and northeast Texas
are the exception to this pattern, as
extended wet periods are the primary
meteorological drivers best correlated
with extensive river flooding. This
would imply that antecedent soil
conditions- as controlled by seasonal
precipitation and evaporation- play a
dominant role in flood events in those
areas.

Research has found a historical

decrease in the magnitude of
flood events at many river gauges,
resulting from abrupt drops in flood
magnitude.56 This can be attributed
to the construction of reservoirs
and dams for flood management
throughout the 20th century.
As substantial future reservoir
construction is not anticipated,
this trend driver will not continue.
Projected increases in temperatures
and precipitation intensity will have
competing effects on river flooding.
Increased precipitation intensity will
lead to more precipitation-runoff
events that suggest more river
flooding in the future. In contrast,
increased temperatures would
mitigate flooding by decreasing
soil moisture and increasing the
capacity for soil to hold new rainfall.
This would limit the amount of
precipitation-runoff that would go
into rivers and increases in river
streamflow would be lower.

17
 At the seasonal scale, there are mixed historical trends, In summary, river flooding in Texas is projected to
with increased peak streamflow in the winter and have no substantial change through 2036. This is in
decreased peak streamflow in the fall and spring that vary large part due to the construction of dams and reservoirs
by climate zone.57 For the Trinity and Neches River Basins, for flood management in the 20th century. There is a
both in east Texas where annual precipitation is greatest mixture of historical trends categorized by season, and
within the state, most gauges show statistically significant this does not bring forth a clear and coherent trend to
increases in winter extreme streamflow post-1965. project. Also, meteorological drivers of river flooding
While the other three seasons do not have statistically (increased rainfall intensity, decreased soil moisture) are
significant trends, there is still a general increasing trend projected to have competing influences. On balance, if an
in extreme streamflow for the Neches and Trinity river increasing trend is present in river flooding, it will be at
gauges. For the Colorado and Brazos River Basins (where the most extreme flood events or in the wettest parts of
annual precipitation decreases towards the west), there is the state where there is so much rainfall that a decrease
an increasing trend in extreme streamflow in the winter in soil moisture would have little mitigating impact.
since 1965, however, the spring and fall show a general
decreasing trend.

Location and the decade of occurrence of stations that experienced

an abrupt change in the mean of annual maximum floods. Of these,
22 out of the 24 experienced a decrease in the mean. (Adapted from
Figure 4 of Villarini and Smith, 2013)

a) b)

Panels a) and b) shows the

locations of selected river
gauges relative to its river
basin and climate zone, and c)
shows the statistical trends of
1-day 95th percentile seasonal
streamflow trends of each
gauge after 1965. A Mann
Kendall statistic greater than
1.96 indicates a significant
increasing trend while a Mann
Kendall statistic less than
-1.96 indicates a significant
decreasing trend. (Adapted
from Figures 1 and 14 of
c) Mishra et al., 2011)

16

18
URBAN FLOODING Local trends in the future may not necessarily reflect what
has happened in the past. Many of the decreasing trend
Urban flooding differs from river flooding in that gauges are generally the gauges with a comparatively
catchments tend to be much smaller and the effects of longer record of data (over 50 years) which are usually on
urbanization are relevant for the projection of flooding major rivers which have been affected by dam/reservoir
in metropolitan areas. In general, urbanization greatly construction over the record period.60 As mentioned
decreases or even eliminates the infiltration rates of the in the river flood section, little substantial reservoir
soil through the construction of impervious surfaces. construction is anticipated in the future. In addition,
For a given precipitation event, this greatly increases many of the increasing trend gauges tend to have a
the precipitation runoff within a given basin and results comparatively shorter record of data (25-50 years) where
in higher streamflow and flooding for urban rivers. natural variability will have a larger influence on the
Mitigation measures, such as detention ponds, can wholly calculated trend.
or partly counter this effect.
Regardless of the variable historical trends, Texas’s urban
Historic trends of urban flooding are variable and population has increased by over 2% per year during
determined by local flood control factors. The distribution 2000-2010, resulting in the largest urban area and the
of impervious surfaces within a river basin, pre-existing second largest urban population of the United States.61
land surfaces, alterations of land surfaces, pre-existing As Texas’s population continues to grow, its urban area
flood control impoundments that alter runoff flow, the is likely to continue to expand and become denser, and,
terrain within a basin, etc. all vary from city to city in in combination with the projected increase in intense
Texas. Nearly all metropolitan areas in the state contain precipitation, the effects of urbanization would result in
gauges with increasing, decreasing, and neutral trends in an increase in precipitation runoff and urban flooding.
flooding, the majority of those being neutral.58 The degree Such a trend is supported by results from hydrological
of increasing and decreasing trend gauges varies from model simulations.62
city, with the Houston metropolitan region standing out
as a hotspot for increasing urban flooding trends (with Assuming that the flooding trends in small, rapidly-
over 60% of Texas’s metropolitan increasing trend gauges responding urban basins are driven climatologically
in Harris County alone), whereas decreasing trends are by rainfall intensity, the change in frequency of
spread among many metropolitan regions. A recent extreme rainfall would translate directly to a change
study has argued that the mere existence of Houston has in the expected frequency of urban flooding: 30%-50%
intensified rainfall and increased streamflow twentyfold, more in 2036 relative to climatological expectations
but those findings may not be robust.59 for 1950-1999 and 10%-15% more relative to
2000-2018.

19
 Number of examined stream gauges
exhibiting no statistically significant trends,
downward trends, and upwards trends in
peak river flow, grouped by metropolitan
area. (Figure 4 from Berg, 2018)

Closer examination of statistically significant gauge trends in a) Dallas-Fort Worth-Arlington and

b) Houston-The Woodlands-Sugar Land areas (Figure 5 from Berg, 2018)

18

20
WINTER PRECIPITATION air. Freezing rain is most common in
the northeastern part of the state.
Winter precipitation is rare enough
in Texas to be automatically The climate data record for freezing
disruptive in most areas. Snow and rain observations is quite sparse
sleet are most common across compared to snow and rain. Freezing
northern Texas, the Panhandle, rain trends in Texas during the
west-central Texas, and far west last half of the 20th century were
Texas.63 However, snow has been mixed.67 Because freezing rain is
reported on occasion in almost every strongly dependent on temperature
county in Texas. conditions, one might expect a
decrease in the threat of freezing
In the southern part of the state and rain over time. However, the weather
in coastal regions, snow is rare, but patterns that produce freezing rain
nonetheless, large accumulations are rather specific, and changes
of snow are possible. Notable in the frequency of those weather
events include 1895 from Galveston patterns could easily amplify or
to Beaumont (nearly two feet of neutralize any temperature-driven
snowfall accumulation)64, 1985 in trend. So the expectation of a
San Antonio (over a foot), and 2004 decline in freezing rain frequency
in Victoria (nearly a foot).65 Given is tentative, and the magnitude of
these rare events, it is perhaps not such a decline is as yet unknown.
totally unbelievable that the one-day
snowfall record for Texas is held not
by a Panhandle city but by Hillsboro,
north of Waco, with 26” in 1929.
These sorts of extremely unusual but
highly disruptive snow events have
not been studied in a climate change
context and they are too rare for any
trends to be robustly detected in the
historical record.

Farther to the northwest, where

snow is more common, there is no
noticeable trend in the frequency
of 2” or greater snowfalls from
1930 to 2007.66 Because the
determining factor for snowfall
in northwest Texas is typically
air temperature, a reasonable
expectation is that snowfall
frequency and intensity will
decrease in the future, somewhat
reducing the snow hazard.

Freezing rain is highly disruptive

to vehicular travel and to electric
power transmission. Freezing rain
can occur when a warm layer of air
aloft overlays subfreezing air near
the surface. In Texas, this happens
when a strong cold front stalls across
Texas as an upper-air disturbance
picks up warmer, humid air from
over the Gulf of Mexico and carries Average annual snowfall at county index stations.
it northward above the shallow cold

21
 SEVERE THUNDERSTORMS
There is no reliable, long-term record of severe thunderstorms
or the severe weather they produce: tornadoes, hail, and
strong winds. Reporting methods and magnitude scales have
changed over time for tornadoes and hail events.

Maps of the historic distribution of tornadoes, hail, and strong

winds make it clear that no corner of the state is immune
to severe thunderstorms.68 The dense clusters of observed
events around major metropolitan areas such as Dallas-
Fort Worth reflect the enhanced likelihood that a severe
thunderstorm there will be witnessed or will cause damage. It
is also possible to detect lines of severe weather reports along
some major highways, such as Interstate 20 near Midland. The
strong dependence of report location on population density
gives an indication of how changes in severe thunderstorm
reports over time are more strongly affected by changes
in population than by changes in severe thunderstorms
themselves.69

Not only is the climate data record for severe thunderstorms

poor, severe thunderstorms are too small to be simulated
directly by present-day climate models. Therefore, when
assessing trends in severe thunderstorms, it is necessary to
consider indirect indicators of severe thunderstorm frequency
and intensity such as wind shear and convective instability,
both of which favor severe storms.

As temperatures increase, the amount of energy available to

fuel these storms will increase as temperature and low-level
moisture increase.70 Even though shear will likely decrease
as the temperature gradient from the poles to the equator
weakens, the increase in CAPE outweighs any decrease
in low level shear.71 This results in an overall increase
in the number of days capable of producing severe
thunderstorms.

Regarding the specific hazards of thunderstorms, lightning

occurs most often during the months of May and June. Severe
wind is most prevalent during the summer months from
disorganized storm systems in the High Plains of Texas.72
The most robust trend in tornado activity is a tendency
of more tornadoes in large outbreaks, but the factors
apparently driving that trend are not projected to
continue.73 Warmer temperatures are likely to lead to less
hail overall, particular during the summer, but increases
in available thunderstorm energy may lead to an increase
of the risk of very large hail in springtime.74

22
 21

Top: Locations and magnitudes of tornadoes in Texas, 1950-2017. Data from https://www.spc.noaa.gov/gis/svrgis.
Bottom left: Locations and magnitudes of severe hail in Texas, 1950-2017. Data from https://www.spc.noaa.gov/gis/svrgis.
Bottom right: Locations and magnitudes of potentially damaging thunderstorm winds in Texas, 1950-2017.
Data from https://www.spc.noaa.gov/gis/svrgis.

23
23
22
 HURRICANES AND COASTAL EROSION
Sea level rise and storm intensity both affect coastal flooding and erosion.

The change in ocean height relative to coastal lands, called relative sea level rise, is one of
many factors affecting coastal erosion. Relative sea level rise is a combination of three factors:
eustatic sea level rise, local variations in sea level rise, and relative land motion. Eustatic sea
level rise is the change in global mean ocean height and is primarily the result of increasing
temperatures that cause thermal expansion and melting glaciers and ice sheets. Local
variations are produced by changes in wind patterns and ocean currents, and are minor for the
Gulf of Mexico. Relative land motion in coastal Texas is dominated by coastal subsidence.

Station ID RSLR 95% C

Sabine Pass 8770822 1958-2017 5.80 mm/yr +/- 0.77
Galveston Pier 21 8771450 1904-2017 6.49 mm/yr +/- 0.23
Freeport 8772440 1954-2008 4.43 mm/yr +/- 1.05
Rockport 8774770 1937-2017 5.58 mm/yr +/- 0.49
Port Mansfield 8778490 1963-2017 3.05 mm/yr +/- 0.74
Port Isabel 8779770 1944-2017 3.98 mm/yr +/- 0.34

Relative sea-level rise (RSLR) and 95% confidence interval (95% C) at selected Texas tide gauges through 2017.75

Subsidence is a gradual lowering of land-surface elevation and is the result of the extraction of
groundwater, oil, or gas or increasing sediment loading or infrastructure construction. As the
coast of Texas slowly sinks, water potentially encroaches landward so quickly that it can exceed
natural sediment accretion rates. In the state of Texas, the rate of subsidence ranges from
less than 2 mm/yr to as much as 7 mm/yr.76 The variations are due to historical differences in
oil, gas, or groundwater extraction and sediment loading, resulting in generally larger rates of
subsidence in southeast Texas than in south Texas.77

The combination of local subsidence, eustatic sea level rise, and changes in sediment
deposition and transport have produced a retreat of the Texas coastline along nearly the entire
length of its barrier islands.78 In Galveston Bay and probably other bays and estuaries behind

24
 Net rates of long-term change
for the Texas Gulf shoreline
between Sabine Pass and the
Rio Grande, calculated from
shoreline positions between
the 1930’s and 2012. From
Paine et al. (2014).

the barrier islands, sediment deposition is not keeping year. The places along the coast with the largest rates
up with relative sea level rise, leading to loss of coastal of relative sea level rise may have a doubled storm
wetlands.79 In many areas there has been a decrease surge risk by 2050 relative to the risk at the beginning
in extraction of groundwater and other resources as of the 20th century, purely due to the relative sea level
the problems associated with coastal subsidence have rise itself.
become clearer, resulting in a reduction in coastal
subsidence. However, eustatic sea level has shown An additional element of enhanced risk is provided by
indications of acceleration,80 so on balance the near-term an expected increase in the intensity of very strong
future rate of coastline retreat may be expected to be hurricanes, despite an expected lack of increase, or even
similar to historic rates. a decrease, in hurricane frequency overall.82 Different
research studies have produced some conflicting results,
Rising sea levels lead directly to increased risk
25 of storm and local trends over the western Gulf of Mexico will also
surge from hurricanes, as the storm surge is on top of be affected by changes in wind patterns for which global
an elevated baseline. Given typical return periods for climate models have little predictive skill. At this point,
storm surges along the Gulf Coast,81 a 1 meter relative the enhanced risk is difficult to quantify, but substantial
sea level rise produces a doubling of storm surge risk, scientific progress on this topic is likely as climate models
as a surge that would in the past have been expected become better able to simulate the observed spatial
have a 1% chance of occurring in any given year would distribution, frequency, and intensity of hurricanes.83
in the future have a 2% chance of occurring in any given

25
 WILDFIRES Vapor pressure deficit, an indicator of the
ability of moisture to evaporate, is projected
Fire depends on strong winds, low humidity, to increase as temperatures rise and carbon
surface dryness, and factors affecting the dioxide fertilization reduces transpiration,
amount of vegetation available to burn leading to both lower humidity and
(known as fuel load). increased surface dryness.86 Overall,
increased dryness should extend the
There is weak statistical evidence for a wildfire season in places where the fire
slight decline in the frequency of unusually season is presently constrained by low
strong winds speeds across the southern levels of aridity, such as eastern Texas.87
Great Plains.84 One future climate simulation
predicts an increase in average wind speed Changes in fuel load involve two competing
across most of Texas in all seasons.85 On the effects: increased aridity leading to reduced
whole, expectations for wind speed trends plant growth, and increased carbon dioxide
lack robustness. leading to increased plant growth. Multiple
papers predict wildfires in the Southwest
Low humidity and surface dryness are (including Texas) will change differently
closely related. Low humidity refers to than elsewhere in the country as fuel load
the extremely dry weather accompanying gradually becomes the determining limit on
wildfire outbreaks across Texas, while fires.88 This transition in wildfire regime
surface dryness refers to the lack of suggests that over the next twenty years,
moisture in dry, dormant, or otherwise wildfire risk may increase more slowly in
combustible vegetation. Surface dryness is the Panhandle and Far West Texas than
strongly influenced by weather conditions, elsewhere as increased aridity reduces
but it is a consequence of desiccation of biomass. Meanwhile, the area of the
vegetation over an extended period of time state commonly affected by wildfires
rather than simply lack of humidity on a may expand eastward as fuels become
particular day. drier faster in a warmer climate.

26
 REFERENCES

Al Mukaimi, M. E., T. M. Dellapenna, and J. R. Williams, Camargo, S. J., and A. A. Wing, 2016: Tropical cyclones
2018: Enhanced land subsidence in Galveston Bay, Texas: in climate models. Wiley Interdisciplinary Reviews: Climate
Interaction between sediment accumulation rates and Change, 7, 211-237.
relative sea level rise. Estuarine, Coastal and Shelf Science,
207, 183-193. Capotondi, A., and M. A. Alexander, 2010: Relationship
between Precipitation in the Great Plains of the United
Anderson, C. J., C. K. Wikle, Q. Zhou, and J. A. Royle, 2007: States and Global SSTs: Insights from the IPCC AR4
Population Influences on Tornado Reports in the United Models. Journal of Climate, 23, 2941-2958.
States. Weather and Forecasting, 22, 571-579.
Capotondi, A., and Coauthors, 2015: Understanding ENSO
Ashley, S. T., and W. S. Ashley, 2008: Flood Fatalities in Diversity. Bulletin of the American Meteorological Society,
the United States. Journal of Applied Meteorology and 96, 921-938.
Climatology, 47, 805-818.
Changnon, S. A., and T. R. Karl, 2003: Temporal and Spatial
Berg, A., J. Sheffield, and P. C. D. Milly, 2017: Divergent Variations of Freezing Rain in the Contiguous United States:
surface and total soil moisture projections under global 1948–2000. Journal of Applied Meteorology, 42, 1302-1315.
warming. Geophysical Research Letters, 44, 236-244.
Cleaveland, M. K., T. H. Votteler, D. K. Stahle, R. C. Casteel,
Berg, A., and Coauthors, 2016: Land-atmosphere feedbacks and J. L. Banner, 2011: Extended Chronology of Drought in
amplify aridity increase over land under global warming. South Central, Southeastern and West Texas. Texas Water
Nature Climate Change, 6, 869-+. Journal, 2, 54-96.

Berg, M. D., 2018: Peak flow trends highlight emerging Cohen, J., K. Pfeiffer, and J. A. Francis, 2018: Warm Arctic
urban flooding hotspots in Texas. Texas Water Journal, 9, episodes linked with increased frequency of extreme winter
18-29. weather in the United States. Nature Communications, 9,
869.
Berghuijs, W. R., R. A. Woods, C. J. Hutton, and M.
Sivapalan, 2016: Dominant flood generating mechanisms Cohen, J., and Coauthors, 2014: Recent Arctic amplification
across the United States. Geophysical Research Letters, 43, and extreme mid-latitude weather. Nature Geoscience, 7,
4382-4390. 627.

Blackport, R., J. A. Screen, K. van der Wiel, and R. Bintanja, Collins, M., and Coauthors, 2013: Long-term Climate
2019: Minimal influence of reduced Arctic sea ice on Change: Projections, Commitments and Irreversibility.
coincident cold winters in mid-latitudes. Nature Climate Climate Change 2013: The Physical Science Basis.
Change, 9, 697-704. Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change,
Bomar, G. W., 2017: Weather in Texas: The Essential T. F. Stocker, and Coauthors, Eds., Cambridge University
Handbook. Third Edition ed. University of Texas Press, 290 Press.
pp.
Cook, B. I., T. R. Ault, and J. E. Smerdon, 2015:
Bright, R. M., E. Davin, T. O’Halloran, J. Pongratz, K. G. Unprecedented 21st century drought risk in the American
Zhao, and A. Cescatti, 2017: Local temperature response to Southwest and Central Plains. Science Advances.
land cover and management change driven by non-radiative
processes. Nature Climate Change, 7, 296-+. Dai, A., T. Zhao, and J. Chen, 2018: Climate Change and
Drought: a Precipitation and Evaporation Perspective.
Brimelow, J. C., W. R. Burrows, and J. M. Hanesiak, 2017: Current Climate Change Reports, 4, 301-312.
The changing hail threat over North America in response to
anthropogenic climate change. Nature Climate Change, 7, Dangendorf, S., C. Hay, F. M. Calafat, M. Marcos, C.
516. G. Piecuch, K. Berk, and J. Jensen, 2019: Persistent
acceleration in global sea-level rise since the 1960s. Nature
Brooks, H. E., 2013: Severe thunderstorms and climate Climate Change, 9, 705-710.
change. Atmospheric Research, 123, 129-138.
Davis, R. A. J., 2011: Sea-Level Change in the Gulf of
Bukovsky, M. S., R. R. McCrary, A. Seth, and L. O. Mearns, Mexico. Texas A&M University Press, 171 pp.
2017: A Mechanistically Credible, Poleward Shift in Warm-
Season Precipitation Projected for the U.S. Southern Great Diffenbaugh, N. S., M. Scherer, and R. J. Trapp, 2013:
Plains? Journal of Climate, 30, 8275-8298. Robust increases in severe thunderstorm environments in

27
 response to greenhouse forcing. Proceedings of the National Kunkel, K. E., and Coauthors, 2013: Monitoring and
Academy of Sciences of the United States of America, 110, Understanding Trends in Extreme Storms: State of
16361-16366. Knowledge. Bulletin of the American Meteorological Society,
94, 499-514.
Easterling, D. R., and Coauthors, 2017: Precipitation change
in the United States. Climate Science Special Report: Fourth Kurtzman, D., and B. R. Scanlon, 2007: El Nino-Southern
National Climate Assessment, Volume I, D. J. Wuebbles, D. Oscillation and Pacific Decadal Oscillation impacts on
W. Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart, and T. precipitation in the southern and central United States:
K. Maycock, Eds., U.S. Global Change Research Program, Evaluation of spatial distribution and predictions. Water
207-230. Resources Research, 43.

Elsner, J. B., L. E. Michaels, K. N. Scheitlin, and I. J. Elsner, Kushnir, Y., R. Seager, M. Ting, N. Naik, and J. Nakamura,
2013: The Decreasing Population Bias in Tornado Reports 2010: Mechanisms of Tropical Atlantic SST Influence on
across the Central Plains. Weather, Climate, and Society, 5, North American Precipitation Variability*. Journal of Climate,
221-232. 23, 5610-5628.

Frederikse, T., S. Jevrejeva, R. E. M. Riva, and S. Lanza, M., 2017: Space City Rewind: Houston’s Great Snow
Dangendorf, 2018: A Consistent Sea-Level Reconstruction of 1895. Space City Weather.
and Its Budget on Basin and Global Scales over 1958-2014.
Journal of Climate, 31, 1267-1280. Letetrel, C., M. Karpytchev, M. N. Bouin, M. Marcos, A.
SantamarÍa-Gómez, and G. Wöppelmann, 2015: Estimation
Guyette, R. P., F. R. Thompson, J. Whittier, M. C. of vertical land movement rates along the coasts of the
Stambaugh, and D. C. Dey, 2014: Future Fire Probability Gulf of Mexico over the past decades. Continental Shelf
Modeling with Climate Change Data and Physical Research, 111, 42-51.
Chemistry. Forest Science, 60, 862-870.
Liu, Y., S. L. Goodrick, and J. A. Stanturf, 2013: Future
Huang, Y., S. Wu, and J. O. Kaplan, 2015: Sensitivity of U.S. wildfire potential trends projected using a dynamically
global wildfire occurrences to various factors in the context downscaled climate change scenario. Forest Ecology and
of global change. Atmospheric Environment, 121, 86-92. Management, 294, 120-135.

Janssen, E., D. J. Wuebbles, K. E. Kunkel, S. C. Olsen, and Mahmood, R., and Coauthors, 2014: Land cover changes
A. Goodman, 2014: Observational- and model-based trends and their biogeophysical effects on climate. International
and projections of extreme precipitation over the contiguous Journal of Climatology, 34, 929-953.
United States. Earths Future, 2, 99-113.
Maloney, E. D., and Coauthors, 2014: North American
Kirtman, B., and Coauthors, 2013: Near-term Climate Climate in CMIP5 Experiments: Part III: Assessment of
Change: Projections and Predictability. Climate Change Twenty-First-Century Projections. Journal of Climate, 27,
2013: The Physical Science Basis. Contribution of 2230-2270.
Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, T. F. Stocker, McCabe, G. J., M. A. Palecki, and J. L. Betancourt, 2004:
and Coauthors, Eds., Cambridge University Press. Pacific and Atlantic Ocean influences on multidecadal
drought frequency in the United States. Proceedings of
Kloesel, K., and Coauthors, 2018: Southern Great Plains. the National Academy of Sciences of the United States of
Impacts, Risks, and Adaptation in the United States: Fourth America, 101, 4136-4141.
National Climate Assessment, Volume II, D. R. Reidmiller, C.
W. Avery, D. R. Easterling, K. E. Kunkel, K. L. M. Lewis, T. Meehl, G. A., J. M. Arblaster, and C. T. Y. Chung, 2015:
K. Maycock, and B. C. Stewart, Eds., U.S. Global Change Disappearance of the southeast US “warming hole” with the
Research Program, 987-1035. late 1990s transition of the Interdecadal Pacific Oscillation.
Geophysical Research Letters, 42, 5564-5570.
Kluver, D., and D. Leathers, 2015: Winter snowfall prediction
in the United States using multiple discriminant analysis. Milly, P. C. D., and K. A. Dunne, 2016: Potential
International Journal of Climatology, 35, 2003-2018. evapotranspiration and continental drying. Nature Climate
Change, 6, 946-+.
Kossin, J. P., T. Hall, T. R. Knutson, K. E. Kunkel, R. J. Trapp,
D. E. Waliser, and M. E. Wehner, 2017: Extreme Storms. Mishra, A. K., and V. P. Singh, 2010: Changes in extreme
Climate Science Special Report: Fourth National Climate precipitation in Texas. J Geophys Res-Atmos, 115.
Assessment, Volume I, D. J. Wuebbles, D. W. Fahey, K. A.
Hibbard, D. J. Dokken, B. C. Stewart, and T. K. Maycock, Mishra, A. K., V. P. Singh, and M. Özger, 2011: Seasonal
Eds., U.S. Global Change Research Program, 257-276. streamflow extremes in Texas river basins: Uncertainty,
trends, and teleconnections. Journal of Geophysical
Research, 116.

28
Morice, C. P., J. J. Kennedy, N. A. Rayner, and P. D. Jones, Pryor, S. C., and J. Ledolter, 2010: Addendum to “Wind
2012: Quantifying uncertainties in global and regional speed trends over the contiguous United States”. Journal of
temperature change using an ensemble of observational Geophysical Research: Atmospheres, 115.
estimates: The HadCRUT4 data set. Journal of Geophysical
Research: Atmospheres, 117. Scheff, J., 2018: Drought Indices, Drought Impacts, CO2,
and Warming: a Historical and Geologic Perspective.
Mukherjee, S., A. Mishra, and K. E. Trenberth, 2018: Climate Current Climate Change Reports, 4, 202-209.
Change and Drought: a Perspective on Drought Indices.
Current Climate Change Reports, 4, 145-163. Schmandt, J., G. R. North, and J. Clarkson, 2011: The
Impact of Global Warming on Texas. Second ed. University
Needham, H. F., B. D. Keim, D. Sathiaraj, and M. Shafer, of Texas Press, 318 pp.
2012: Storm Surge Return Periods for the United States Gulf
Coast. Advances in Hurricane Engineering, 715-740. Schubert, S., and Coauthors, 2009: A U.S. CLIVAR Project
to Assess and Compare the Responses of Global Climate
Nerem, R. S., B. D. Beckley, J. T. Fasullo, B. D. Hamlington, Models to Drought-Related SST Forcing Patterns: Overview
D. Masters, and G. T. Mitchum, 2018: Climate-change- and Results. Journal of Climate, 22, 5251-5272.
driven accelerated sea-level rise detected in the altimeter
era. Proceedings of the National Academy of Sciences of Seager, R., A. Hooks, A. P. Williams, B. Cook, J. Nakamura,
the United States of America, 115, 2022-2025. and N. Henderson, 2015: Climatology, Variability, and Trends
in the U.S. Vapor Pressure Deficit, an Important Fire-Related
Nigam, S., B. Guan, and A. Ruiz-Barradas, 2011: Key role Meteorological Quantity. Journal of Applied Meteorology
of the Atlantic Multidecadal Oscillation in 20th century and Climatology, 54, 1121-1141.
drought and wet periods over the Great Plains. Geophysical
Research Letters, 38, n/a-n/a. Smith, B. T., T. E. Castellanos, A. C. Winters, C. M. Mead,
A. R. Dean, and R. L. Thompson, 2013: Measured Severe
Nikiel, C. A., and E. A. B. Eltahir, 2019: Summer Climate Convective Wind Climatology and Associated Convective
Change in the Midwest and Great Plains due to Agricultural Modes of Thunderstorms in the Contiguous United States,
Development during the Twentieth Century. Journal of 2003–09. Weather and Forecasting, 28, 229-236.
Climate, 32, 5583-5599.
Stambaugh, M. C., R. P. Guyette, E. D. Stroh, M. A.
Paciorek, C. J., D. A. Stone, and M. F. Wehner, 2018: Struckhoff, and J. B. Whittier, 2018: Future southcentral US
Quantifying statistical uncertainty in the attribution of human wildfire probability due to climate change. Climatic Change,
influence on severe weather. Weather and Climate Extremes, 147, 617-631.
20, 69-80.
Swann, A. L. S., 2018: Plants and Drought in a Changing
Paine, J. G., T. L. Caudle, and J. R. Andrews, 2014: Climate. Current Climate Change Reports, 4, 192-201.
Shoreline Movement Along the Texas Gulf Coast, 1930’s to
2012, 62 pp. Swann, A. L. S., F. M. Hoffman, C. D. Koven, and J. T.
Randerson, 2016: Plant responses to increasing CO2
Paine, J. G., T. L. Caudle, and J. R. Andrews, 2017: reduce estimates of climate impacts on drought severity.
Shoreline and Sand Storage Dynamics from Annual Airborne Proceedings of the National Academy of Sciences of the
LIDAR Surveys, Texas Gulf Coast. Journal of Coastal United States of America, 113, 10019-10024.
Research, 33, 487-506.
Thomson, A. M., and Coauthors, 2011: RCP4.5: a pathway
Pendergrass, A. G., R. Knutti, F. Lehner, C. Deser, and B. for stabilization of radiative forcing by 2100. Climatic
M. Sanderson, 2017: Precipitation variability increases in a Change, 109, 77-94.
warmer climate. Sci Rep, 7, 17966.
Tippett, M. K., C. Lepore, and J. E. Cohen, 2016: More
Peng, S.-S., and Coauthors, 2014: Afforestation in China tornadoes in the most extreme U.S. tornado outbreaks.
cools local land surface temperature. Proceedings of the Science, 354, 1419-1423.
National Academy of Sciences, 111, 2915-2919.
Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B.
Perica, S., S. Pavlovic, M. St. Laurent, C. Trypaluk, D. Parsons, 2003: The Changing Character of Precipitation.
Unruh, and O. Wilhite, 2018: NOAA Atlas 14: Precipitation- Bulletin of the American Meteorological Society, 84, 1205-
Frequency Atlas of the United States, Volume 11 Version 1218.
2.0: Texas, 40 pp.
TSHA, 2018: Texas Almanac 2018-2019. Texas State
Pfahl, S., P. A. O’Gorman, and E. M. Fischer, 2017: Historical Association, 752 pp.
Understanding the regional pattern of projected future
changes in extreme precipitation. Nature Climate Change,
7, 423.

29
 USGCRP, 2017: Climate Science Special Report: Fourth Climate Assessment, Volume I, D. J. Wuebbles, D. W.
National Climate Assessment, Volume I. U.S. Global Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart, and T.
Change Research Program, 470 pp. K. Maycock, Eds., U.S. Global Change Research Program,
185-206.
Van Klooster, S. L., and P. J. Roebber, 2009: Surface-Based
Convective Potential in the Contiguous United States in a Vose, R. S., and Coauthors, 2014: Improved historical
Business-as-Usual Future Climate. Journal of Climate, 22, temperature and precipitation time series for U.S. climate
3317-3330. divisions. Journal of Applied Meteorology and Climatology,
53, 1232-1251.
van Oldenborgh, G. J., E. Mitchell-Larson, G. A. Vecchi,
H. de Vries, R. Vautard, and F. Otto, 2019: Cold waves are Wang, H., S. Schubert, M. Suarez, and R. Koster, 2010: The
getting milder in the northern midlatitudes. Environmental Physical Mechanisms by Which the Leading Patterns of SST
Research Letters, 14, 114004. Variability Impact U.S. Precipitation. Journal of Climate, 23,
1815-1836.
van Vuuren, D. P., and Coauthors, 2011: The representative
concentration pathways: an overview. Climatic Change, 109, Wood, A. W., L. R. Leung, V. Sridhar, and D. P. Lettenmaier,
5-31. 2004: Hydrologic implications of dynamical and statistical
approaches to downscaling climate model outputs. Climatic
Villarini, G., and J. A. Smith, 2013: Flooding in Texas: Change, 62, 189-216.
Examination of temporal changes and impacts of tropical
cyclones. Journal of the American Water Resources Zhang, W., G. Villarini, G. A. Vecchi, and J. A. Smith, 2018:
Association, 49, 825-837. Urbanization exacerbated the rainfall and flooding caused
by hurricane Harvey in Houston. Nature, 563, 384-388.
Vogel, M. M., R. Orth, F. Cheruy, S. Hagemann, R. Lorenz, B.
J. J. M. van den Hurk, and S. I. Seneviratne, 2017: Regional Zhao, G., H. L. Gao, and L. Cuo, 2016: Effects of
amplification of projected changes in extreme temperatures Urbanization and Climate Change on Peak Flows
strongly controlled by soil moisture-temperature feedbacks. over the San Antonio River Basin, Texas. Journal of
Geophysical Research Letters, 44, 1511-1519. Hydrometeorology, 17, 2371-2389.

Vose, R. S., D. R. Easterling, K. E. Kunkel, A. N. LeGrande,

and M. F. Wehner, 2017: Temperature changes in the United
States. Climate Science Special Report: Fourth National

30
 ENDNOTES
Endnotes

i
More general discussions of climate change and its impact on Texas may be found in:
Schmandt, J., G. R. North, and J. Clarkson, 2011: The Impact of Global Warming on Texas. Second ed. University
of Texas Press, 318 pp.
USGCRP, 2017: Climate Science Special Report: Fourth National Climate Assessment, Volume I. U.S. Global
Change Research Program, 470 pp.
Kloesel, K., and Coauthors, 2018: Southern Great Plains. Impacts, Risks, and Adaptation in the United States:
Fourth National Climate Assessment, Volume II, D. R. Reidmiller, C. W. Avery, D. R. Easterling, K. E. Kunkel, K.
L. M. Lewis, T. K. Maycock, and B. C. Stewart, Eds., U.S. Global Change Research Program, 987-1035.
ii
Online data archive: ftp://ftp.ncdc.noaa.gov/pub/data/cirs/climdiv/
Vose, R. S., and Coauthors, 2014: Improved historical temperature and precipitation time series for U.S. climate
divisions. Journal of Applied Meteorology and Climatology, 53, 1232-1251.
iii
Global temperatures are from the HadCRUT4 data set. All linear trends in this report are calculated using
ordinary least squares.
Morice, C. P., J. J. Kennedy, N. A. Rayner, and P. D. Jones, 2012: Quantifying uncertainties in global and regional
temperature change using an ensemble of observational estimates: The HadCRUT4 data set. Journal of Geophysical
Research: Atmospheres, 117.
iv
Meehl, G. A., J. M. Arblaster, and C. T. Y. Chung, 2015: Disappearance of the southeast US "warming hole" with
the late 1990s transition of the Interdecadal Pacific Oscillation. Geophysical Research Letters, 42, 5564-5570, Vose,
R. S., D. R. Easterling, K. E. Kunkel, A. N. LeGrande, and M. F. Wehner, 2017: Temperature changes in the United
States. Climate Science Special Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles, D. W.
Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research
Program, 185-206.
v
The historical simulations and projections are based on CMIP5 multi-model ensemble output from the Downscaled
CMIP3 and CMIP5 Climate and Hydrology Projections archive at https://gdo-
dcp.ucllnl.org/downscaled_cmip_projections
The first available ensemble member from the combined Historic and RCP 4.5 was used from each of the following
models for temperature and from the combined Historic and RCP 8.5 for precipitation. A ^ means only RCP 4.5
simulations were available from a particular model, so it was not included in the precipitation ensemble.

Modeling Center (or Group) Institute ID Model Name

Commonwealth Scientific and Industrial Research
ACCESS1.0
Organization (CSIRO) and Bureau of Meteorology (BOM), CSIRO-BOM
Australia ACCESS1.3^

BCC-CSM1.1
Beijing Climate Center, China Meteorological Administration BCC BCC-
CSM1.1(m)
Canadian Centre for Climate Modelling and Analysis CCCMA CanESM2
National Center for Atmospheric Research NCAR CCSM4
CESM1(BGC)
Community Earth System Model Contributors NSF-DOE-NCAR
CESM1(CAM5)
Centro Euro-Mediterraneo per I Cambiamenti Climatici CMCC CMCC-CM
Centre National de Recherches Météorologiques / Centre
Européen de Recherche et Formation Avancée en Calcul CNRM-CERFACS CNRM-CM5
Scientifique
Commonwealth Scientific and Industrial Research
Organization in collaboration with Queensland Climate CSIRO-QCCCE CSIRO-Mk3.6.0
Change Centre of Excellence

43
31
 EC-EARTH consortium EC-EARTH EC-EARTH^
LASG, Institute of Atmospheric Physics, Chinese Academy of
LASG-CESS FGOALS-g2
Sciences and CESS,Tsinghua University
LASG, Institute of Atmospheric Physics, Chinese Academy of
LASG-IAP FGOALS-s2^
Sciences

The First Institute of Oceanography, SOA, China FIO FIO-ESM

GFDL-CM3
NOAA Geophysical Fluid Dynamics Laboratory NOAA GFDL GFDL-ESM2G
GFDL-ESM2M
GISS-E2-H-CC^
NASA Goddard Institute for Space Studies NASA GISS GISS-E2-R
GISS-E2-R-CC^
National Institute of Meteorological Research/Korea
NIMR/KMA HadGEM2-AO
Meteorological Administration
Met Office Hadley Centre (additional HadGEM2-ES MOHC (additional
HadGEM2-CC
realizations contributed by Instituto Nacional de Pesquisas realizations by
Espaciais) INPE) HadGEM2-ES

Institute for Numerical Mathematics INM INM-CM4

IPSL-CM5A-LR^
Institut Pierre-Simon Laplace IPSL IPSL-CM5A-MR
IPSL-CM5B-LR
Japan Agency for Marine-Earth Science and Technology, MIROC-ESM
Atmosphere and Ocean Research Institute (The University of MIROC MIROC-ESM-
Tokyo), and National Institute for Environmental Studies CHEM
Atmosphere and Ocean Research Institute (The University of
Tokyo), National Institute for Environmental Studies, and MIROC MIROC5
Japan Agency for Marine-Earth Science and Technology
Max-Planck-Institut für Meteorologie (Max Planck Institute for MPI-ESM-MR
MPI-M
Meteorology) MPI-ESM-LR
Meteorological Research Institute MRI MRI-CGCM3

NorESM1-M
Norwegian Climate Centre NCC
NorESM1-ME^

The 1/8-degree BCSD (bias-corrected and statistically-downscaled) model output was downloaded as a
geographically averaged ensemble mean over a latitude-longitude box centered on Texas: 27.5°-35.5°N, 94.5°-
104.0°W, inclusive.
Wood, A. W., L. R. Leung, V. Sridhar, and D. P. Lettenmaier, 2004: Hydrologic implications of dynamical and
statistical approaches to downscaling climate model outputs. Climatic Change, 62, 189-216.
vi
van Vuuren, D. P., and Coauthors, 2011: The representative concentration pathways: an overviewibid., 109, 5-31.
Thomson, A. M., and Coauthorsibid.: RCP4.5: a pathway for stabilization of radiative forcing by 2100, 77-94.
vii
Kirtman, B., and Coauthors, 2013: Near-term Climate Change: Projections and Predictability. Climate Change
2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, T. F. Stocker, and Coauthors, Eds., Cambridge University Press.
viii
1900-1999 average: 64.6 °F. 2000-2018 average: 66.0 °F. 2036 projection: 67.6 °F. Warmest year on record:
2012 (67.8 °F).

44

32
ix
To our knowledge, no studies have specifically investigated this issue for Texas, but the effect has been noted
elsewhere.
Mahmood, R., and Coauthors, 2014: Land cover changes and their biogeophysical effects on climate. International
Journal of Climatology, 34, 929-953.
Peng, S.-S., and Coauthors, 2014: Afforestation in China cools local land surface temperature. Proceedings of the
National Academy of Sciences, 111, 2915-2919.
Bright, R. M., E. Davin, T. O'Halloran, J. Pongratz, K. G. Zhao, and A. Cescatti, 2017: Local temperature response
to land cover and management change driven by non-radiative processes. Nature Climate Change, 7, 296-+.
x
The stations are chosen for their geographical representativeness, data completeness, and lack of substantial
changes in station location, and they are characterized as Urban (within metropolitan areas of at least 500,000
population), Semi-Urban (within cities of 10,000-100,000 or near cities of 100,000-500,000), and Rural (within
cities of less than 10,000 or well outside of or upwind of larger cities). These stations, and which of four regions of
the state in which they are found (Northwest, Northeast, South, and Coastal), are listed here:
Urban: El Paso Intl Airport (NW); San Antonio Intl Airport (S); Dallas Love Field (NE); Houston Hobby Airport
(C); McAllen (S)
Semi-Urban: Hereford (NW); Huntsville (NE); College Station Easterwood Field (S); Harlingen (C); Corpus Christi
Intl Airport (C); Brownfield (NW)
Rural: Dalhart FAA Airport (NW); Luling (S); Midland Intl Airport (NW); Lufkin Angelina Co Airport (NE);
Amarillo Intl Airport (NW); Victoria Regional Airport (C); Beeville 5 NE (R); Crosbyton (NW); Centerville (NE)
In order to retain a sufficient number of homogeneous stations, the index data record only covers the period 1950-
present.
xi
The county temperature stations are created as follows: using SC-ACIS web services, every daily
maximum/minimum temperature observation is retrieved for a given county. Each year, starting in 1890, a station
with no more than ten missing days is identified. If no such station exists, data for that year is marked as missing. If
at least one such station exists, data from the station with the longest period-of-record is chosen. This data becomes
the year’s data for that county. The process is repeated for each year through 2018. Compared to the index stations,
the county stations are much less homogeneous. In particular, different stations would have taken observations at
different times, which can skew average temperatures. Because of this, we only use the county stations to identify
trends in monthly temperature extremes. Such block maxima are relatively insensitive to time-of-observation
artifacts.
xii
Although El Paso is part of the NW region, its summertime temperatures are markedly higher than other stations
within the region. So that El Paso's large number of 100 °F days would not have an outside influence on regional
trends, triple-digit days at all other index stations in the NW region were averaged for the period 1961-1990, and a
temperature threshold at El Paso was selected that is reached about as often as 100 °F was reached at other NW
stations. That threshold turned out to be 103 °F, so urban triple-digit days in NW Texas are estimated by the
number of days reaching 103 °F at El Paso.
xiii
Vogel, M. M., R. Orth, F. Cheruy, S. Hagemann, R. Lorenz, B. J. J. M. van den Hurk, and S. I. Seneviratne,
2017: Regional amplification of projected changes in extreme temperatures strongly controlled by soil moisture-
temperature feedbacks. Geophysical Research Letters, 44, 1511-1519.
xiv
Each year’s value consists of the average of four temperatures: the hottest temperature in June, the hottest
temperature in July, the hottest temperature in August, and the hottest temperature in September. For example, if at
the station being used for Travis County the highest temperature during June 2008 was 95, in July 2008 98, in
August 2008 100, and in September 2008 97, the average hottest day in 2008 would be 97.5. The value for index
stations is calculated in a similar manner, except that only June-August are used.
xv
USGCRP, 2017: Climate Science Special Report: Fourth National Climate Assessment, Volume I. U.S. Global
Change Research Program, 470 pp.
xvi
Nikiel, C. A., and E. A. B. Eltahir, 2019: Summer Climate Change in the Midwest and Great Plains due to
Agricultural Development during the Twentieth Century. Journal of Climate, 32, 5583-5599.
xvii
Calculated from county stations as the average of the lowest temperatures recorded during each of January,
February, and December.
xviii
Cohen, J., and Coauthors, 2014: Recent Arctic amplification and extreme mid-latitude weather. Nature
Geoscience, 7, 627.
Cohen, J., K. Pfeiffer, and J. A. Francis, 2018: Warm Arctic episodes linked with increased frequency of extreme
winter weather in the United States. Nature Communications, 9, 869.

45

33
34
Blackport, R., J. A. Screen, K. van der Wiel, and R. Bintanja, 2019: Minimal influence of reduced Arctic sea ice on
coincident cold winters in mid-latitudes. Nature Climate Change, 9, 697-704.
xix
van Oldenborgh, G. J., E. Mitchell-Larson, G. A. Vecchi, H. de Vries, R. Vautard, and F. Otto, 2019: Cold waves
are getting milder in the northern midlatitudes. Environmental Research Letters, 14, 114004.
xx
Those summers are 1990, 1998, 2009, and 2011.
xxi
Those winters are 1931 (27.1 °F), 1941 (25.1 °F), 1992 (24.7 °F), 1991 (24.3 °F), and 1907 (24.2 °F).
xxii
The county precipitation stations were created following the same method as the county temperature stations.
xxiii
Precipitation from single ensemble members from the Historical+RCP4.5 CMIP5 runs were averaged over the
box 25°N-37.5°N, 95°W-105°W and downloaded from the KNMI Climate Explorer. The models were: ACCESS1-
0 ACCESS1-3 bcc-csm1-1 bcc-csm1-1-m BNU-ESM CanESM2 CCSM4 CESM1-BGC CESM1-CAM5 CMCC-
CM CMCC-CMS CNRM-CM5 CSIRO-Mk3-6-0 EC-EARTH FGOALS-g2 FIO-ESM GFDL-CM3 GFDL-ESM2G
GFDL-ESM2M GISS-E2-H GISS-E2-H GISS-E2-H GISS-E2-H-CC GISS-E2-R GISS-E2-R GISS-E2-R GISS-E2-
R-CC HadGEM2-AO HadGEM2-CC HadGEM2-ES inmcm4 IPSL-CM5A-LR IPSL-CM5A-MR IPSL-CM5B-LR
MIROC5 MIROC-ESM MIROC-ESM-CHEM MPI-ESM-LR MPI-ESM-MR MRI-CGCM3 NorESM1-M
NorESM1-ME.
xxiv
Maloney, E. D., and Coauthors, 2014: North American Climate in CMIP5 Experiments: Part III: Assessment of
Twenty-First-Century Projections. Journal of Climate, 27, 2230-2270.
xxv
McCabe, G. J., M. A. Palecki, and J. L. Betancourt, 2004: Pacific and Atlantic Ocean influences on multidecadal
drought frequency in the United States. Proceedings of the National Academy of Sciences of the United States of
America, 101, 4136-4141.
Kurtzman, D., and B. R. Scanlon, 2007: El Nino-Southern Oscillation and Pacific Decadal Oscillation impacts on
precipitation in the southern and central United States: Evaluation of spatial distribution and predictions. Water
Resources Research, 43.
Schubert, S., and Coauthors, 2009: A U.S. CLIVAR Project to Assess and Compare the Responses of Global
Climate Models to Drought-Related SST Forcing Patterns: Overview and Results. Journal of Climate, 22, 5251-
5272.
Kushnir, Y., R. Seager, M. Ting, N. Naik, and J. Nakamura, 2010: Mechanisms of Tropical Atlantic SST Influence
on North American Precipitation Variability*ibid., 23, 5610-5628.
Wang, H., S. Schubert, M. Suarez, and R. Kosteribid.: The Physical Mechanisms by Which the Leading Patterns of
SST Variability Impact U.S. Precipitation, 1815-1836.
Capotondi, A., and M. A. Alexanderibid.: Relationship between Precipitation in the Great Plains of the United States
and Global SSTs: Insights from the IPCC AR4 Models, 2941-2958.
Nigam, S., B. Guan, and A. Ruiz-Barradas, 2011: Key role of the Atlantic Multidecadal Oscillation in 20th century
drought and wet periods over the Great Plains. Geophysical Research Letters, 38, n/a-n/a.
xxvi
Texas precipitation data is from the nClimDiv dataset of the National Centers for Environmental Information.
The AMO and PDO were obtained from the KNMI Climate Explorer using ERSSTv5 sea surface temperatures and,
in the case of the AMO, the AMO residual from a regression against global mean surface temperature.
xxvii
Bukovsky, M. S., R. R. McCrary, A. Seth, and L. O. Mearns, 2017: A Mechanistically Credible, Poleward Shift
in Warm-Season Precipitation Projected for the U.S. Southern Great Plains? Journal of Climate, 30, 8275-8298.
xxviii
Easterling, D. R., and Coauthors, 2017: Precipitation change in the United States. Climate Science Special
Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles, D. W. Fahey, K. A. Hibbard, D. J. Dokken,
B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research Program, 207-230.
Janssen, E., D. J. Wuebbles, K. E. Kunkel, S. C. Olsen, and A. Goodman, 2014: Observational- and model-based
trends and projections of extreme precipitation over the contiguous United States. Earths Future, 2, 99-113.
Kunkel, K. E., and Coauthors, 2013: Monitoring and Understanding Trends in Extreme Storms: State of Knowledge.
Bulletin of the American Meteorological Society, 94, 499-514.
Mishra, A. K., and V. P. Singh, 2010: Changes in extreme precipitation in Texas. J Geophys Res-Atmos, 115.
xxix
Easterling, D. R., and Coauthors, 2017: Precipitation change in the United States. Climate Science Special
Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles, D. W. Fahey, K. A. Hibbard, D. J. Dokken,
B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research Program, 207-230.
xxx
One year block maxima of 1-day precipitation totals were fit to a GEV distribution with linearly time-dependent
location and scale parameters using the climetRemes Python library. The trend in frequency of extreme rainfall was
calculated using the time-dependent PDF. Citation for software package:
Paciorek, C. J., D. A. Stone, and M. F. Wehner, 2018: Quantifying statistical uncertainty in the attribution of human
influence on severe weather. Weather and Climate Extremes, 20, 69-80.

46

35
 xxxi
Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B. Parsons, 2003: The Changing Character of Precipitation.
Bulletin of the American Meteorological Society, 84, 1205-1218.
xxxii
Pfahl, S., P. A. O’Gorman, and E. M. Fischer, 2017: Understanding the regional pattern of projected future
changes in extreme precipitation. Nature Climate Change, 7, 423.
xxxiii
Easterling, D. R., and Coauthors, 2017: Precipitation change in the United States. Climate Science Special
Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles, D. W. Fahey, K. A. Hibbard, D. J. Dokken,
B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research Program, 207-230.
xxxiv
Based on the 2035 global temperature projection by Kirtman et al. (0.4 °C increase by 2036 compared to 2010-
2018) and the ratio of Gulf of Mexico to global temperature increases by Collins et al. (75%-100%).
Kirtman, B., and Coauthors, 2013: Near-term Climate Change: Projections and Predictability. Climate Change
2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change, T. F. Stocker, and Coauthors, Eds., Cambridge University Press.
Capotondi, A., and Coauthors, 2015: Understanding ENSO Diversity. Bulletin of the American Meteorological
Society, 96, 921-938, Collins, M., and Coauthors, 2013: Long-term Climate Change: Projections, Commitments and
Irreversibility. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change, T. F. Stocker, and Coauthors, Eds.,
Cambridge University Press.
xxxv
Based on the county precipitation station analysis, the odds of extreme (once per decade or so) one-day
precipitation decrease by a factor of two for every 20% increase in precipitation amount.
xxxvi
Based on the updated NOAA Atlas 14 values for Texas, which are used for flood risk calculations.
Perica, S., S. Pavlovic, M. St. Laurent, C. Trypaluk, D. Unruh, and O. Wilhite, 2018: NOAA Atlas 14: Precipitation-
Frequency Atlas of the United States, Volume 11 Version 2.0: Texas, 40 pp.
xxxvii
Online data archive: ftp://ftp.ncdc.noaa.gov/pub/data/cirs/climdiv/
Vose, R. S., and Coauthors, 2014: Improved historical temperature and precipitation time series for U.S. climate
divisions. Journal of Applied Meteorology and Climatology, 53, 1232-1251.
xxxviii
Data and analysis archive label: climdiv-sp09st-v1.0.0-20190204 and SP09
Online data archive: ftp://ftp.ncdc.noaa.gov/pub/data/cirs/climdiv/
Overall yearly drought severity is computed by summing the negative statewide 9-month SPI values for a given year
and dividing by 12 to obtain an average annual drought intensity.
xxxix
The historic simulations and projections are based on the CMIP5 multi-model ensemble output, as with
Footnote iv for temperature, with the following differences: Model simulations using the RCP8.5 scenario are
extracted, to improve signal-to-noise ratio. Precipitation is summed for each water year, October-September, rather
than each calendar year, in order to better reflect the potential for dryness during the growing season and so as to not
divide cool-season precipitation variability associated with El Niño across two separate years. The water year
precipitation is converted to anomalies by subtracting the centered 29-year mean for each simulation, with the 1950-
1978 mean used for calculating anomalies in 1950-1964. Interannual precipitation variance is then calculated using
the same 29-year intervals, and variance time series are created from the overlapping 29-year periods.
xl
Pendergrass, A. G., R. Knutti, F. Lehner, C. Deser, and B. M. Sanderson, 2017: Precipitation variability increases
in a warmer climate. Sci Rep, 7, 17966.
xli
Mukherjee, S., A. Mishra, and K. E. Trenberth, 2018: Climate Change and Drought: a Perspective on Drought
Indices. Current Climate Change Reports, 4, 145-163.
xlii
Data and analysis archive label: climdiv-pdsist.v1.0-.0-20190204
Online data archive: ftp://ftp.ncdc.noaa.gov/pub/data/cirs/climdiv/
Vose, R. S., and Coauthors, 2014: Improved historical temperature and precipitation time series for U.S. climate
divisions. Journal of Applied Meteorology and Climatology, 53, 1232-1251.
xliii
Milly, P. C. D., and K. A. Dunne, 2016: Potential evapotranspiration and continental drying. Nature Climate
Change, 6, 946-+.
xliv
Swann, A. L. S., F. M. Hoffman, C. D. Koven, and J. T. Randerson, 2016: Plant responses to increasing CO2
reduce estimates of climate impacts on drought severity. Proceedings of the National Academy of Sciences of the
United States of America, 113, 10019-10024.
Swann, A. L. S., 2018: Plants and Drought in a Changing Climate. Current Climate Change Reports, 4, 192-201.
Scheff, J.Ibid.: Drought Indices, Drought Impacts, CO2, and Warming: a Historical and Geologic Perspective, 202-
209.
xlv
Dai, A., T. Zhao, and J. Chenibid.: Climate Change and Drought: a Precipitation and Evaporation Perspective,
301-312.

47

36
Berg, A., and Coauthors, 2016: Land-atmosphere feedbacks amplify aridity increase over land under global
warming. Nature Climate Change, 6, 869-+.
xlvi
Swann, A. L. S., 2018: Plants and Drought in a Changing Climate. Current Climate Change Reports, 4, 192-201.
xlvii
Cook, B. I., T. R. Ault, and J. E. Smerdon, 2015: Unprecedented 21st century drought risk in the American
Southwest and Central Plains. Science Advances.
xlviii
Seager, R., A. Hooks, A. P. Williams, B. Cook, J. Nakamura, and N. Henderson, 2015: Climatology, Variability,
and Trends in the U.S. Vapor Pressure Deficit, an Important Fire-Related Meteorological Quantity. Journal of
Applied Meteorology and Climatology, 54, 1121-1141.
xlix
Basis for estimate: Given a statewide average 1 hPa increase in vapor pressure deficit from 1960-2013 and a
climatological mean warm season vapor pressure deficit of 14 hPa (both from Seager et al. 2015), the fractional
trend is about -1/14 in 50 years, or -1/7 in 100 years. Over the 26 or so years from 2000-2018 to 2036, a
continuation of the trend would yield a change of about -1/28, or -4%.
l
Berg, A., J. Sheffield, and P. C. D. Milly, 2017: Divergent surface and total soil moisture projections under global
warming. Geophysical Research Letters, 44, 236-244.
li
Pendergrass, A. G., R. Knutti, F. Lehner, C. Deser, and B. M. Sanderson, 2017: Precipitation variability increases
in a warmer climate. Sci Rep, 7, 17966.
lii
Bukovsky, M. S., R. R. McCrary, A. Seth, and L. O. Mearns, 2017: A Mechanistically Credible, Poleward Shift in
Warm-Season Precipitation Projected for the U.S. Southern Great Plains? Journal of Climate, 30, 8275-8298.
liii
Cleaveland, M. K., T. H. Votteler, D. K. Stahle, R. C. Casteel, and J. L. Banner, 2011: Extended Chronology of
Drought in South Central, Southeastern and West Texas. Texas Water Journal, 2, 54-96.
liv
Ashley, S. T., and W. S. Ashley, 2008: Flood Fatalities in the United States. Journal of Applied Meteorology and
Climatology, 47, 805-818.
lv
Berghuijs, W. R., R. A. Woods, C. J. Hutton, and M. Sivapalan, 2016: Dominant flood generating mechanisms
across the United States. Geophysical Research Letters, 43, 4382-4390.
lvi
Villarini, G., and J. A. Smith, 2013: Flooding in Texas: Examination of temporal changes and impacts of tropical
cyclones. Journal of the American Water Resources Association, 49, 825-837.
lvii
Mishra, A. K., V. P. Singh, and M. Özger, 2011: Seasonal streamflow extremes in Texas river basins:
Uncertainty, trends, and teleconnections. Journal of Geophysical Research, 116.
lviii
Berg, M. D., 2018: Peak flow trends highlight emerging urban flooding hotspots in Texas. Texas Water Journal,
9, 18-29.
lix
Zhang, W., G. Villarini, G. A. Vecchi, and J. A. Smith, 2018: Urbanization exacerbated the rainfall and flooding
caused by hurricane Harvey in Houston. Nature, 563, 384-388.
Among our concerns with this study are the use of generic building heights rather than actual lidar-measured heights
to represent Houston urbanization and a method of aggregating streamflow trends that gives undue influence to
outliers.
lx
Berg, M. D., 2018: Peak flow trends highlight emerging urban flooding hotspots in Texas. Texas Water Journal, 9,
18-29, Villarini, G., and J. A. Smith, 2013: Flooding in Texas: Examination of temporal changes and impacts of
tropical cyclones. Journal of the American Water Resources Association, 49, 825-837.
lxi
Data accessed from U.S. Census Bureau, https://www.census.gov/geo/reference/ua/urban-rural-2010.html, 18
February 2019.
lxii
Zhao, G., H. L. Gao, and L. Cuo, 2016: Effects of Urbanization and Climate Change on Peak Flows over the San
Antonio River Basin, Texas. Journal of Hydrometeorology, 17, 2371-2389.
lxiii
The county-scale historic snowfall analysis is generated from county-specific composite station data, as with the
county-scale temperature and precipitation analyses discussed earlier in the report, except that the number of
allowed missing data points per year is increased to 27 to allow for the lower reliability of reporting of null snowfall
amounts. Also, for snowfall a year is defined as beginning on July 1 and ending on June 30.
lxiv
Lanza, M., 2017: Space City Rewind: Houston’s Great Snow of 1895. Space City Weather. [Accessed March 16,
2019]
lxv
Bomar, G. W., 2017: Weather in Texas: The Essential Handbook. Third Edition ed. University of Texas Press,
290 pp, TSHA, 2018: Texas Almanac 2018-2019. Texas State Historical Association, 752 pp.
lxvi
Kluver, D., and D. Leathers, 2015: Winter snowfall prediction in the United States using multiple discriminant
analysis. International Journal of Climatology, 35, 2003-2018.
lxvii
Changnon, S. A., and T. R. Karl, 2003: Temporal and Spatial Variations of Freezing Rain in the Contiguous
United States: 1948–2000. Journal of Applied Meteorology, 42, 1302-1315.

48

37
 lxviii
Data source: Storm Prediction Center, National Oceanic and Atmospheric Administration,
https://www.spc.noaa.gov/gis/svrgis/
lxix
For example, see:
Anderson, C. J., C. K. Wikle, Q. Zhou, and J. A. Royle, 2007: Population Influences on Tornado Reports in the
United States. Weather and Forecasting, 22, 571-579.
Elsner, J. B., L. E. Michaels, K. N. Scheitlin, and I. J. Elsner, 2013: The Decreasing Population Bias in Tornado
Reports across the Central Plains. Weather, Climate, and Society, 5, 221-232.
lxx
Van Klooster, S. L., and P. J. Roebber, 2009: Surface-Based Convective Potential in the Contiguous United
States in a Business-as-Usual Future Climate. Journal of Climate, 22, 3317-3330.
lxxi
Brooks, H. E., 2013: Severe thunderstorms and climate change. Atmospheric Research, 123, 129-138,
Diffenbaugh, N. S., M. Scherer, and R. J. Trapp, 2013: Robust increases in severe thunderstorm environments in
response to greenhouse forcing. Proceedings of the National Academy of Sciences of the United States of America,
110, 16361-16366.
lxxii
Smith, B. T., T. E. Castellanos, A. C. Winters, C. M. Mead, A. R. Dean, and R. L. Thompson, 2013: Measured
Severe Convective Wind Climatology and Associated Convective Modes of Thunderstorms in the Contiguous
United States, 2003–09. Weather and Forecasting, 28, 229-236.
lxxiii
Tippett, M. K., C. Lepore, and J. E. Cohen, 2016: More tornadoes in the most extreme U.S. tornado outbreaks.
Science, 354, 1419-1423.
lxxiv
Brimelow, J. C., W. R. Burrows, and J. M. Hanesiak, 2017: The changing hail threat over North America in
response to anthropogenic climate change. Nature Climate Change, 7, 516.
lxxv
Tide gauge data is from the National Oceanic and Atmospheric Administration,
https://tidesandcurrents.noaa.gov/sltrends/
lxxvi
Letetrel, C., M. Karpytchev, M. N. Bouin, M. Marcos, A. SantamarÍa-Gómez, and G. Wöppelmann, 2015:
Estimation of vertical land movement rates along the coasts of the Gulf of Mexico over the past decades.
Continental Shelf Research, 111, 42-51.
lxxvii
Davis, R. A. J., 2011: Sea-Level Change in the Gulf of Mexico. Texas A&M University Press, 171 pp.
lxxviii
Paine, J. G., T. L. Caudle, and J. R. Andrews, 2017: Shoreline and Sand Storage Dynamics from Annual
Airborne LIDAR Surveys, Texas Gulf Coast. Journal of Coastal Research, 33, 487-506.
Paine, J. G., T. L. Caudle, and J. R. Andrews, 2014: Shoreline Movement Along the Texas Gulf Coast, 1930’s to
2012, 62 pp.
lxxix
Al Mukaimi, M. E., T. M. Dellapenna, and J. R. Williams, 2018: Enhanced land subsidence in Galveston Bay,
Texas: Interaction between sediment accumulation rates and relative sea level rise. Estuarine, Coastal and Shelf
Science, 207, 183-193.
lxxx
Frederikse, T., S. Jevrejeva, R. E. M. Riva, and S. Dangendorf, 2018: A Consistent Sea-Level Reconstruction
and Its Budget on Basin and Global Scales over 1958-2014. Journal of Climate, 31, 1267-1280.
Nerem, R. S., B. D. Beckley, J. T. Fasullo, B. D. Hamlington, D. Masters, and G. T. Mitchum, 2018: Climate-
change-driven accelerated sea-level rise detected in the altimeter era. Proceedings of the National Academy of
Sciences of the United States of America, 115, 2022-2025.
Dangendorf, S., C. Hay, F. M. Calafat, M. Marcos, C. G. Piecuch, K. Berk, and J. Jensen, 2019: Persistent
acceleration in global sea-level rise since the 1960s. Nature Climate Change, 9, 705-710.
lxxxi
Kossin, J. P., T. Hall, T. R. Knutson, K. E. Kunkel, R. J. Trapp, D. E. Waliser, and M. E. Wehner, 2017:
Extreme Storms. Climate Science Special Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles,
D. W. Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research
Program, 257-276, Needham, H. F., B. D. Keim, D. Sathiaraj, and M. Shafer, 2012: Storm Surge Return Periods for
the United States Gulf Coast. Advances in Hurricane Engineering, 715-740.
lxxxii
Kossin, J. P., T. Hall, T. R. Knutson, K. E. Kunkel, R. J. Trapp, D. E. Waliser, and M. E. Wehner, 2017:
Extreme Storms. Climate Science Special Report: Fourth National Climate Assessment, Volume I, D. J. Wuebbles,
D. W. Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart, and T. K. Maycock, Eds., U.S. Global Change Research
Program, 257-276.
lxxxiii
Camargo, S. J., and A. A. Wing, 2016: Tropical cyclones in climate models. Wiley Interdisciplinary Reviews:
Climate Change, 7, 211-237.
lxxxiv
Pryor, S. C., and J. Ledolter, 2010: Addendum to “Wind speed trends over the contiguous United States”.
Journal of Geophysical Research: Atmospheres, 115.
lxxxv
Liu, Y., S. L. Goodrick, and J. A. Stanturf, 2013: Future U.S. wildfire potential trends projected using a
dynamically downscaled climate change scenario. Forest Ecology and Management, 294, 120-135.

49

38
lxxxvi
Huang, Y., S. Wu, and J. O. Kaplan, 2015: Sensitivity of global wildfire occurrences to various factors in
the context of global change. Atmospheric Environment, 121, 86-92.
lxxxvii
Liu, Y., S. L. Goodrick, and J. A. Stanturf, 2013: Future U.S. wildfire potential trends projected using a
dynamically downscaled climate change scenario. Forest Ecology and Management, 294, 120-135.
lxxxviii
Guyette, R. P., F. R. Thompson, J. Whittier, M. C. Stambaugh, and D. C. Dey, 2014: Future Fire Probability
Modeling with Climate Change Data and Physical Chemistry. Forest Science, 60, 862-870, Stambaugh, M. C., R. P.
Guyette, E. D. Stroh, M. A. Struckhoff, and J. B. Whittier, 2018: Future southcentral US wildfire probability due to
climate change. Climatic Change, 147, 617-631.

50

39
 TEXAS A&M UNIVERSITY

College of Geosciences

40