Border Climate Summary

%RUGHU&OLPDWH6XPPDU\ 5HVXPHQGHO&OLPDGHOD)URQWHUD ,VVXHG)HEUXDU\ Tree Rings and the Monsoon in the Southwestern U.S. B" D#$%&' G(%))%$, S*+,,' ,) G&- To better grasp ,-(#.+" #$/ D&0&',.1&$2 #$/ T+& the full range of L#3,(#2,(" ,) T(&&-R%$- R&4&#(*+, spatial and tem- U$%0&(4%2" ,) A(%5,$# poral variability that is pos- 6e monsoon is a major component of sible under natu- southwestern North America’s climate ral (non-human) regime and arguably one of the most conditions, re- anticipated regional climate events of searchers at the the year, delivering varying doses of life- University of Ar- giving rains to the U.S.-Mexico border re- izona are study- (: /: (: /: (: (: /: gion each summer. On average, the mon- ing the long- Dry Dry Dry soon brings three-fourths of northwestern term climate Wet Wet Wet Wet Winter Winter Winter Winter Mexico’s annual rainfall and up to half of history of the Summer Summer Summer Pre-Monson? Pre-Monsoon? the annual rain in the U.S. Southwest. monsoon using Dry Summer annual growth Figure 1. This photomicrograph illustrates a sequence of four Douglas- !r tree rings (1871–1874) from southwestern New Mexico. Each annual Variability in monsoon rainfall in time rings from long- growth ring is composed of light-colored earlywood (EW) and dark-colored and location increases northward with lived, moisture- latewood (LW). Both 1871 and 1872 contain intra-annual density variations distance from the “core region” in west- stressed trees. known as “false rings,” which are probably related to reduced soil moisture ern and northwestern Mexico and is in the May–June pre-monsoon period. Note the variability of EW-width, a Tree Rings and function of cool-season precipitation, and the independent variability of LW notably dramatic along the border and width, which corresponds to warm-season precipitation. into the southwestern U.S. 6ese year- the Monsoon to-year changes in!uence ecosystems, Our understanding of the monsoon excellent records for reconstructing en- rangelands, agriculture, public health, climate system has improved great- vironmental history, including natu- water resources, and water demand, ly in recent years, largely through the ral climate variability at time-scales yet the mechanisms behind the chang- cooperative e7orts of research associ- of years to decades to centuries. es are not completely understood. ated with the North American Mon- soon Experiment (http://www.eol. Southwestern North America has a ucar.edu/projects/name/) and im- dense network of tree-ring collec- proved seasonal forecasts. For lon- tions that extend back more than 400 Table of Contents: ger time scale climate change projec- years. Historically, scientists have used 1 Tree-Rings and the Monsoon in tions, global climate models robust- these records to learn about the long- the Southwestern U.S. ly predict that the region’s cool season term history of drought and wet- 3 A look back at the 2009 summer will be drier in the future, but predic- ness in the winter season, but these monsoon tions of future North American Mon- tree-ring samples also can be used soon precipitation, in response to hu- to study moisture history associated Recent Conditions man-caused climate change, are varied. with the North American Monsoon. 6 Temperature 7 Precipitation Natural proxy records in the form 6e annual growth rings in many south- 8 North American Drought Monitor of tree rings can help put these pro- western conifer species, such as pine 9 S o n o r a n Reservoir Levels jections and future changes in long- and 8r trees, are composed of light col- Forecasts term perspective. As described in ored “earlywood” that forms in the 10 Stream!ow Forecast the July 2009 Border Climate Sum- spring and dark colored “latewood” that 11 Precipitation Forecast mary (available online: http://www. forms in the summer (Figure 1). In the 12 ENSO climas.arizona.edu/forecasts/bor- Southwest U.S. and northwest Mexico, der/archive.html), tree rings provide continued on page 4

7KHLQIRUPDWLRQLQWKLVSDFNHWLVDYDLODEOHRQWKHZHEKWWSFOLPDVDUL]RQDHGXIRUHFDVWVERUGHUVXPPDU\KWPO _([HFXWLYH6XPPDU\

'LVFODLPHU–6is packet contains o9cial and non-o9cial forecasts, as well as other information. Executive Summary While we make every e7ort to verify this information, please understand that we do not warrant the accuracy of any of these materials. 6e user assumes the entire risk related to the use of this data. In General – El Niño is expected to in!uence the winter storm track and bring an CLIMAS disclaims any and all warranties, whether increased chance of above-average precipitation to the U.S. Southwest and north- expressed or implied, including (without limitation) any implied warranties of merchantability or 8tness western Mexico in the February–April period. Servicio Meteorológico Nacional pre- for a particular purpose. In no event will CLIMAS dicts well above-average precipitation for the Baja California peninsula and north- or 6e University of Arizona be liable to you or to any third party for any direct, indirect, incidental, western Sonora in March and April. Northwestern Mexico and the southwestern consequential, special or exemplary damages or lost United States received above-average precipitation during the second half of January. pro8t resulting from any use or misuse of this data.

Temperature – In northwestern Mexico and the southwestern U.S., temperatures were mostly average to above average during June–November 2009.

Precipitation – 6anks to several tropical storms in September and October, Baja California Sur and Sonora received above-average fall precipitation. November was mostly a dry month for the entire region.

Precipitation Forecast – SMN predicts well above-average precipitation for the Baja California peninsula and northwestern Sonora in March and April, and well above- average precipitation for Baja California Norte and northwestern Sonora in May.

ENSO – 6e chance of El Niño persisting through at least April stands at more than 90 percent, according to the latest IRI ENSO forecast. Neutral conditions are expected to return rapidly in the May–July period, which is typical of El Niño events.

Sta!: Aaron Banas, UA Graduate Research Assistant Jason Criscio, UA graduate research assistant Stephanie Doster, Institute of the Environment Associate Editor Luis Farfan, CICESE Research Scientist Gregg Gar!n, Institute of the Environment Deputy Director for Outreach David Gochis, NCAR Research Scientist Rebecca Macaulay, Graphic Artist Funding for the Border Climate Summary/Resumen del Clima de la Frontera was provided Andrea Ray, NOAA Research Scientist by Inter American Institute for Global Change Research (IAI) and the NOAA Sector Applications Research Program.

%RUGHU&OLPDWH6XPPDU\ _$UWLFOHV A look back at the 2009 summer monsoon

B" D#0%/ J. G,*+%4, N#2%,$#' Observed Accumulated Precipitation (mm) C&$2&( ),( A21,4.+&(%* R&4&#(*+ 140 140 120 120 On May 21, 2009, the monsoon mes- 100 100 sage was clear. During an online brie8ng, 80 80 Below Normal or webinar, forecasters and researchers 60 60 outlined their forecasts and expectations 40 40 Above Normal for the 2009 North American Mon- 20 20 soon (NAM) season (http://ag.arizona. 0 0 edu/climate/ws/052109.htm): a wet 1 Jul 6 Jul 11 Jul 16 Jul 21 Jul 26 Jul 1 Aug 6 Aug 11 Aug 16 Aug 21 Aug 26 Aug 1 Sep 6 Sep 11 Sep 16 Sep 21 Sep start to the monsoon tempered by un- Daily Precipitation and Normal Precipitation (mm/day) 6 6 certainty beginning in August. Yet as daily normal the season progressed, conditions took 5 5 an unexpectedly severe turn as the rains 4 4 all but disappeared. So, what happened during the 2009 NAM season, and 3 3 how accurate were those predictions? 2 2 1 1 6e consensus statement from the webinar was for an early and robust on- 0 1 Jul 6 Jul 11 Jul 16 Jul 21 Jul 26 Jul 1 Aug 6 Aug 11 Aug 16 Aug 21 Aug 26 Aug 1 Sep 6 Sep 11 Sep 16 Sep 21 Sep 0 set of monsoon rains in the Southwest Figure 1. Summer monsoon precipitation in 2009 for northwest Mexico and the Southwest U.S. U.S. and northwest Mexico, with con- Top: green indicates cumulative precipitation greater than the 1979–1995 average, and brown ditions becoming very uncertain by indicates cumulative precipitation below the average. By the end of the monsoon season, the average cumulative de!cit for the region was around 40 mm. Bottom: green bars indicate the daily mid-season due to the development of precipitation rate (mm/day) for the region; the black curve shows the average daily precipitation an El Niño event (warmer-than-aver- rate. Note the substantial decline in precipitation rate between July 5 and August 10, and the near age ocean temperatures) in the tropi- cessation of precipitation between July 27 and August 10. cal eastern Paci8c Ocean. 6is consensus statement was based on indepen- (departures from average conditions) for As suspected at the time of the sea- dent forecasts made by the U.S. and the period ending September 24. Ar- sonal forecast webinar, El Niño ap- Mexican national weather services, as eas of northeast and eastern Mexico pears to have played a signi8cant role well as analyses by climate researchers. (Coahuila, Nuevo León, Tamaulipas, in suppressing monsoon rainfall, par- and Veracruz) and portions of south- ticularly over southern and eastern To a certain degree, this general fore- ern Mexico (Yucatán, Chiapas, Guer- Mexico. Figure 1c shows the aver- cast turned out to be accurate for a rero, and Michoacán) were drier than age sea surface temperature anoma- limited region of the NAM, includ- average nearly all summer and faced lies for June 2b through September 16. ing the border regions of Arizona, So- some of the worst drought conditions A strong band of warmer-than-aver- nora, New Mexico, and Chihuahua. in recent history. Similarly, parts of age sea surface temperatures (around Figure 1 shows the accumulated rain- western Arizona and the broader re- 2 degrees Celsius above average) per- fall averaged throughout southern Ari- gions of the lower Colorado River re- sisted across much of the tropical east- zona and northern Sonora from July 1 gion experienced rainfall totals ap- ern Paci8c Ocean during the summer through September 24, and evidence proaching only 50 percent of average. months and has continued into 2010. exists of the forecasted early and robust onset of monsoon rainfall. Although it is always di9cult to de- A broader region of generally warmer- termine exactly what combination than-average sea surface temperatures However, what was not predicted was of meteorological events lead to ob- also existed across the eastern Pacif- the severity and extent of dry con- served seasonal patterns of rainfall, a ic, Gulf of Mexico, and Intra-America ditions that enveloped much of the basic analysis of regional atmospher- Seas regions. Past research has shown NAM region through July into early ic circulation features (weather pat- that such conditions tend to weak- September. Figure 1a shows the map terns) o7ers some explanation. en the overall monsoon circulation, of the 90-day precipitation anomalies continued on page 5

%RUGHU&OLPDWH6XPPDU\ _$UWLFOHV tree-rings, continued the widths of these components of tree 6e scientists’ pri-

rings can be distinguished using a mi- mary goal is to croscope, and they are most often re- study the spatial lated to seasonal moisture conditions; and temporal his-

wide earlywood bands result from wet tory of the mon- winters and wide latewood bands form soon over the

in wet summers. Conversely, earlywood southwestern U.S. is narrow following dry winters and for the last 500 latewood is narrow in dry summers. 8QLWHG6WDWHV years and address 0p[LFR

many research

Earlywood- and latewood-widths vary questions: When independently, and these widths can be were the worst measured to produce distinct records of droughts during cool- and warm-season precipitation vari- the last 8ve cen- ability. 6is concept was demonstrated turies? How ex- by David Meko and Christopher Baisan treme were these of the UA’s Laboratory of Tre e - R i n g Re- summer-season search in a small-scale pilot study that fo- droughts, and how PL cused on southeastern Arizona (Meko and long did they last? NP Baisan 2001). An early tree-ring study of Were there pro- IW the monsoon in Mexico was published in longed periods of P 2002 by Matthew 6errell, David Stahle, wetter-than-av- and Malcolm Cleaveland from the Uni- erage monsoons? Figure 2. Circles on this map illustrate the monsoon-sensitive tree-ring versity of Arkansas Tre e - R i n g Laborato- How does the chronology network currently being developed for the southwestern United States by researchers from the University of Arizona’s Laboratory ry, and Jose Villanueva-Diaz of the Insti- monsoon moisture of Tree-Ring Research. Triangles illustrate the complementary Mexican tuto Nacional de Investigaciones Fores- history compare chronology network developed by our colleagues from the Instituto tales, Agrícolas, y Pecuarias (INIFAP) in to that of the win- Nacional de Investigaciones Forestales, Agrícolas, y Pecuarias (INIFAP) in Durango, Mexico (6errell et al. 2002). ter season? What Mexico and the University of Arkansas Tree-Ring Laboratory. 6eir research in Mexico is ongoing. happened to the monsoon during the provide information that is useful for severe winter-season droughts (such as resource management. In the summer Current Research the late 1500s megadrought)? How fre- of 2011, once the long-term monsoon In 2008, with funding from the Na- quently has major drought occurred reconstruction datasets are complete, tional Science Foundation, a team of during both the cool and warm seasons? the researchers plan to host a technical experts from the UA’s Laboratory of What is the relationship between the workshop for stakeholders in the greater Tree-Ring Research and the Depart- monsoon and large-scale circulation fea- border region. In the meantime, infor- ment of Atmospheric Sciences initi- tures such as the El Niño Southern Os- mation will be available at ated a new large-scale research proj- cillation and the Paci8c Decadal Oscil- http://monsoon.ltrr.arizona.edu. ect to study tree rings and the history lation? How well do regional climate of the monsoon in the southwestern models reproduce the full range of vari- U.S. 6is three-year project will pro- ability evident in the tree-ring data? For more information, contact the author at: duce the 8rst network of monsoon- 409 Harvill Building, University of sensitive tree-ring chronologies for A primary thrust of the project is to Arizona,Tucson 85721-0076 the southwestern U.S. (Figure 2). work with regional stakeholders to [email protected] References 6e researchers have collected samples Meko, D.M., and C.H. Baisan. 2001. Pilot study of latewood-width of confers as from the original tree-ring sites to update an indicator of variability of summer rainfall in the north American Monsoon the chronologies to the 2008 calendar region: International Journal of Climatology, 21: 697–708. year and are measuring the earlywood- and latewood-widths. In previously col- 6errell, M.D., D.W. Stahle, M.K. Cleaveland, and J.Villanueva-Diaz, 2002. Warm lected tree-ring records, only total ring season tree growth and precipitation over Mexico. Journal of Geophysical Re- width was measured by earlier researchers. search, 107, no. D14, 4205, 10.1029/2001JD000851.

%RUGHU&OLPDWH6XPPDU\ _$UWLFOHV summer monsoon, continued because one of the key factors in!u- Another factor that inhibited monsoon steer moisture into Southwest U.S. and encing monsoon strength is the con- rainfall during summer 2009 was an northwestern Mexico, but the eastward trast between ocean and land tem- eastward displacement of high pressure displacement of high pressure resulted peratures, with relatively cool ocean near the top of the atmosphere between in drier-than-average !ow from the west and warm land temperatures lead- June 24 and September 16. Upper at- and north into the monsoon region. ing to increased monsoon activity. mosphere high pressure usually helps 90-Day Accumulation Anomalies Ending August 24, 2009 Related to this upper atmosphere cir- 55°N culation pattern is the fact that fewer- than-average tropical storms hit land 50°N in both the eastern and western por- 45°N tions of the NAM region. In 2008, four 250 named Paci8c systems made direct land- 175 40°N 100 fall or came within 50 miles of Baja 50 and the west coast of mainland Mex- 35°N 25 ico, whereas only two named systems -25 30°N (Hurricane Andres on June 21–24 and -50 -100 Hurricane Jimena on August 29–Sep- 25°N -175 tember 4) made landfall from the Pa- -250 ci8c during the 2009 monsoon season. 20°N

15°N Forecasts from computer models (“dynamical model forecasts”) were one 10°N 140°W 130°W 120°W 110°W 100°W 90°W 80°W 70°W 60°W component of the consensus forecast described above. While the dynamical Figure 2a. Accumulated precipitation anomalies (observed precipitation minus the 1979–1995 average) for June 26–September 24, 2009. Note the widespread region of below-average pre- models predicted a later-than-observed cipitation across most of Mexico and the Southwest U.S. onset of the monsoon, they did a rea- sonable job of predicting the spatial pat- Average SST Anomalies 24 Jun, 2009–16 Sep 2009 terns of wet (northwest Mexico) and dry conditions (southern and eastern Mex- 40°N ico) during July. However, by mid Au- gust they had largely underestimated the increasing degree of dryness through- °C 20°N out much of the monsoon region. 6is 3 shortcoming in predicting the magni- 2 tude of wet and dry periods is somewhat 1 similar to the 2008 monsoon forecasts 0° 0.5 (see the October 2008 Border Climate -0.5 Summary at http://www.climas.arizo- -1 na.edu/forecasts/border/archive.html). 20°S -2 -3 In summary, the 2009 monsoon season was characterized by drier-than- 40°S average conditions across much of the region. While seasonal forecasts showed some skill during 2009, much room for improvement re- 60°S 140°W 120°W 100°W 80°W 60°W 40°W 20°W 0° mains in predicting the severity of wet- Figure 2b. Sea surface temperature (SST) anomalies (observed minus 1979–1995 average) ter- or drier-than-average conditions. for the 2009 monsoon season. The large region of above-average SSTs along the equator in the eastern Paci!c Ocean is characteristic of an El Niño event. El Niño usually inhibits summer monsoon precipitation in northern Mexico and the Southwest U.S., but it enhances winter precipitation in the region.

%RUGHU&OLPDWH6XPPDU\ _5HFHQW&RQGLWLRQV

Temperature

Looking back at June–November 2009, temperatures were were mostly average to above average during June–November well above average across the entire Mexico-U.S. border re- 2009. Temperatures were mostly above average during Sep- gion during the summer months (not shown); areas in south- tember-November (Figure 1d), with some cooler than aver- eastern Arizona across to west Texas were 2 degrees Celsius age temperatures in Durango during November (Figure 1c). warmer than the long-term average. For September–Novem- ber (Figure 1b), above-average temperatures occurred mostly from California to southwestern New Mexico, while some areas of south-central New Mexico to west Texas saw average to below-average temperatures. Above-average temperatures in the southwestern U.S. were associated with below-average summer monsoon precipitation and long, relatively dry spells Notes: from mid-July until late August. Some of this can be as- Maps of recent temperature conditions were produced by the National cribed to the El Niño episode that began during the summer; Oceanic and Atmospheric Administration’s Climate Prediction Center El Niño has a tendency to suppress the northward extent of (NOAA-CPC). Temperature anomalies refer to departures from the monsoon moisture. In northwestern Mexico, temperatures 1971–2000 arithmetic average of data for that period.

Figure 1a. Mean temperature at 2-m elevation Figure 1b. Mean temperature at 2-m elevation for November 2009. for September–November 2009.

5 10 15 20 25 30 35 5 10 15 20 25 30 35 Degrees Celsius Degrees Celsius Figure 1c. Mean temperature departure from 1971-2000 Figure 1d. Mean temperature departure from 1971-2000 average at 2-m elevation for November 2009. average at 2-m elevation for September–November 2009.

-4 -3 -2 -1 1 2 3 4 -4 -3 -2 -1 1 2 3 4 Degrees Celsius Degrees Celsius

On the Web: For more information: http://www.cpc.ncep.noaa.gov/products/Drought/ Atm_Circ/2m_Temp.shtml

%RUGHU&OLPDWH6XPPDU\ _5HFHQW&RQGLWLRQV

Precipitation

Precipitation was mostly below average during the sum- and Empalme were cut o7 from basic services and at least 30 mer and early fall months north of the border but ranged neighborhoods of both cities were !ooded when more than from below average (Baja California Sur) to above average 632.5 millimeters of rain fell during Hurricane Jimena be- (Chihuahua) south of the border. Near the border, areas of tween August 29–September 4. Florencio Díaz Armenta, a above-average summer precipitation included southeastern delegate from the Comisión Nacional del Agua (Conagua), New Mexico, northern Chihuahua, and west Texas. Further noted that precipitation near the port surpassed the record south, June was wetter than average in Chihuahua, Duran- set in 1948 of 340 millimeters (El Imperial, September 4). go, Sinaloa and Sonora, and statewide summer precipitation was near average for most of northern Mexico, with the following exceptions: Chihuahua recorded above-average pre- Notes: cipitation and Sinaloa recorded below average. 6anks to sev- Maps of recent precipitation conditions were produced using data from the National Oceanic and Atmospheric Administration’s Climate Prediction eral tropical storms in September and October, Baja Califor- Center (CPC). Precipitation anomalies refer to departures from the 1971– nia Sur and Sonora received above-average fall precipitation. 2000 arithmetic average of data for that period. Percentage of normal is In rain-related news, around 200,000 people in Guaymas masked out where normal precipitation is less than 0.1 mm per day.

Figure 2a. United States and Mexico precipitation Figure 2b. United States and Mexico precipitation percent percent of 1971-2000 average for June–August 2009. of 1971-2000 average for September–November 2009.

25 25 50 50 75 75 125 125 150 percent 150 percent 175 175

Figure 2c. United States and Mexico precipitation for Figure 2d. United States and Mexico precipitation for September–November 2009. November 2009.

100 25 200 50 300 100 mm 400 mm 150 500 200

On the Web: OnFor more the information:Web: Forhttp://www.cpc.ncep.noaa.gov/products/Drought/ more information: http://www.cpc.ncep.noaa.gov/cgi-bin/US_anom_realtime.shAtm_Circ/2m_Temp.shtml

%RUGHU&OLPDWH6XPPDU\ _5HFHQW&RQGLWLRQV

North American Drought Monitor

6e North American Drought Monitor map (Figure 3) Figure 3. North American Drought Monitor (released shows drought across the southwestern U.S. and north- November 30). western Mexico. Following a moderately wet 2008–09 winter but exceedingly dry 2009 summer and fall, Arizona and southern California exhibited severe drought conditions. 6e major impacts in Arizona included draw down of the San Carlos Reservoir on the Upper Gila River to al- most historically low levels and degradation of rangeland conditions. Between August and November, reservoir levels increased in most reservoirs in Sonora (not shown; based on data from Comisión Nacional del Agua), with the exception of the most southern Sonoran reservoirs.

In drought-related news, water savings from the moisture accumulation from previous rainfall reached nearly 80 million cubic meters, according to the Irrigation District of the Rio Yaqui (IDRY). 6e director of opera- tions, Humberto Borbón, estimated that the volume not used by the reservoir system would irrigate between 7,000 and 8,000 hectares (about 17,300 to 19,700 acres) during an entire crop cycle (El Imperial, November 2).

On the U.S. side of the border, University of California re- Drought Intensity searchers estimated that 2009 water shortages led to loss- D0 Abnormally Dry es of 21,000 jobs and $703 million in central California’s D1 Moderate Drought San Joaquin Valley (California Drought Update, September 30). In other California drought news, the California Build- D2 Severe Drought ing Standards Commission approved the California Dual D3 Extreme Drought Plumbing Code, which delineates statewide standards to in- D4 Exceptional stall both potable and recycled water plumbing systems in certain types of buildings. 6e code applies to commercial, retail, and o9ce buildings, and several other public build- Drought Impact Types ings. Recycled water will be used for building cooling and Delineates Dominant Impacts toilet !ushing (California Drought Update, November 30). A = Agricultural (crops, pastures, grasslands) H = Hydrological (water) AH = Agricultural and Hydrological Notes: The North American Drought Monitor maps are based on expert assess- ment of variables including (but not limited to) the Standardized Precipi- tation Index, soil moisture, stream"ow, precipitation, and measures of vegetation stress, as well as reports of drought impacts. It is a joint e#ort of several agencies, including NOAA’s National Climatic Data Center, NOAA’s Climate Prediction Center, the U.S. Department of Agriculture, the U.S. National Drought Mitigation Center, Agriculture and Agrifood Canada, the Meteorological Service of Canada, and the National Meteo- rological Service of México (SMN - Servicio Meteorológico Nacional).

On the Web: For more information: For more information: http://www.cpc.ncep.noaa.gov/cgi-bin/US_anom_realtime.sh http://www.ncdc.noaa.gov/oa/climate/monitoring/drought/nadm/

%RUGHU&OLPDWH6XPPDU\ _5HFHQW&RQGLWLRQV

Sonoran Reservoir Levels

Most of Sonora’s reservoirs are well below maximum Notes: storage capacity but several, including Lázaro Cárde- The map gives a representation of current storage levels for reservoirs in Sonora. Reservoir locations are numbered within the blue circles on nas, P. Elías Calles, Álvaro Obregón, Abraham González, the map, corresponding to the reservoirs listed in the table. The cup and A. Ruiz Cortines, are at near- or above-average stor- next to each reservoir shows the current storage level (blue !ll) as a per- age capacity (Figure 4). November storage levels in cent of total capacity. Note that while the size of each cup varies with Lázaro Cárdenas and Abraham González were above the size of the reservoir, these are representational and not to scale. 2008 levels, but other reservoirs were below 2008 lev- The table details more exactly the current capacity level (listed as a els. Total November 2009 reservoir storage for Sono- percent of maximum storage, and percent of average storage). Current ra was 170 cubic hectometers below 2008 levels. and maximum storage levels are given in cubic hectometers for each reservoir. One cubic hectometer is one billion liters.

This map is based on reservoir reports updated daily in El Imparcial (http://www.elimparcial.com), using data provided by Comisión Nacio- nal del Agua.

Figure 4. Sonoran reservoir levels for November 2009 as a percent of capacity. The table lists current, average, and maximum storage levels.

Legend 100% Reservoir Average size of cups is representational of reservoir 50% size, but not to scale Current Level 8 0% R io 1 Co

ncep ción

e p

Reservoir Capacity Current Max Average a

Name Level Storage* Storage* Storage* i 2 R 1. Cuauhtémoc 3% 2.3 66.1 42.5 3 7 2. El Molinito 2% 4.4 233.9 130.2 a r

3. A.L. Rodríguez 0% 0.0 284.5 219.5 n e

p S 4 a t 4. I.R. Alatorre 15% 4.7 31.1 17.8 i o a R Rio M 5. Álvaro Obregón 63% 2,635.2 4,200.0 2,989.2 5 6. A. Ruiz Cortines 52% 944.4 1,822.6 950.3 i u q 7. P. Elías Calles 72% 2,874.1 a 2,620.8 3,628.6 Y o Rio y 6 a 8. Lazaro Cárdenas 66% 735.0 1,116.5 703.4 M Rio * cubic hectometers CLIMAS www.climas.arizona.edu On the Web: For more information: For more information: http://www.cpc.ncep.noaa.gov/cgi-bin/US_anom_realtime.sh http://www.elimparcial.com

%RUGHU&OLPDWH6XPPDU\ _)RUHFDVWV

Stream!ow Forecast

6e six-month Ensemble Stream!ow Prediction (ESP) fore- Notes: cast, released January 18 by the University of Washing- The forecast information provided in Figure 5 is updated monthly ton and Princeton University, predicts mostly below-av- by the University of Washington and Princeton University using ensemble stream"ow prediction (ESP) techniques. The average of a erage stream!ow for the period January–June (ranging group (ensemble) of forecasts is generated by using recent meteo- from 30 percent below average to 10 percent above aver- rology to initialize the Variable In!ltration Capacity (VIC) hydrologic age) for gages on both sides of the Mexico-U.S. border re- model. Stream"ow volume estimates are based on 40 VIC model runs, gion. 6e average !ow is based on the period from 1960 using meteorological data from the period 1960–1999. These estimates, shown in Figure 5, are expressed in terms of the percent of the through 1999. Most of the stream!ow forecasts for the re- 1960–1999 average stream"ow at each gage. gion in Figure 5 predict six-month stream!ow volumes of 70 to 90 percent of average. Even lower !ows are predicted for the gages at Imuris (#15) in northern Sonora and Las Sardinas in northern Durango (#36). Low stream!ows also are forecast for tributaries of the Colorado River in the U.S. It is very likely that most of these predictions will change when the forecasts are updated, because these forecasts were initiated before a series of late January and early February storms brought large amounts of precipitation to the Lower Colorado River Basin and northern Mexico.

Figure 5. United States and Mexico stream#ow forecast for January–June.

1. San Joaquin 21. Paso Nacori 2. Boquilla 22. Angostura 3. Hoover Dam 23. Guadalupe 4. Davis Dam 24. Huapaca 5. Parker Dam 25. Casas Grandes 6. Alamo Dam 26. Near Mcphee, CO 7. Imperial Dam 27. Near Bay!eld, CO 8. Virgin, UT 28. Navajo Reservoir 9. Near Hurricane, UT 29. Abraham Gonzalez % of normal 10. Little!eld, AZ 30. Albuquerque 11. Lee’s Ferry 31. Ixpalino >170 12. Desert View, AZ 32. Near Del Norte, CO 150–170 13. Near Cameron, AZ 33. Near Lobatos, CO 130–150 14. Near Grand Canyon 34. Near Chamita 110–130 15. Imuris 35. Villalba 90–110 16. Near Blu", UT 36. Las Sardinas 70–90 17. Oviachic 37. Zacatecas 50–70 18. La Junta 30–50 19. Novillo <30 20. Cubil

On the Web: For more information: http://www.hydro.washington.edu/forecast/westwide/s"ow/ index.6mons.shtml#seas_vol

%RUGHU&OLPDWH6XPPDU\ _)RUHFDVWV

Precipitation Forecast

6e Servicio Meteorológico Nacional (SMN) forecasts, is- Figure 6a. Precipitation forecast for March 2010 (released February 1, 2010). sued in early February, are based on years with similar patterns of precipitation, atmospheric circulation, and ocean temperatures, which a7ect the climate of the region; for this forecast, the years are 1969, 1980, 1988, and 1992. SMN predicts well above-average precipitation for the Baja Cali- fornia peninsula and northwestern Sonora in March and

April, and well above-average precipitation for Baja Califor- nia Norte and northwestern Sonora in May (Figures 6a–c). Forecasts for April and May show increasing chances of below-average precipitation in Baja California Sur. SMN forecasts average to above-average precipitation for most of Chi- huahua and Durango, and more than 50 percent greater- than-average precipitation for these states in May, a month Figure 6b. Precipitation forecast for April 2010 (released with a relatively low percent of the annual total precipita- February 1, 2010). tion for these states. Forecasts of below-average precipitation for states along continental Mexico’s west coast are consis- tent with climate patterns for El Niño events. 6ese forecasts from the NOAA-Climate Prediction Center (not shown) agree well with forecasts for the U.S.-Mexico border states.

Notes:

This forecast was prepared by the Servicio Meteorológico Nacional (SMN). The forecast methodology was developed by Dr. Arthur Douglas (Creighton University, retired) in collaboration with SMN scientists. Figure 6c. Precipitation forecast for May 2010 (released The forecasts are based on the average of precipitation values from February 1, 2010). analogous years in the historical record. Selection of analogous years is based on statistical analysis of factors in oceanic and atmospheric circulation known to in"uence precipitation in Mexico. Unique com- binations of climate indices are used in the forecasts each month. A statistical method known as cluster analysis is used to identify evolving climate patterns observed in the historic record and place each year in historical context; the years with the evolving climate patterns most similar to the current year are selected. Average atmospheric "ow patterns and surface precipitation anomalies are constructed with the historic data and compared with the climatological average.

Examples of atmospheric and oceanic factors used in identifying ana- logue years, include: Paci!c and Atlantic Ocean temperatures, tropical upper atmosphere oscillations, the position and strength of persistent high and low atmospheric pressure centers, and other factors.

The maps show predicted percent of monthly average precipitation. The legend shows the ranges of predicted percent of average precipitation associated with each color. Blues and greens indicate above-average precipitation; yellows and reds indicate below-average precipitation. White indicates precipitation within 20% of the climatological average (based on data from 1941-2002).

On the Web: For more information: http://smn.cna.gob.mx/productos/map-lluv/p-clim02.gif

%RUGHU&OLPDWH6XPPDU\ _)RUHFDVWV

ENSO (El Niño – Southern Oscillation)

Moderate El Niño conditions in the equatorial Paci8c Ocean continued through December and January, with sea surface temperatures (SSTs) remaining 1.5 degrees Celsius above average across much of the central and eastern parts of the basin. 6e International Research Institute for Climate and So- ciety (IRI) notes that the pattern of above-average SSTs has become more organized; thus this initially weak El Niño event has evolved into a pattern more typical of past El Figure 7a. The standardized values of the Southern Niño events. 6e Southern Oscillation Index (SOI; Figure Oscillation Index from January 1980–December 2009. La 7a) remained negative in December, indicating that the at- Niña/El Niño occurs when values are greater than 0.5 (blue) mosphere was responding to the above-average SSTs. In re- or less than -0.5 (red) respectively. Values between these cent weeks, subsurface water temperatures along the equa- thresholds are relatively neutral (green). tor also have remained well above average, indicating that 2.5 this warm water will help increase SSTs in coming weeks. 2.0 La Niña 1.5 1.0 All of these signs point towards the continuation of at least 0.5 weak to moderate El Niño conditions for the next sever- 0 -0.5 al months. 6e chance of El Niño persisting through at -1.0 least April stands at more than 90 percent, according to the -1.5

January 21 IRI ENSO forecast (Figure 7b). Neutral con- Value SOI -2.0 -2.5 ditions are expected to return rapidly in the May–July pe- -3.0 riod, which is typical of El Niño events, because they nor- -3.5 mally dissipate in the spring. In the meantime, El Niño -4.0 CLIMAS El Niño -4.5 www.climas.arizona.edu is expected to impact winter weather in the U.S.-Mexico 0 0 9 0 9 0 1980 1982 1984 1986 1988 1992 1994 1996 1998 2002 2004 2006 2008 2010 border region over the next several months. A strong sub- 1 2 tropical jet stream, forming in response to the unusual- Year ly warm waters in the eastern Paci8c, is expected to in!u- Figure 7b. IRI probabilistic ENSO forecast for El Niño 3.4 ence the winter storm track and bring an increased chance monitoring region (released January 21). Colored lines of above-average precipitation to the U.S. Southwest and represent average historical probability of El Niño, La Niña, northwestern Mexico in the February–April period. and neutral. 100 Notes: El Niño Figure 7a shows the standardized three-month running average values of 90 Neutral the Southern Oscillation Index (SOI) from January 1980 through May 2009. La Niña 80 The SOI measures the atmospheric response to sea surface temperature (SST) changes across the tropical Paci!c Ocean. The SOI is strongly associ- 70 ated with climate e#ects in parts of Mexico and the United States. Values 60 greater than 0.5 represent La Niña conditions, which are frequently associated with dry winters and sometimes with wet summers in the southwest- 50 ern U.S. and northwestern Mexico. Values less than -0.5 represent El Niño 40 conditions, which are often associated with wet winters in those regions. (%) Probability 30 Figure 7b shows the IRI probabilistic El Niño-Southern Oscillation (ENSO) 20 forecast for overlapping three month seasons. The forecast expresses the 10 probabilities (chances) of the occurrence of three ocean conditions in the 0 ENSO-sensitive Niño 3.4 region, as follows: El Niño is de!ned as the warm- Jan– Feb– Mar– Apr– May– Jun– Jul– Aug– Sep– Oct– Mar Apr May Jun Jul Aug Sep Oct Nov Dec est 25 percent of Niño 3.4 SSTs during the three month period in question, 2010 2010 La Niña is de!ned as the coolest 25 percent of Niño 3.4 SSTs, and neutral Time Period conditions are de!ned as SSTs falling within the remaining 50 percent of observations. The IRI probabilistic ENSO forecast is a subjective assess- ment of monthly model forecasts of Niño 3.4 SSTs. The forecast takes into On the Web: account the indications of the individual forecast models (including expert For more information: knowledge of model skill), an average of the models, and other factors. http://iri.columbia.edu/climate/ENSO/currentinfo/update.html

%RUGHU&OLPDWH6XPPDU\