Hydrology and Water Resources in Arizona and the Southwest, Volume 38 (2008)

Item Type text; Proceedings

Publisher Arizona-Nevada Academy of Science

Journal Hydrology and Water Resources in Arizona and the Southwest

Rights Copyright ©, where appropriate, is held by the author.

Download date 26/09/2021 13:49:17

Link to Item http://hdl.handle.net/10150/296676 Volume 38

HYDROLOGY AND WATER RESOURCES IN ARIZONA AND THE SOUTHWEST

Proceedings of the 2008 Meetings of the

Hydrology Section Arizona -Nevada Academy of Science

March 29, 2008, Southwestern University, Phoenix, Arizona ACKNOWLEDGMENTS

The Spring 2008 meeting of the Hydrology Section of the Arizona- Nevada Academy of Science was held at Southwestern University, Phoenix, Arizona, on March 29, 2008. The organizers wish to thank Boris Poff, the Chairperson for the 2008 Hydrology Section meeting. Appreciation is also extended to Cody L. Stropki, School of Natural Resources, University of Arizona, for preparing these proceedings of the meeting.

ii CONTENTS

LONG -TERM CHANGES IN PEAK SNOWPACK ACCUMULATIONS ON ARIZONA WATERSHEDS 1 Peter F. Ffolliott

INTERNATIONAL CO- OPERATIVE PROGRAM ON ASSESSMENT AND MONITORING OF AIR POLLUTION EFFECTS ON FORESTS: THE SIERRA 5 ANCHA EXPERIMENTAL FOREST, ARIZONA Boris Poff and Daniel G. Neary

A CONTRAST AMONG NATIONAL FOREST WATERSHED PROGRAMS: 1978 - 2008 11 Robert E. Lefevre

SOUTH -TO -NORTH WATER DIVERSION PROJECT IN CHINA Hui Chen and Peter F. Ffolliott 17

TRANSPIRATION OF OAK TREES IN THE OAK SAVANNAS OF THE SOUTHWESTERN BORDERLANDS REGION 23 Peter F. Ffolliott, Cody L. Stropki, Aaron T Kauffman, and Gerald J. Gottfried

HOW USEFUL IS LiDAR IN ESTABLISHING A STREAM GAUGING NETWORK IN A TROPICAL EXPIRMENTAL FOREST 29 Boris Poff, Daniel G. Neary, and Gregory P. Asner

COMPARING BEDLOAD CONDITIONS IN THE CASCABEL WATERSHEDS, CORONADO NATIONAL FOREST. 33 Karen A. Koestner, Daniel G. Neary, Gerald J. Gottfried

CHARACTERISTICS AND BEHAVIOR OF A COOL -SEASON PRESCRIBED FIRE IN THEOAK SAVANNAS OF THE SOUTHWESTERN BORDERLANDS 41 Karen A. Koestner , Daniel G. Neary, Gerald J. Gottfried , and Ruben Morales\

HYDROLOGY AND EROSION IMPACTS OF MINING DERIVED COASTAL SAND DUNES, CHAÑARAL BAY, CHILE 47 Daniel G. Neary and Pablo Garcia -Chevesich

CLIMATE CHANGE IMPACTS ON MUNICIPAL, MINING, AND AGRICULTURAL WATER SUPPLIES IN CHILE 53 Daniel G. Neary and Pablo Garcia -Chevesich

The papers contained in these proceedings reflect the author(s) interpretations of the data sets, inferences, and interpretations presented therein. Editing of these papers by the compilers of the proceedings consisted largely of formatting for consistency of presentation. LONG -TERM CHANGES IN PEAK SNOWPACK ACCUMULATIONS ON ARIZONA WATERSHEDS

Peter F. Ffolliott'

Field observations and computer predictions indicate would (hopefully) retain a measurable snowpack that the magnitudes and timing of peak snowpack throughout the snowmelt- runoff season. accumulations in the western states might be The changing number of snow courses through changing, with magnitudes less and timing earlier in time is another issue.The first snow courses in the snowmelt- runoff season. These changes haveArizona were established in the middle 1930s. often been attributed to changes in the regional Additional snow courses were addedinthe climate. Because snowmelt -runoff is a major source following years, while other snow courses were of streamflows from Arizona watersheds (Ffolliott eliminated through time because of a lack of and Baker 2000, Baker and Ffolliott 2003), a study (statistical)significance in contributing to the of the long -term magnitudes and timing of peakregression models for predicting snowmelt- runoff snowpack accumulations has been initiated to volumes.The results presented in this paper, determine if thereported changes in snowpack therefore, must be interpreted within the framework conditions is also occurring in this state. Data sets of this variable sample. representing measurements from the network of Water equivalents on the snow courses were snow courses maintained by the Natural Resources originally measured with a snow tube on a bi- weekly ConservationService(formerlytheSoilbasis on the beginning and mid -point of a month Conservation Service) and their cooperators are the beginning in January and continuing to the end of basis for this study. Preliminary results of the study snowmelt -runoff season. Due to unfavorable are presented in this paper. weather conditions, the availability of personnel (especially cooperators), and other uncontrollable SOURCE DATA factors, however, this bi- monthly schedule could not Source data represent nearly 70 years ofalways be maintained. Snowpack measurements at snowpack measurements and, therefore, are thethe prescribed bi- weekly time interval were not longest record of snowpack conditions in the state. always possible as a consequence. It should be noted, however, that these data are Measurements of water equivalents with a snow biased because of the selected locations of the snow tube have been largely replaced in recent years by courses, varying number of snow courses in the transmitting data collected on selected snow courses network, snowpack measurement intervals, andby pressure pillows (snow pillows) to a central changes in measurement technique. processing center by SNOWTEL (the Snow Data The snow courses were established to provide Telemetry System).The data obtained by this an index of snowpack magnitudes (in inches ofmethod represent the real -time conditions on a snow snowmelt -runoff waterequivalent)topredict course and, importantly, are available to be accessed volumes, not necessarily to represent the average and retrieved for almost analysis at any time (Brooks snowpack conditions on the watershed in question. et al. 2003). Therefore, the snow courses were located in The magnitudes and timing of peak snowpack accessible areas to facilitate obtaining the necessary accumulations,therefore,are more accurately snowpack measurements and they were situated on estimated with these data sets, therefore, than was sites that were relatively flat, protected from thepossible with the earlier estimates obtained with a prevailing wind to minimize blowing of snow, and snow tube on a bi- weekly basis.

'School ofNatural Resources, University of Arizona, Tucson, Arizona 2

ANALYTICAL PROCEDURES of conserving and augmenting the states water The magnitudes and timing of peak snowpack supplies to accommodate the growing population of accumulations on snow courses with the longest people (Fox et al. 2000). continuous records were averaged for 10 year A steady decline in the magnitudes of peak intervals(thatis,1930,1940, 1950,etc.)for snowpack accumulation between the decades of analysis. However, there were less than 10 years of 1960 and 2000 is indicated in figure 1, reflecting record in the 1930 decade because the first snowabout a 50 percent reduction in water equivalents for courses were established in the middle 1930s, while the period. This reduction is similar to that reported the record for the 2000 decade remains incomplete. for other western states. Whether this decline will Another consideration of importance in analyzingcontinue into the future is open to conjecture. the data sets centered on the varying number of snow coursesinthesamplefora specified snowmelt- runoff season. The sampling frame was not constant throughout the study period because of the establishment of additional snow courses and the discontinued monitoring of some of the established courses through the years of data collection. Snow courses selected for analysis in this preliminary study were all located on watersheds within the major river basins in Arizona including the Gila, Salt, Verde, San Francisco, and Little Colorado. They ranged in elevations from about 6,500 to over 8,500 feet, an elevational strata representing the historically expected snowfall snowpack region in the state (Beschta 1974). While some ofFigure 1. Magnitudesofpeak these snow courses were situated on sites with aaccumulations on selected snow courses. Average forest overstory, most of the courses had beenwater equivalents for the 10 year intervals of 1930 to established in openings adjacent to a forest structure.2000 are shown. *Less than 10 years of record.

RESULTS AND DISCUSSION It should be noted that the severe drought of the Magnitudes of Peak Snowpack Accumulations 1950s is illustrated by the magnitude of peak The magnitudes ofpeak snowpack snowpack accumulation for this decade. This accumulations (in inches of water equivalents) for drought was followed by a decade of above average the study period are presented in figure 1.The precipitation, again, reflected by the high magnitude magnitudes prior to 1950 are likely to be indicative of peak snowpack accumulation for this decade. of the A historical variability in peak snowpackOnce again, the magnitudes of peak snowpack accumulations normally expected in the state. accumulation before 1950 are likely indicative of the While some of the snow courses in a samplenormal variability to be expected in the state. period in this time span attained seasonal peaks on a particular date, other snow course often attained this Timing of Peak Snowpack Accumulations condition either before or after that date, with the The timing of peak snowpack accumulations averagesnowpackconditionsreflectingthis areshown in figure 2.That the timing of peak variability.It is also important to note that thesnowpack accumulations is occurring earlier in the decade of the 1950s was a period of prolongedsnowmelt- runoff season appears inconclusive. A drought throughout the state. Parenthetically, it was clearly defined trend in the timing of peak snowpack duringthisdrought periodthattheArizona accumulations for the study period is not shown in Watershed Program was initiated to intensify effortsthis figure. 3

Snowpacks are largely intermittent in Arizona, the snowpacks in Arizona generally above average with alternating cycles of snow accumulation and when the spring snowmelt- runoff season began. melt. The snowpacks are also relatively shallow in The results obtained on the timing of peak comparison to other regions in the western states. snowpackaccumulationsareconsidered Furthermore, peak snowpack accumulations in the inconclusive by the author at this time. The state are likely to occur at different times in different variability of the snowpack conditions normally locations through the snowfall region largely in observedthroughoutthestatemightbea response totheinherent elevational gradientscontributing factor to this situation. Itis also (Beschta 1974).It could be, therefore, that the unknown if a longer data set would allow a clearer timing of peak snowpack accumulations illustratedpicture of the timing to be obtained. in figure 2 is simply a measure of the expected variability. REFERENCES Baker, M. B., Jr., and P. F. Ffolliott. 2003. Role of 8-Mar snow hydrology in watershed management. Journal of the Arizona -Nevada Academy of 29-Feb - Science, Special Issue: Watershed Management 21-Feb - in Arizona 35(1):42 -47. m 13-Feb - Beschta, R. L. 1974. Climatology of the ponderosa E 5-Feb - pine type in central Arizona. Technical Bulletin 228, Arizona Agricultural Experiment Station, 28,Jan - Tucson, Arizona. 20-Jan - Brooks, K. N., P. F. Ffolliott, H. M. Gregersen, and

12 -Jan L. F. DeBano. 2003. Hydrology and the 930* 1940195019601970198019902000* management of watersheds. Iowa State Press, Decade Ames, Iowa. Ffolliott, P. F., and M. B. Baker, Jr. 2000. Figure 2. Approximate timing of peak snowpack Snowpack hydrology in the southwestern United accumulations on selected snow courses. Average States: Contributions to watershed management. dates for the 10 year intervals of 1930 to 2000 In: Ffolliott, P. F., M. B. Baker, Jr., C. B. shown. *Less than 10 years of record. Edminster, M. C. Dillon, and K. L. Mora, technical coordinators. Land stewardship in the 21st century: The contributions of watershed CONCLUSIONS management. U.S.ForestService, Rocky The results of this preliminary study confirm MountainResearchStation,Proceedings (to some extent) the findings of some of the other RMRS -P -13, pp. 274 -276. studies in the western states of the long -term decline Fox, K. M., P. F. Ffolliott, M. B. Baker, Jr., and L. in the magnitudes of peak snowpack accumulations. F. DeBano. 2000. More water for Arizona: A How long the observed decline will continue into the history of the Arizona Watershed Program and future is unknown. The winter of 2007 -2008 - which the Arizona Water Resources Committee. Primer was not included in this study - produced the best Publishers, Phoenix, Arizona. overall snowfall of the preceding eight years, with INTERNATIONAL CO- OPERATIVE PROGRAM ON ASSESSMENT AND MONITORING OF AIR POLLUTION EFFECTS ON FORESTS: THE SIERRA ANCHA EXPERIMENTAL FOREST, ARIZONA

Boris Poff' and Daniel G. Neary2

At the end of the 2007 Fiscal Year, the Experimental mapping will contribute to the calculation of critical Forests and Ranges (EFR) Synthesis Network levels /loadsandtheirexceedancesinforest Committee awarded funds to 18 sites to establish a ecosystems. Ideally this will improve collaboration strategic ICP Level II (described below) synthesis with other environmental monitoring programs as network in the United States. Eleven Experimental well and contribute to the monitoring activities to Forest were selected to be included in the network, other aspects of relevance for forest policy at as well as seven Long Term Ecological Researchnational and global level, such as effects of climate (LTER) sites. This will give the USFS R &D a ICP changes on forests, sustainable forest management Level II Network starting this year, the only one in and biodiversity in forests. Lastly, the network will North America (see Appendix). Each site will provide policy- makers and the general public with include a NADP weather station, UV Radiationrelevant information. Monitors and Ozone Sensors. The Sierra Ancha The Sierra Ancha EF in particular plays three Experimental Forest (EF) was chosen to be part ofimportant roles inthis network: (1)itisthe this network, because it is the most southern EF in southernmost EF in the contiguous US; (2) it is the contiguous US and because it is downwind fromdownwindfromthenation'sfourthlargest Phoenix, Arizona, one of thenation'slargest metropolitan area; and (3) it is the driest EF in the metropolitan areas. network.

OBJECTIVES BACKGROUND The main objective of the proposed network is The InternationalCo- operative Program on to strengthen the role of the EFR by providing moreAssessment and Monitoring of Air Pollution Effects internationally standardized data which will benefit on Forests (ICP) was launched in 1985 under the the entire US EFR network -all 77 sites. TheConvention on Long -range Transboundary Air network will contribute to a better understanding ofPollutionoftheUnitedNationsEconomic the relationships between the condition of forestCommission for Europe (UNECE) due to the ecosystems and anthropogenic (in particular air growing public awareness ofpossible adverse effects pollution) as well as natural stress factors throughof air pollution on forests. ICP Forests monitors the intensive monitoring on a number of selected forest condition in Europe, in cooperation with the permanent observation plots and to study theEuropean Union using two different monitoring development of important forest ecosystems. intensity levels. The first grid (called Level I) is Further, the network will provide a deeperbased on around 6000 observation plots on a insight into the interactions between the varioussystematic transnationalgrid of 16 x 16 km components of forest ecosystems by compiling throughout Europe. The intensive monitoring level available information from related studies, and incomprises around 800 Level II plots in selected close co- operation with the ICP on modeling and forest ecosystems in Europe. Currently 40 European countries and the US participate in the ICP Forests network. 'Mojave National Preserve, National Park Service, Barstow, California 'Rocky Mountain Research Station U.S. Forest Service, Flagstaff, Arizona 6

METHODS ranged in size from several square meters to Study Site complete watersheds comprising several thousand The Sierra Ancha Experimental Forest, located hectares. The Sierra Ancha is still maintained as a on the Tonto National Forest about 48 km northeast research site under the administration of the Rocky of Globe, Arizona, was established in 1932 as aMountain Research Station. Many of the earlier researchareadevotedtostudying watershedwatershed studies have been concluded and the management. This 5,364 -ha experimental area isresults published. typical of watershed and vegetation conditions throughout the Southwest, particularly in Arizona. Set -up Considerations The climate, soil, and physiography are typical One important selection criterion is that the Level of much of the southwestern region, and are II plots in a country should be located in such way particularly representative of the Verde, Salt, andthat the most important forest species and most Upper Gila watersheds. The Sierra Ancha lies along widespread growing conditions in the respective the crest of the Sierra Ancha Mountain range andcountry are represented. Within theplot,the includes areas between 1,082 to 2,354 m insituation shall be as homogeneous as possible elevation. Vegetation types within the forest range regardingtreespecies,standtypeandsite from semidesert shrub and grassland to the pine -firconditions. Whenever possible, plots should be forests at higher elevations. selected that have been monitored during the last years. The great advantage of existing data on air Climate quality and meteorological parameters from nearby Precipitation averages about 850 mm at the stations should be taken into consideration whenever higher elevations at Workman's Creek, 635 mm atestablishing Level II plots. the intermediate elevations (1,460 to 1,830 m) The plot has a minimum size of 0.25 ha. Each plot surrounding the headquarters, and 410 mm at theis surrounded by a buffer zone with a minimum lower elevations. width of 10 m, if possible. There should be no differences in the management of the plot, its buffer Soils zone and surrounding forest,e.g, management Geology of the rangeiscomplex with operations should be comparable and fencing should sedimentary,metamorphic, and igneousrocks be limited to a minimum. However, the disturbance uplifted in a dome like structure. Thick formations of the monitoring activities should be minimized. of Dripping Springs quartzite, dissected by deepTrees felled in the plot or in the buffer zone should canyons or with intrusions of diabase and basalt be registered and if possible used for increment plugs and sills are common in much of the forest. analysis. The standard Level II -plot design is shown Troy sandstone occurs at higher elevations. in the appendix. In principle, all trees in the total plot are to be Vegetation included in the sample for the tree assessment, e.g, Eight vegetation types have been identified oncrown inventory, increment assessment. In the case the Sierra Ancha including, from the high elevations that the plot has many trees in a dense stand, a to low; mixed conifer, mountain park, ponderosa sub -plot maybe defined to be used for these surveys. pine, chaparral, oak woodland, desert grassland, The size of the sub -plot at the time of the installation desert shrub, and riparian. Fifty -seven percent of the of the plots should be large enough to give reliable vegetation is covered by chaparral shrubs. estimates for theses surveys for a minimum of 20 years, preferably throughout the life of the stand. Research, Past and Present A minimum of at least 20 trees in the sub -plot Research studies on watershed managementshould be available in this period. The installation of problems in woodlands, chaparral, ponderosa pine,a plot comprises its detailed description, including and pine -fir forests were conducted on the sites that stand and site characteristics and other available 7 information on the history of the plot, or otherinstalled and set -up two station personnel will travel nearby monitoring stations. probably on a monthly basis to the Sierra Ancha EF Efforts to complete the set of data on as many todownloadthedatawhichiscollected plots as possible are important. The best option is tocontinuously. This data includes: carry out all the surveys on as many plots as possible. If it is not possible to equip all plots in a Meteorology country, it is strongly recommended to concentrate Meteorological measurements will be taken on a the continuous measurements on soilsolution, continuous basis using an automated weather station meteorological parameters, deposition and ambientdescribed below. A NADP type wet/dry collector air quality at a smaller number of plots. These plots will also be deployed. are best selected taking into account the need of statistical analysis. Water Flow If all measurements mentioned under points a - Streamflowwillbemeasuredonthree k (Table 1) are carried out at the same plot, this plot supercritical, trapezoidal flumes located on the is then called a key plot. All countries are invited to Workman Creek North, Middle and South Forks. establish at least 10% of their Level II plots as key Water yield based on stage height and flume ratings plots. While all Level II plots contribute at a certainwill be measured with modern electronic stage degreetoimprovetheunderstanding of the height sensors. Stream stage heights are converted to cause -effect relationship (objective b) the key plots flow volumes based on hydraulic rating formulas.

Table 1: Surveys carried out on Level II plots Survey Frequency Intensity Crown condition at least annually all plots Soil (solid phase) every 10 years all plots Soil solution continuously part of the plots Foliage every 2 years all plots Deposition continuously part of the plots Ambient air quality continuously part of the plots Meteorology continuously part of the plots Forest growth every 5 years all plots Ground Vegetation every 5 years all plots Phenology several times per year optional Litterfall continuously part of the plots Remote Sensing preferably at plot installation optional willprovidethesupplementaryinformation necessary to fulfill objective c (to provide a deeper Ambient Air Quality insight into the interactions between the various Ambient air quality is measured by two methods: components of forest ecosystems by compiling (1) a Ozone monitor measures atmospheric ozone in available information from related studies). the concentration range 1.5 ppbv to 100 ppmv using the established technique of UV absorption at 254 Data Collection nrn. These data are recorded with a data logger The exact protocol for data collection in the US which will be downloaded duringsite visits. (2) ICP Level II network is still to be determined. Once Passive samplers will collect samples for nitric the instrumentation described below has been oxide, nitrogen dioxide, sulfur dioxide & ammonia. 8

Most likely these samples will be shipped to the BENEFITS OF THE RESEARCH PSW Riverside station for testing and analysis. ThedatacollectedattheSierraAncha Again the exact protocol is yet to be determined. Experimental Forest will contribute to a better understanding of the relationships between the Soils condition of forest ecosystems and anthropogenic (in Soils are evaluated every ten years (solid phase) particular air pollution) as well as natural stress as well as continuously (liquid phase). The solid factors through intensive monitoring on a number of phase evaluation will begin in 2008, after monitoring selected permanent observation plots and to study instrumentation has been installed. Soil solutions the development of important forest ecosystems. willbecollectedtogetherwiththeother Further, the results will provide a deeper insight into continuously compiled data and will is collected by the interactions between the various components of an integratedsoiland groundwater pollutionforestecosystemsbycompilingavailable monitoring unit. The sampler will collect soil water,information from related studies, and in close soil gas, monitor soil moisture, temperature andco- operation with the ICP on modeling and mapping conductivity, and store soil water samples in the soil willcontributetothecalculationof critical until further analysis can be performed. levels /loadsandtheirexceedancesinforest ecosystems. Ideally this will improve collaboration Vegetation with other environmental monitoring programs as Overstory composition, understory, and forest well and contribute to the monitoring activities to floorwill be determined on theestablished other aspects of relevance for forest policy at monitoring site at given time intervals (Table 1). national and global level, such as effects of climate changes on forests, sustainable forest management Litterfall and biodiversity in forests. Lastly, the network will Throughfall collection systems will be located provide policy -makers and the general public with throughout the monitoring site and will be co- located relevant information. with the lysimeters and near the deposition samplers. The physical samples will be collected on a monthly basis with the remaining continuously recorded data. 9 Appendix: Standard Level II plot design (example)

n forest stand Id

ísa lb FI (ss'::....ib Fxb S o $ o Y S o a o

S o s 5si F3 s S o o o o Y Id s S o 4sSj S a ° ° ß Fh Ys S o s s 0 o O S o lb o o S S o S Id 0 o S lb F4 lb F2 Ys Y Ib Y !,5)

path

road In nearby open area

PS Yb M

Crown assessment Foliar o Tree for crown assessment (yearly) Fx Tree x for foliar sampling (2- yearly)

Soil (every 10 years) Increment s Location for soil sampling (minimal o Trees for increment measurement DBH disturbance) (5- yearly) S Location for soil pit Ib Trees for increment measurement bores (once) SS Soil solution samplers Id Trees for increment measurement disks (once)

Deposition and ambient air quality Meteorology Y Throughfall collector M Meteorological equipment Ys Stemflow collector (in beech mandatory) Yb Bulk or wet -only collector Ground vegetation PS Passive Samplers ? ? ? Ground vegetation sampling areas (always located outside fence) A CONTRAST AMONG NATIONAL FOREST WATERSHED PROGRAMS: 1978 - 2008

Robert E. Lefevre'

Watershed management program on the Coronado that all the Forests in New Mexico and Arizona had National Forest has changed dramatically in theto research all water uses and determine if any thirty years between 1978 and 2008. This paperneeded to be certified through the respective State presents the changes brought about by changes inwater codes. Over 2,000 water uses on the national leadership, program emphasis, fundingCoronadoNationalForestalonerequired level, and technology.The Coronado National certification. Fortunately, Arizona had laws in effect Forest is located in Southeastern Arizona andat the time that allowed for registration of existing Southwestern New Mexico, in the Southwesternuses.Nearly all of the hydrologists' time on the Region. The watershed management program on theForest was consumed bytheresearchand Coronado National Forest has been influenced byregistration of water rights during 1978 and 1979. four factors throughout the period of 1978 -2008: National Leadership;local Program Emphasis; Funding Level Funding Level for the program; and Technology Budget not an issue.A staff of four Soil available. Each of these factors is addressed briefly Scientists and two Hydrologists was adequately below. The meaning of "National Leadership" has funded.Timber and range programs were well been identified by the Chief of the Forest Service in funded and drove most activity, while watershed office at the time. fundsprovidedforanaggressivewatershed restoration program. The majority of all water uses 1978 -1979: Chief John McGuire were for range allotment administration, and range National Leadership funds paid for the registration costs. Chief McGuire was nearing the end of his term in office in 1978. The National Forest Management Technology Act of 1976 was just beginning to be implemented. The entire Forest Service had one computer, McGuire is quoted as saying "perhaps the greatestlocated at the Fort Collins Computer Center. Land challenge facing forestry today is the calendar -Management Planning efforts tried to use that namely the arrival of the 21st century.My questioncomputer by requiring the use of "FORPLAN" (U. is, will American forestry be ready to meet the 21st S. Congress, Office of Technology Assessment, century ?" (http: / /www.foresthistory.org/, 2004). 1992). With nearly all National Forests competing for time, scheduling difficulties and computer time . Program Emphasis backlogs were common.The planning process A Supreme Court Decision shaped local dragged on for years. program emphasis. The Mimbres Decision of 1978 took the Forest Service by surprise (Gillilan and 1979 -1987: Chief Max Peterson Brown, 1997).The decision said that while the National Leadership Multiple -Use Sustained -Yield Act of 1960 (Public Chief Peterson took on the task of implementing Law 86 -517) was intended to broaden the purposes the National Forest Management Act of 1976 from for whichnational forests had previously beenJohn McGuire.He was concerned about the administered, Congress did not intend to reserve potential for controversy and was quoted as saying additional water in forests previously withdrawn "...the resources of these (National Forest) lands are under the 1897 Act. The effect of the decision was wanted by a large number of diverse users who see them as critical to meeting their future needs. Many 'Coronado National Forest' U.S. Forest Service, also see their own desired use as either exclusive of Tucson, Arizona other potential users or at least incompatible with 12 them.In any language, that spells controversy" Computer Center. By the end of 1986, almost all (http: / /www.foresthistory.org/ 2004). Forest Service employees had a Data General Computer on their desk. Electronic messages Program Emphasis (e -mail) and cell phones began to play a role in Water rights issues continued, as water rightseveryday work. registrations were winding down, the Gila River Adjudication proceedings began. More filings were 1987 -1993: Chief Dale Robertson required. The budget droveother program National Leadership components, first a large amount of erosion control In response to the arrival of controversy as and other watershed restoration work, then addingpredicted by Chief Peterson, Chief Robertson led other assignments not directly related to watershed. efforts by the Forest Service to find new and creative A third hydrologist was added to the Coronado staff. ways to manage the national forests, especially by The Forest Plan on the Coronado National Forest emphasizing non -commodity (non- timber) resources, was completed in 1986, and the Forest set aboutnew forestry, new perspectives, and ecosystem trying to implement it. The Plan included standards management.Chief Robertson is credited with for riparian areas and direction to determine howcoining the "caring for the land and serving people" many Coronado National Forest riparian areas met slogan (http: / /www.foresthistory.org/ 2004) those stands. It also called for at least 50 percent of all riparian areas to be in satisfactory condition by Program Emphasis the end of the first 10 -year period. No definitions The wildfire season during 1987 in the west was for "satisfactory condition" were given, and thehuge and widely spread, including Yellowstone standards referred to "natural conditions" with noNational Park and multiple National Forests in definition for what "natural" was. A Forest protocolCalifornia. Locally on the Coronado National for riparian area assessment was developed in 1984, Forest,a prescribed burn intheChiricahua based almost entirely on vegetation parameters evenMountains escaped and caused concern about future thoughthePlan hadstandardsforchannel prescribedburning programsandeffectson morphology as well. watershed components. Regional Forester Sotero Muniz called for better management of riparian Funding Level areas, resulting in a regional level effort to develop The 1980 budget came with money and targets a riparian area assessment and monitoring protocol. for watershed restoration. Burchard Heede of theThe nationalcallfor ecosystem management Rocky Mountain Research Station had publishedresulted in the range management program on the more than 80 articles on gully rehabilitation and soil Coronado to come under intense scrutiny, especially erosion control (USDA Forest Service, Rocky with regard to how it affected riparian areas. The Mountain Station, 1993, Stream Notes, 1993). His new Regional protocol for riparian area assessment work was used to design erosion control projects.called "Riparian Area Survey and Evaluation" However, by 1987 budgets had become an issue for (RASES) came out in 1989 and was pressed into the watershed management program. The Coronado service to respond to the demand for information and NationalForestreducedstaffingfromthree proposals to improve riparian areas that didn't hydrologistsand foursoilscientiststoone measure up.It included channel morphology and hydrologist and one soil scientist, then assigned soil parameters in addition to vegetation.The other duties including forestry program management Coronado, along with the Tonto National Forest and special uses management to these individuals. modified RASES to fit local situations. The initial response to the assessments was to fence cattle out Technology of riparian areas, but criticism mounted, as it seemed Computers became more available, but using to be contrary to the national direction for holistic them to design projects still required the Fort Collins ecosystem management.The upland watersheds 13 began to suffer a threat in addition to grazing andprovide detailed soil and vegetation information was wildfire as a bark beetle attack killed a large number looked to, as was the RASES information and of mature ponderosa pine trees. Arizona Department of Environmental Quality Water Assessments. The Rescission Act of 1995 Funding Level (Rescission Act of 1995 (Public Law 104 -19 Section Budgets began to become a serious concern as 504(a)) required range allotment permits to have an the timber and range programs lost their power, and environmentalanalysisundertheNational the budget process did not include "ecosystem Environmental Policy Act (NEPA) completed before management"asa lineitemortargetthey could be issued.This caused watershed accomplishment. Thesolutionswerenot assessments to be prioritized in favor of those immediately apparent. allotments needing analysisfirst. Nearly all watershed program activities in 1996 were in Technology response to range allotment NEPA. Computers, pagers, and cell phones became An unprecedented event in 1994 occurred when a increasingly important to day -to -day operations. lightningignitedwildfireintheChiricahua Typing pools, which traditionally had carried out Mountains burned over 27,000 acres mostly in most tasks involving the preparation and filing ofmixed conifer forest, affecting every watershed on documents, disappeared and all employees were the range.Rucker Lake, the popular recreation expected to do their own document preparation on development in Rucker Canyon was completely their computers. The time required not only to do filled with sediment following the fire, and was these added tasks, but to get trained on the rapidly abandoned.(Lefevre and Neary, 1999).Several developing technology was becoming a greater large forest fires occurred in the Rincon Mountains fraction of allemployees', not just watershedand Huachuca Mountains, requiring emergency management personnel, work time. burned area assessments.

1993 -1996: Chief Jack Ward Thomas Funding Level National Leadership Range program funding began to increase to Chief Thomas supported the implementation of respond to the NEPA analyses. Watershed program a new ecosystem management approach on thepersonnel were funded through this, as watershed national forests and grasslands. He is quoted as line items declined. saying "ecosystem management is not just a timber sale; it's putting the timber sale into a bigger picture, Technology including the watersheds,wildlife,roads, and In addition to more computer technology, all people'sneedsandvalues..."terrain vehicles were being added to our fleet. (http: / /www.foresthistory.org/ 2005). Improved mobility and access made the assessments somewhat more efficient, but training to keep up Program Emphasis with new technology offset some of the benefits. Watersheds were often thought of as a starting point to define areas in which to accomplish 1996 -2001: Chief Mike Dombeck "ecosystem management." Ecosystem management National Leadership resulted in a need for more information about Chief Dombeck focused his efforts on promoting watershed conditions at a variety of scales. Most ofpartnerships,collaborativestewardship, the data needed to do assessments was not available, accountability, and financial health.An avid so searches for data began in earnest. The General fisherman, he considered watershed management an Ecosystem Survey, completed at the Regional level important task: "Watershed maintenance and for Arizona and New Mexico in 1990 as an oversight restoration are the oldest and highest callings of the part of the terrestrial ecosystem survey which would Forest Service" ( http : / /www.foresthistory.org/2004). 14

Program Emphasis 2001 -2007: Chief Dale Bosworth In spite of the call for watershed programs to National Leadership take their place at the center of National Forest Chief Bosworth continued the recent trend in management, anincreasedemphasis onfuel nationaldirectiontobuildrelationshipsand management arose in response to large wildfirespartnerships. He tried to get more traditional both locally and on the national scene. This in turnauthority to the District Rangers by saying the resulted in a need to assess watershed impacts from "decisions need to be made at the lowest level that moremechanicalvegetationtreatmentsand they can. We need to build better relationships with prescribed burns in addition to the continuedlocal communities and with states,tribes, and demandforanalysisofrangeallotments. others" (http: / /www.foresthistory.org/, 2007). Endangered aquatic species on the Coronado National Forest began to take center stage as drought Program Emphasis caused habitats to shrink The last remaining As national emphasis pointed toward more local populationsof nativefrogsandfisheswere relationships, the Coronado was forced into intense frequently foundinwatersfoundtoharbor local situations that required relationships with the introduced species and disease in addition to beingcounties and communities. Wildfires in the Santa located in range allotments. The demand for dataCatalina, Pinaleño, Tumacacori, Huachuca, and and analysis to develop proposed actions to save Santa Rita Mountains required detailed burned area these habitats became intense.The RASES andassessments. Large burned area emergency response newly developed "Soil Quality Assessment" (FSH(BAER) teams were formed, absorbing large 2509.18, 1999) were stretched to their limits in amounts of watershed personnel time, and that after attempts to help decision makers. extended tours on suppression assignments by the same individuals.Implementation of the BAER Funding Level projects, followed by long term restoration needs Funding levels stayed flat in the face of rising and project implementation resulted in the watershed costs of doing business. Local watershed programprogram being dominated by wildfires. The personnel,alreadyassignedotherprograms unexpected number of large fires nationally resulted including forestry and special uses, diversified by in a renewed call for action to manage forests. The getting training to support wildfire suppressionHealthy Forest Restoration Act (HFRA) (PL actions. 108 -148) of 2003 called for specific action that requiredsomewatershedassessment. The Technology Rescission Act schedule also required continued Geographic Information Systems (GIS) began assessment to complete range allotment analyses. to become the accepted way to conduct and display nearly all analyses. The programs required training Funding Level and space to run, and computers began to fail as The HFRA proj ects and range management NEPA more and more demands were placed on them.provided funding, as did a Southwestern Region Bigger, more powerful computers began to replaceinitiative to gather data for potential instream flow the original Data General models, and by 1997 manywater rights.New hydrologist and silviculturist documents created on the Data General computerspositions were added to the Coronado organization were no longer available.Hard copies became to assist in these programs. invaluable as it was discovered that we could no longer read or print documents created as recently as 1994. 15

Technology CONCLUSIONS Digital cameras began to be more common, as Thirty years of change, especially in national did global positioning system (GPS) units. Both ofleadership and technology, have had little effect on these new technologies could be linked to the GISlocal programs for the Coronado National Forest. projects, creating complex data storage situations Water rights and riparian areas continue to dominate requiring even more powerful computers. the program. Plan revision offers the opportunity to make riparian area assessment and management 2007 -present: Chief Abigail Kimbell more meaningful, but how to accomplish it remains National Leadership difficult. As population and demands on the Forest Chief Kimbell has started her term continuing increase while drought threatens day -to -day supply, the emphasis on ecosystem management.She is water use and rights remain important topics. quoted as saying "we have to manage for the health of the whole landscape- for clean water, for wildlife REFERENCES habitat,for healthy vegetation,for recreation" Gillilan, David M. and T. C. Brown. 1997. (http: / /www.foresthistory.org/ 2008). Instream Flow Protection: Seeking a Balance in Western Water Use. Island Press, 417 pages. Program Emphasis http: / /www.foresthistory.org/. 2004 Forest restoration with an emphasis on fuelshttp: / /www.foresthistory.org/ 2005 management and range allotment NEPA continue tohttp: / /www.foresthistory.org/ 2007 demand much of the watershed program attention. http: / /www.foresthistory.org/ 2008 Burned area restoration and response to situationsLefevre, Robert E. and Daniel G. 1999. Rucker within the burned areas also commands attention. Lake: A history of recent conditions affecting a Forest Plan Revision is requiring a new look at water southeastern Arizona watershed. Hydrology and resources. The Gila River adjudication and other Water Resources in Arizona and the Southwest continuing water rightsissues(refer to1978 29: 23 -30. discussion) are still in progress, thirty years after Multiple -Use Sustained -Yield Act of 1960. they began, with no end in sight. Public Law 86 -517 Rescission Act of 1995. Public Law 104 -19 Funding Level The Healthy Forest Restoration Act (HFRA). Budget levels remainstatic. Watershed 2003. (PL 108- 148).USDA Forest Service, personnel continue to respond to projects and 1999. FSH 2509.18 - SoilManagement programs, and are funded by them. In addition to Handbook, R3 Supplement 2509.18 -99 -1 forest restoration and range management, new United States v. New Mexico, 438 U.S. 696. mineral development is starting to command a lot of 1978. Mimbres Decision of 1978. 698 -718. attention and time. U. S. Congress, Office of Technology Assessment.1992. Forest Service planning: Technology accommodating uses, producing outputs, and All the computer hardware purchased and sustaining ecosystems. OTA -F -505 maintained over the last decade is slowly being USDA Forest Service, Rocky Mountain Station. upgraded and replaced. Software support has been 1993. Stream Notes April 1993, p.7. centralized. SOUTH -TO -NORTH WATER DIVERSION PROJECT IN CHINA

Hui Chen' and Peter F. Ffolliott'

The South -to -North Water Diversion Project incompatible with the distribution of the population China is the largest water project of its kind ever and agricultural, industrial, and social development. undertaken in the world.It was proposed in 1952The level of water shortage in China in 2000, and the main construction was started in 2002 afterrepresentingthewaterresource,socialand 50 years of baseline inventorying and study. Theeconomical development, water supply,water project is expected to be completed in .2050 with a demand, water shortage and water quality, indicates total investment of approximately 486 billion yuans sufficient water resources in southern China but a (1 dollar = 8 yuans in 2002). The South -to -Northdeficiency of water resources in northern China Water Diversion Project will not only relieve the(Figure 1).The shortage of water resources in severe water shortage in northern China, but, more northern China will only get worse with the importantly,itwillefficiently manage wateranticipated social and economical development. The resourcesbyestablishingastrategicwater South to North Water Diversion Project is expected distribution and security network for the Yangtze to relieve this problem. River, Yellow River, Huai River, and Hai River. The project consists of three routes - the Western THREE ROUTE PROJECT Route, the Middle Route, and the Eastern Route - Since beginning to consider the feasibility of diverting water from upstream, mid- stream, andimplementing the South -to -North Water Diversion downstream of the Yangtze River, respectively. The Project in early 1950s, more than 150 alternative total water to be transferred in the project is layouts for the project had been proposed, with the expected to be nearly 44.8 billion m3 annually by ThreeRouteProjectultimatelychosenfor 2050, with 17.0 billion m3, 13.0 billion m3, and 14.8implementation. The Three Route Project (Figure billion m3 of water transferred by the Western Route,2), including the Western Route Project, the Middle theMiddleRoute,andtheEasternRoute, Route Project, and the Eastern Route Project, will respectively. divert water from upstream, mid- stream,and This paper on the South -to -North Projectdownstream of the Yangtze River, respectively, to consists of a background explanation of why Chinaconnect the four major rivers - the Yangtze River, needs this water project; a section on how the Three Yellow River, Huai River, and Hai River - into a Route Project will be constructed; the expectednational water resource network. social,economical, and ecology benefits;and existing and potential future impacts of the project Eastern Route Project on ecology and the environment. Construction of the Eastern Route Project started in December 2002 and is planned for BACKGROUND completion in 2016, with a total investment of 65 The large population in China (about 1.3 billion billion yuans. The Eastern Route Project will divert people in 2002) is unevenly distributed, with a high water from a downstream branch of the Yangtze density of people in eastern China and a low density River to the eastern Huang- Huai -Hai Plain, with the in western China. The spatial distribution of watertermination in the Tianjin Municipality via the resource is also uneven, with higher resources in the existing Beijing -Hangzhou Grand Canal, which is south and east than in the north and west.The the oldest and longest canal in the world.The current distribution of water resources is not Eastern Route Project is expected to relieve water shortagesintheTianjinMunicipality,the Heilonggang Yundong regionsintheHebei 'School ofNatural Resources, University of Arizona, Tucson, Arizona 18

2000 *710 g AA*1413z.ft VA 3 ( WA. 41 T )

j > o. 65 ¡ j0. 5 - 0. 65 0 0.35 - 0.5 ¡-]

cP111zMqyk0,4#144ft,.Ki-ìtFlr

Figure 1. Distribution of the water shortage index in China in 2000. (Source: (http: / /sdinfo. chinawater. net. cn/waterresources /ww007. j pg)) province, and the north and southwest part of Middle Route Project Jiaodong Peninsula of Shandong Province.As The Middle Route Project was started in already stated, it will divert water from the YangtzeDecember 2003 and its completion is expected in River, which normally has mean annual water flow 2050, with a total investment of 117 billion yuans. of 960 billion m3 entering the Eastern Sea and more The Middle Route Project will divert water from the than 760 billion m3 of water flow in extremely dryDanjiangkou Reservoir on Han River to the Beijing years. The length of the completed water -course Municipality in the short-run. In the long -run, the will be about 1,560 km from the Yangtze River toMiddle Route Project will divert water from the the Tianjin Municipality. The Eastern Route Project Three Gorges Reservoir or downstream of the dam involves establishing 13 pumping stations south ofon the main branch of the Yangtze River to the the Yellow River where the water will then flow to Beijing Municipality, the Tianjing Municipality, and the Tianjin Municipality by gravity. The Easternthe provinces of Hebei, Henan, and Hubei. The Route Project also involves the construction of a main advantage of this project is that water will flow nearly 9 km tunnel beneath Yellow River. When by the forces of gravity. Two key components of the completed, the Eastern Route Project is expected to construction effort for this project are heightening divert approximately 14.8 billion m3 of water perthe Danjiangkou Reservoir dam from 97 to 170 m year. and building two tunnels of 8.5 m in internal 19

MONGOLIA 1.-vevEr4 Beijingng :'b1 ...toGOLfr. ... _ .. ,T i rt SCUT H Cr .:50 Cenbrat route rSH C k`iisr SeR fi:w-1A:Ne`xE Western routesî Hat!, Eastern .._.... mute. Shang#laá East CPen a r.'_ Sa7,6! á. I F.l':1F., f gr :7e i40': G! . IWAN T.tïret, Cit?x'+.`á 00011' (unr1iltr construction) C IA. IIsI DIA H g Korig r SrAi`rr. China Sea

tu1 N'ta IN t/1 .p 4n' I CTtJA%.,r, Figure 2. Layout of the Three Route Project of South -to -North Water Diversion. (Source: (http: / /www.water- technology. net / projects /south_north /south_north l .html)) diameter and about 7 km long with a planned flow It will divert water from the upper reach of the capacity of 500 m3 /s.After the completion of the Yangtze River to Yellow River. The Western Route DanjiangkouReservoir,themeanannually Project is planned to transfer 20 billion m3 of water transferred water will be increased from 12.0 to 14.0 annually from three tributaries of the Yangtze River, billion m3 in normal years and 6.2 billion m3 in dry including 10 billion m3 from the Tongtian River in years. Another benefit of Danjiangkou Damthe upper reach of Yangtze River, 5 billion m3 from Extension Project is that it will increase the level ofthe Yalong River, and 5 billion m3 from the Dadu flood control in the middle and lower Han River and River. Two main parts of the construction effort for enhance the safety of Wuhan City and the plain to the West Route Project are building a dam with a the north of the Han River. The diversion route ofheight more than 200 m and digging a tunnel through the Middle Route Project will be about 1,274 km in the Bayangela Mountain with the length of more length, starting at the head of the Taocha canal and than100 km. This project willnecessitate terminating at the Yuyuan Pool in Beijing.The overcomingmajorengineeringandclimatic Middle Route Project will supply water to Beijingchallengesbecauseitissituatedonthe for the 2008 Olympic Games. The transferred water Qinghai -TibetPlateau where theelevationis is temporarily from reservoirs in the Hebei province, between 3,000 and 5,000 m and is characterized by with an annul flow rate of 0.4 billion m3. complicated geological structures, extremely low temperatures, and frequent and severe earthquakes Western Route Project up to 8 to 9 degrees on Richter scale in some areas. The Western Route Project is expected to start in 2010, with the investment of 300 billion yuans, an BENEFITS amount accounting for 60% of total investment for The South -to -North Water Diversion Project the South -to -North Water Diversion Project. The will eventually divert about 38 to 48 billion m3 of Western Route Project is expected to relieve thewater per year.It is expected to have important water shortage in the north and northwest of China. social, economic and environmental benefits for the 20

Huang- Huai -Hai District by alleviating the existing areas.The project will also increase the water watershortage,promotingsocio- economic supply to agriculture and livestock purposes, adjust development, and improving living conditions for theagriculturalcrop components toimprove the nearly 300 million residents of urban and rural efficiency,and increasetheoveralllevelof areas. Some of these benefits are discussed below. agricultural production. The diverted clean water will replace sewage water for agricultural irrigation Social Benefits purposes and, in doing so, reduce pollution of The South -to -North Water Diversion Project agricultural lands and improve the quality of will have social benefits to the water district. Foragricultural production. The increasing availability example, Beijing, the capital of China, is the nation's of clean water will also reduce excessive extraction center of political, cultural, financial and diplomaticof groundwater and, as a consequence, alleviate the activities.Tianjin is the largest industrial base inproblem of declining aquifers. By reasonable water northern China and, additionally, an important tradedistribution,it will be possible to recharge dry port.The western and northwest region is thewetlands, creeks, and streams, which will likely country's base of energy resources, raw materials, increase the ability of self -dilution and self -cleaning and heavy chemical industries. The South -to -North of water resources and promote the development of Project will ease the competition for water between aquatic biological resources. geographic regions and between the agricultural and industrial sectors.It will also reduce or relieve ECOLOGICAL AND ENVIRONMENTAL regional health problems such as fluoride bone and IMPACTS thyroid problems due to drinking contaminated deep As the largest water project in the world, the groundwater. South -to -North Water Diversion Project certainly has existing problems and could have problems in Economical Benefits thefuturesuchasfinancial,technical,and It is planned that 40% of 17 billion m3 of watermechanical problems and,asdiscussedhere, transferred by the Eastern Route Project will be ecological and environmental problems. Poor water allocated to urban and industrial uses and 60% to quality for the Eastern Route Project, severe soil rural and agricultural uses. About 65% of 13 billion erosion for the Middle Route Project, and feasibility m3 of water transferred by the Middle Route Project and reliability for Western Route Project are among will help to satisfy urban and industrial needs and the most crucial of these possible problems. 35% will be made available to rural and agricultural uses. One -half of the 14.8 billionm3 of water to be Eastern Route Transfer Project transferred by the Western Route Project will go to The main problem with the Eastern Route urban and industrial uses and 50% to the rural and Project is the poor quality in the water output agricultural sectors. Based on recognized criteria for district.The project uses the Beijing -Hangzhou agricultural and industrial outputs and the currentGrand Canal and other natural rivers and lakes as its price level, the average annual economic benefitwatercourse. Main sources of pollution at this time from the South -to -North Diversion Project will be are the point- discharges of sewage pollution from approximately 60 to 80 billion yuans. Importantly, living and industry; surface discharges of water it is also expected that the financial inputs andcontaining chemical fertilizers and pesticides that outputs for this project will come into balance in less are applied widely on agricultural lands around the than 10 years. Huai River, the Yellow River, and the Hai River; the high sediment concentrations resulting from the Ecological and Environmental Benefits movement of large sediment deposits within the The transferred water will increase the supply transferred water; and pollution resulting from the of water to people and the industrial sector, improve heavy traffic of shipping that takes place on the sanitation, and enhance the vegetative cover in urban Beijing -Hangzhou Grand Canal. Water pollution 21

control and its treatment will be the first step to occurrence of higher degrees of earthquakes in this insuring that the diverted water meets established area; constructing a tunnel of a 100 km length and a quality standards. Planned pollution control projects dam with a height of 200 m is a technical and include building urban sewage treatment system, mechanicalchallengeconsideringthe industrialsewer diversion,adjustment of the geomorphology characteristics in this mountain area; industrial structure, and comprehensive management and a recognition that the natural ecological system of industrial and watershed treatment. is fragile, and, as a consequence, the ability of the ecosystem to sustain itself is unknown after the Middle Route Project massive of grassland and agricultural lands are The major problem for this project is that the destroyed. And finally, the Tibetans living in this ecological environmentisdeteriorating. The area of the project worship at highly regarded mountain areas with steep slopes and a high level oftemples; the high mountains of the region have ecosystem fragmentation are prone to severe soil special meaning to the people; and the natural waters erosion.Another concern for the Middle Route of the area have a great value. Of particular note, Project isthat the quality of water is gettingrelocation and rebuilding of the temples is another progressively worse due to the increasing dischargeslarge problem. of living and industrial wastewater into the Han River and Dan River. Treatment of soil erosion and CONCLUDING COMMENT wastewater, therefore, will be the highest priority in Taking50yearsfromconceptionto diverting clear water to the north. commencement, the South -to -North Water Diversion Project is expected to require another 50 years to Western Route Project complete its construction. After it is completed in The Western Route Project will divert water 2050, at a total investment of 486 billion yuans, this from the upstream of the Yangtze River to the large project will eventually transfer 44.8 billion m3 Yellow River. Itislocated on the eastern of critically needed water per year to benefit the Qinghai -Tibet Plateau, with elevations of 3,000 togrowing population centers of the drier regions of 5,000 m. There is inherently a high diversity ofnorthern China. It is indeed at unique undertaking. geomorphology and rich natural resources in the project area. Fortunately, the change of water flow REFERENCES should not have significantly adverse impacts on the Information for this paper was obtained from the biological stock of the region that relies mainly onfollowing Chinese government Web sites: precipitationinputsforsurvival. Although Sustainable Development in China construction on the project will begin in 2010, its (http://www.sdinfo.net.cn) feasibility and reliability is still debated, since it is Chinese South -to -North Water Diversion Project both an inter -basin water transfer project and a (http://www.nsbd.gov.cn) large -scale ecological and environmental project. National Bureau of Statistics of China Current arguments are centered around the main (http://www.stats.gov.cn) issues that building the high dam could trigger the TRANSPIRATION OF OAK TREES IN THE OAK SAVANNAS OF THE SOUTHWESTERN BORDERLANDS REGION

Peter F. Ffolliott,' Cody L. Stropki,' Aaron T. Kauffman,' and Gerald J. Gottfriede

Transpirationof oaktreesontheCascabelCascabel Watersheds have been described by watersheds in the savannas on the eastern slope ofGottfried et al. (2007) and, therefore, are not the Peloncillo Mountains in southwestern New included in this paper. Mexico has been estimated by the sap -flow method. Transpiration represents the largest loss of gross STUDY METHODS precipitationfallingonawatershedin Instantaneous transpiration of 16 Emory oak approximations of water budgets for the more (Quercus emoryi) trees, the dominant oak species on densely stocked oak woodlands of the Southwestern the Cascabel Watersheds, was measured with a Borderlands region (Ffolliott 2000, 2004, Shipek et sap -flow meter (Swanson and Whitfield 1981, al. 2004).Knowledge of transpiration is alsoSwanson 1994, Schaeffer et . al. 2000) in the late important, therefore, in developing a general water spring, summer, and early fall of 2004. The budget for this more open ecosystem. An initialmeasured trees were located on two transects estimate of the transpiration of oak trees in the oak oriented perpendicular to the main stream channel savannas of the Southwestern Borderlands region is on each of two Cascabel Watersheds, specifically, presented in this paper. Watersheds E and I (Gottfried et al. 2007). Two of the trees were situated on the southerly aspect and STUDY AREA two trees on the northerly aspect of each transect for Twelve small watershedshave been a total of 16 measured trees. The trees ranged in drc instrumentedandsampledinthePeloncillo (diameter rootcollar)from 6to14inches. Mountains to study the effects of cool seasonParenthetically, errors of 5 to15 percent are (November- April) and warm- season (May- October) commonly associated with the measurement of prescribed burning on the multiple resources of oakinstantaneous transpiration of individual plants with savannas in the Southwestern Borderlands.The a sap -flow meter (Shuttleworth 2008). watersheds are located near the Cascabel Ranch in The instantaneous measurements of transpiration the eastern part of the Coronado National Forest of the Emory oak trees measured were transformed near the Animas Valley of southwestern New intoapproximations of daily transpiration by Mexico. The watersheds are 5,380 to 5,590 feet inapplying the equations of Barret et al. (1995) and elevation.The weather station at the CascabelSchaeffer et al. (2000). The resulting estimates of Ranchheadquartersindicatesthatannual daily transpiration were considered to be a proxy for precipitation in the vicinity of the watershedsother oak species on the Cascabel Watersheds. averages 21.8 " 1.2 inches, with more than one -halfEmory oak represented 75.5 percent of the oak trees falling in the summer monsoon from late Juneon these watersheds, with Arizona white oak (Q. through early September. However, a prolongedarizonica) and Tourney oak (Q. toumeyi) comprising drought impacted the Southwestern Borderlands 14.3 and 10.2 percent, respectively. region from the middle 1990s to the present time. Earlierstudies of transpirationinthe oak Annual precipitation during this drought period has woodlands on the southern slope of the Huachuca averaged14.9inches. Geological,edaphic, Mountainsindicatedthatvaluesofdaily hydrologic, and vegetative characteristics of the transpiration of oak trees cycled through time on an annual basis, ranged from relatively high values of 'School of Natural Resources, University of transpiration following the summer monsoons and Arizona, Tucson, Arizona winter rains to intervening low values (Ffolliott and 'Rocky Mountain Research Station, U.S. Forest Gottfried 1999). It was hypothesized by the authors Service, Phoenix, Arizona of this paper that this cycle of annual transpiration 24 rates also applies in the oak savannas. To test this and their respective drc values was coupled with the hypothesis, the estimates of daily transpiration rates Cascabel stand table, which presented tree frequency obtained forthe oak trees on the Cascabel peracre,toobtaintheestimateof annual Watersheds were plotted on a figure representing transpiration in area -inches.Information on the this cycle in the oak woodlands of the Huachucafrequency of oak trees was obtained from an earlier Mountains (Figure 1). inventory of the tree overstories on the watersheds (Gottfried et al. 2007).

15C, 200 3,00 353 135 .,ta,. Jut: ;ìn date Figure 1. Average daily transpiration rates (with standard deviations) by oak tress in the woodlands of the Huachuca Mountains. (Source Ffolliott and Gottfried 1999.)

RESULTS AND DISCUSSION All of these plotted estimates fell within the The estimated annual transpiration of the oak trees confidence intervals in this figure. It was concluded, in the oak savannas on the Cascabel Watersheds is therefore, that the cycle illustrated in the figure is nearly 4.8 area- inches. This estimate is about 60 applicable to the oak savannas.It was furtherpercent of the annual transpiration of mature oak concluded that the relationship between annual trees (60 years and older) in a stand on the Huachuca transpiration and the drc of oak trees in the oak Mountains (Ffolliott and Gottfried 1999, Ffolliott et woodlands on the Huachuca Mountains (Figure 2) is al. 2003). The last (known) harvest in the Huachuca also representative of the oak savannas on the stand selectively removed larger oak trees for mining Cascabel Watersheds. Relationship illustrated intimbers in 1895, and, therefore, the stand appeared figure 2 became a basis for obtaining an estimate ofto be largely representative of structure of uncut annual transpiration of the oak trees in the oak stands in the region when sap -flow measurements to savannas in terms of area -inches. The relationship estimate transpiration were made in 1997 -1998. The between estimated annual transpiration of oak trees lower value for annual transpiration in the oak 25

WÑ 40uC

O 4x r =0886 3.=Mt)

IC 25 30 40 drc (centimeters)

Figure 2. Relationship between annual transpiration and drc (diameter root collar) of oak trees in the woodlands of the Huachuca Mountains. savannas is attributed largely to the smaller numberecosystems in the region. Accepting this value as a of oak trees in the oak savannas compared to the oakbasis for developing a similar expression for the woodlands. The average number of oak trees ontranspiration of oak trees in the oak savannas theCascabel Watershed is 42.6 stems per acre indicated that the annual transpiration of the oak (Ffolliott and Gottfried 2005), while that in thetrees on the Cascabel Watersheds is approximately sampled stand in the oak woodlands on the 30 percent of annual precipitation. Comparisons of Huachuca Mountains was 183.1 stems per acre this value with the relationships obtained for the (Ffolliott et al. 2003). differentstandconditionsandsilvicultural Annual transpiration of oak trees can also betreatments in the oak woodlands are shown in figure expressed as a percent of annual precipitation. An 3. arbitraryannualprecipitationvalueof 450 The most valid comparison presented in figure 3 millimeters (about 17.7 inches) had been selectedis that between the annual transpiration of oak trees earlier as a basis to present this percentage for in the oak savannas on the Cascabel Watersheds and varying stand conditions and silvicultural treatments the oak trees in the uncut stand of mature oak trees in the oak woodlands on the Huachuca Mountains inin the oak woodlands of the Huachuca Mountains earlier studies (see Table 1 in Ffolliott 2004). This (Ffolliott and Gottfried 1999, Ffolliott et al. 2003). precipitation value is within a range of the variable These two sites are generally similar in the past and annual precipitation amounts occurring in the oakpresent land -use practices imposed on them. The 26

Transpiration by Oak Trees

OAK SAVANNAS All trees OAK WOODLANDS Mature trees Partial (65 %) harvesting Sprouts Unthinned stump sprouts Three stump sprouts Two stump sprouts One stump sprout

o 20 40 60 80 100 Percent of Net Annual Precipitation

Figure 3. Annual transpiration of oak trees in the oak savannas as a percent of the (arbitrarily) selected annual precipitation value of 450 millimeters (about 17.7 inches) and varying stand conditions and silvicultural treatments in the oak woodlands on the Huachuca Mountains.Percentages for the oak woodlands were obtained from the studies reported by Ffolliott and Gottfried (1999), Ffolliott et al. (2003), Ffolliott (2004), and Shipek et al. (2004). other comparisons presented in the figure are offered trees in the harvested stand were thinned to one, two, for information only. and three of the dominant sprouts in 1984 to The partial firewood harvest removed 65 evaluate the effects of coppice thinning treatments percent of the mature oak trees in the oak woodlands on residual growth and volume (Bennett 1988, of the Huachuca Mountains in 1981. TranspirationTouchan and Ffolliott 1999, Farah et al. 2003). of the residual mature oak trees and the resulting Transpiration was measured with a sap -flow meter stump sprouts (combined) was measured with a sapon 16 of these rootstocks representing each of these flow meter in 1997 -1998 approximately 16 yearsthinning treatments and an unthinned control in after the firewood harvest. One factor contributing 2000,approximately 20yearsfollowingthe to the greater estimated transpiration in this stand firewood harvest (Shipek et al. 2004). Expressions was the large number of stump sprouts (Ffolliottand of these annual transpiration values as a percent of Gottfried 1999). Although the number of residual the annual precipitation amount selected in the mature oak trees in this stand was (statistically)earlier studies are also shown in figure 3. Note that similar to the number of trees in the uncut stand ofthe transpiration of stump sprouts on the unthinned mature oak trees, there was a (very) large number ofroot -stocks is similar to the transpiration of stump post -harvest stump sprouts. Even though the sprouts in the partially harvested site, because of the estimated transpiration of an individual stump sprout overwhelming dominance of stump sprouts. was comparatively small, the large numberof sprouts in the stand translated into the greater annual CONCLUSIONS transpiration value (Figure 3).There were 370.2 Annual transpiration of oak trees in the oak stump sprouts per acre in the stand when the savannasisabout 60 percent of theannual measurements of transpiration were made (Ffolliotttranspiration in a stand of mature oak trees in the oak and Gottfried, Ffolliott et al. 2003). This number ofwoodlands. The lower value estimated for the oak sprouts was considerably larger than that normallysavannas was likely due to differences in the found in the oak woodlands (Touchan 1988). densities of oak trees in the two ecosystems. The The stump- sprouts on rootstocks of some of the number of oak trees in the oak savannas is less than 27 that in the oak woodlands.Further study is Resources in Arizona and the Southwest 30:37 -41 necessary, however, to verify or refine this initialFfolliott, P. F. 2004. A water budget for Emory estimate. Also, other tree species including alligator oak woodlands of southeastern Arizona: An juniper,border pinyon, and thetree -form of expansionoftheinitialapproximation. mesquite are often present in the oak savannas, and, Hydrology and Water Resources in Arizona and therefore, an estimate of the transpiration for these the Southwest 34:11 -14. trees species is required to obtain a comprehensive Ffolliott, P. F., and G. J. Gottfried. 1999.Water use estimate of transpiration in this ecosystem. by EmoryoakinsoutheasternArizona. Knowledge of the transpiration by herbaceous Hydrology and Water Resources in Arizona and plants on the Cascabel Watersheds would strengthen the Southwest 29:43 -48. our understanding of water use by most components Ffolliott, P. F., and G. J. Gottfried. 2005. of the vegetative community. These data combined Vegetative characteristics of oak savannas in the with existing measurements of precipitation and Southwestern UnitedStates: A comparative streamflow would then facilitate more accurate analysis with oak woodlands in the region.In: estimates of the water balance of the Southwestern Gottfried, G. J., B. S. Gebow, L. G. Eskew, and C. Borderlands Region. B. Edminster, compilers. Connecting mountain islandsanddesertseas:Biodiversityand ACKNOWLEDGMENT management of the Madrean Archipelago. U.S. This study and the preparation of this paper was ForestService,Proceedings RMRS -P -36, supportedbytheSouthwesternBorderlands 399 -402. Ecosystem Management Project of the RockyFfolliott, P. F., G. J. Gottfried, Y. Cohen, and G. Mountain Research Station, U.S. Forest Service, Schiller. 2003. Transpiration by dryland Phoenix, Arizona, and the Arizona Agricultural oaks: Studies in the south -western United States Experiment Station, University of Arizona, Tucson, and northern Israel. Journal of Arid Environments Arizona. 55 (2003):595 -605. Gottfried, G. J., D. G. Neary, and P. F. Ffolliott. REFERENCES 2007. An ecosystem approach to determining the Barret, D. J., T. J. Hatton, J. E. Ash, and M. C. effectsof prescribedfireon Southwestern Ball. 1995.Evaluation of the heat pulse Borderlands oak savannas: A baseline study. In: velocity technique for measurement of sapflow Masters, R. E., and K. E. M. Galley, editors. Fire in rainforest and eucalypt forest species of ingrasslandandshrublandecosystems. southeasternAustralia. Plant,Celland Proceedings of the 23rd Tall Timber Fire Ecology Environment 18:463 -469. Workshop,TallTimbersResearchStation, Bennett, D. A. 1988. Effects of coppice treatment Tallahassee, Florida, pp. 140 -146. on Emory oak.In: Ffolliott, P. F., and J. D. Shipek, D. C., P. F. Ffolliott, G. J. Gottfried, and L. Hasbrouck,editors. Oak woodland F. DeBano. 2004. Transpiration and multiple use management: Proceedings of the workshop. management of thinned Emory oak coppice. U.S. School of Renewable Natural Resources, Forest Service, Research Paper RMRP- RP -48, 8 University of Arizona, Tucson, Arizona, pp. p. 31 -37. Schaeffer, S. M., D. G. Williams, and D. C. Farah, M. H., P. F. Ffolliott, and G. J. Gottfried. Goodrich. 2000. Transpiration of 2003.Growth and volume of Emory oak cottonwood/willow forest estimated from sap flux. coppice 10 years after thinning.Western AgricultureandForestMeteorology Journal of Applied Forestry 18:77 -80. 105(2000):257 -270. Ffolliott, P. F. 2000. An annual water budget for Shuttleworth, W. J. 2008. Evapotranspiration Emory oak woodlands: An initial measurement methods.Southwest Hydrology approximation. HydrologyandWater 7(1):22 -23. 28

Swanson, R. H. 1994. Significant historical Touchan, R. 1988. Growth and yield of Emory developments inthermalmethodsfor oak.In: Ffolliott, P. F., and J. D. Hasbrouck, measuring sap flow in trees. Agriculture and editors. Oak woodland management: Proceedings Forest Meteorology 72:113 -132. of the workshop. School of Renewable Natural Swanson, R. H., D. W. A. Whitfield. 1981. A Resources,Universityof Arizona,Tucson, numerical analysis of heat pulse velocity theory Arizona, pp. 11 -18. and practice. Journal of Experimental Botany Touchan, R., and P. F. Ffolliott. 1999. Thinning 32(126):221 -239. of Emory oak coppice: Effects on growth, yield, and harvestingcycles. The Southwestern Naturalist 44:1 -5. HOW USEFUL IS LIDAR IN ESTABLISHING A STREAM GAUGING NETWORK IN A TROPICAL EXPERIMENTAL FOREST

Boris Poff,' Daniel G. Neary,' and Gregory P. Asner'

In the late summer of 2007 the Institute for PacificThree streams have cut their stream beds through Islands Forestry (IPIF), which is part of the US65,000 year old rock formations, while the other ForestServicePacificSouthwestern Researchchannels exist on newer bedrock, which formed Station, asked the USFS Rocky Mountain Research between 4,000 and 65,000 years ago. All channels Station's (RMRS) Air, Water and Aquatic Program'sare located in Akaka soils generic class, with the (AWA) Southwest Watershed Science Team forexception of the Kaawalii stream, which is located assistance in the establishing baseline data in the on Honokaa soils. initial phase of a long term research project in the newly established Hawaiian Experimental Forest. RESEARCH NEEDS RMRS and its predecessors have a history of the None of the stream channels in the HETF had long standing watershed related research in the previouslybeensurveyed.Sinceitsrecent southwestern United States which was establisheddesignation as an experimental forest, it became experimental watersheds on volcanic rock over 50 prudent to properly inventory, classify and survey years ago. Both systems, in Hawaii and Arizona, the streams channels in the HETF. These activities have mostly washes that have flashy precipitationare also necessary to select possible locations for events driven flows. Funding availability, personnel stream gauging stations along those channels. While shortages and other ongoing collaborative projects certain criteria for a stream gauge location can be between experimentalforests of theresearchevaluated using a GIS, such as proximity to roads, branches of the U.S. Forest Service facilitated the forest boundaries etc. other criteria, such as channel joint effort in this project. morphology still needs to be evaluated in the field. The first stream selected to be surveyed was the STUDY SITE Kaawalii Stream in the Wet Forest vegetation zone The 12,343 -acre Laupahoehoe Unit of theof the HETF. This is one of five streams on Hawaii Experimental Tropical Forest (HETF) isLaupahoehoe Unit and is the number one priority for located in North Hilo on the Isle of Hawaii and was future research efforts, mostly due to its accessibility formally established on state land on March 23,and proximity to one of the few roads in the HETF. 2007.It is one of two experimental forests in the The first 2 km reach of the Kaawalii Stream fall HETF network (Figure 1). within the Experimental Forest. The objective of this Located upslope of former sugarcane lands, the study is to complete a Kaawalii Stream baseline sitealso contains timber plantations, degradedgeomorphic survey and add one experimental pastures and numerous streams. Its location and size gauged watershed to the EF. Gathered data will be make it an ideal global location for global climateused for future studies of the effects of climate change research studies of native forest restoration, change on hydrology channel substrates, aquatic invasivespeciescontrol,andwatershed organisms, and coarse woody debris. Required work management. The Laupahoehoe unit contains fiveconsists of the establishment of cross -section, GPS streams, of which only two are perennial and contain monumenting,photodocumentation,channel aquatic organisms. classification, geologic substrate, and pebble counts. Thisincludesrunninglevelsurveysof the 'Rocky Mountain Research Station, US Forest cross -sections to provide a baseline of channel Service, Flagstaff, Arizona conditions, and ground -truth data for the LiDAR 'Carnegie Institution, Stanford University, Palo Alto, survey, establishing of coarse woody debris status in California thestreamsand conductingbaselineaquatic organism surveys. Location of the Laupahoehoe 30

Experimental Forest in the Hawaii Experimental Field Methods Tropical Forest as shown in figure 1. The lower 900 m of the 2 km reach of the Kaawalii Stream within the HETF were surveyed duringthelastweekofSeptember2007. Cross -sectionswereestablishedatrandomly Hawai'i Experimental Tropical Forest

Estaässiird AdaroH 23. 2C07 pre -selected locations on successive 100 m segments

The Maa7íS E.}perimn* Trap+aat wrast oorssn of h4a urdn, of the stream. The cross -sections were permanently 1a. Laupesl+oaHoa Una ¡Amax 12,343 awn) art eta Pita LVa'aw'a :MO (wpm 38.93,! aorrcs;. monumented, photo -documented, located with GPS coordinates,channel -typedusingtheRosgen method, and measured for substrate composition using the pebble -count method. These cross -sections constitute ground -truth data for a LiDAR database being constructed for the HETF.

DISCUSSION OF METHOD While the CAO LiDARwas intended to specialize invegetation,itcanalsoidentifyhidden geomorphologic features in the landscape. The CAO scanindicatesaseriesof depressions -water catchments and channels on the forest floor. These features have not been documented to this extent and detail as a whole. The CAO covers a large area and provides levels of detail beyond what can be ei 0 achieved on foot, but the question is: Can it be used uenar 1s.- vw in establishing a stream gauging network and Data s.cacr. Stan o/llama O$. xuaawc .eaameaa Lana ud Natant R.so.rws determining other channel characteristics? The flata a PmtuaLOn; ASay23, 20C? Maat Hawaiian Islandsfi I:anlaa RonaldCant/ink. OAR Map hèaMuFW.D709r6 .. simple answer is no. The LiDAR output is helpful in b 25 V s ,o ss ?c 25 Hann rA±.a identifying the main stream channels on the map and displaying major geomorphic features such as Figure 1. Tthe Laupahoehoe Experimental Forest in the waterfalls and channel junctions. However, field Hawaii Experimental Tropical Forest. work isstill necessary to identify geomorphic features that are required to establish a stream gauge. METHODS These include classifying the substrate, channel The methods used in this study consist of a classification, substrate size range, slope, and degree LiDAR mission flown over the HETF and field work of entrenchment. Some stream characteristics, such conducted by a survey crew in the Kaawalii stream as sinuosity can be determined byLiDAR. The CAO channel. Both projects were carried out in 2007. data can be helpful in planning the field work, determining access points, channel paths, major LiDAR channel alterations etc., but it can not serve as a In the summer of 2007, the Carnegie Airbornereplacement for the actual field work. Aside from its Observatory (CAO) LiDAR was flown at 2000 muse in planning for stream gaugingactivities, the a.g.l., at 50 khz pulse repetition rate, which resulted LiDAR could be more useful if collected at higher in 1.0 m spacing between laser spot centers. The spatial resolution (e.g. « 1 m), which can easily be resolution of each laser spot was 1.2 m. Maximum accomplished by flying the system at lower altitude. scan angle was limited to 18 degreesoff -nadir. For the current data set, the LiDAR was flown at high altitude (12,000 'm.s.l.) to cover the entire 4,995 hectare reserve in a few hours. 31

Future work is proposed for the 2008 Fiscal ACKNOWLEDGMENTS Year to complete the Kaawalii stream surveyThe Carnegie Airborne Observatory (CAO) is (includinglevelsurveys,cross -sectionfunded by the W.M. Keck Foundation, William establishment,GPSmonumenting,photoHearst III, and the endowment of the Carnegie documentation, channelclassification,geologic Institution for Science. substrates, pebble counts, coarse woody debris status) and complete one additional stream survey within the HETF, which also includes a baseline aquatic organism survey for the perennial stream. COMPARING BEDLOAD CONDITIONS IN THE CASCABEL WATERSHEDS, CORONADO NATIONAL FOREST

Karen A. Koestner,' Daniel G. Neary,' Gerald J. Gottfried'

Oak savannas and woodlands are a significant ResourcesConservationService,andtheir ecosystem type of the Southwestern Borderlandsassociates has amassed considerable hydrological spanning approximately 800,000km' (31,000 mi2). and ecological information on the oak savannas of However, there is little hydrologic data available tothe Peloncillo Mountains since 2002. This paper aid in the informed management of these lands compares changing bedload conditions between (Gottfried and others2001). Therefore,an 2003 and 2006 on four of the Cascabel watersheds ecosystem -scale experimental watershed study was scheduled for treatment. This informationis established to study the hydrology and ecology ofnecessary to the understanding of hydrologic and southwestern oak savannas, as well as evaluate the geomorphologic characteristics of these watersheds effects of cool and warm season prescribed fires onprior to treatment.These observations are also multiple ecosystem components. important for they may influence the treatment Firewasthemostsignificantnatural assignment blocking design used for this study disturbance in southwestern oak savannas prior to pending further analysis. European settlement. However, due topast over -grazing and fire suppression practices, fire has SOUTHWESTERN OAK SAVANNAS been far less frequent on the landscape (Neary and This ecotypeisdominant in the Coronado Gottfried 2004). These management activities have National Forest of southern Arizona and New caused ecosystem changes and increased fuel Mexicocoveringapproximately342,800ha accumulations which could contribute to stand (847,000ac).Theseoaksavanna/woodland replacing fires (Kruse and others 1996). Prescribed ecosystems span elevations from 1,220 to 2,225 m fire is a proposed management technique to restore (4,000 -7,300ft),withannualextremesin natural processes within oak savannas by reducing precipitation ranging from 305 to 1,016 mm (12 to woody species density, increasing herbaceous plant 40 in) (Gottfried and others 2002). production, and creating vegetative mosaics on the Hydrologic information pertaining to the oak landscape. Questions concerning the seasonality ofsavannas is limited (Lopes and Ffolliott 1992). The burn treatments and the overall effects of these bulk of hydrologic research in the region has been treatments on hydrologic and ecologic processes conductedintheChihuahuanDesertnear need to be addressed prior to broad management Tombstone, AZ (Osterkamp 1999). There is a lack application. Twelve small watersheds on the eastern ofinformationconcerningsurfacerunoff slope of the Peloncillo Mountains of southwestern characteristics in oak savannas (Gottfried and others New Mexico, termed the Cascabel Watersheds, were2002).Surface runoff is affected by spring and selected to address these questions. winter rainfall and snowmelt events; however, the A collaborative research effort between the US majority of runoff is from high intensity summer Forest Service's Rocky Mountain Research Station rainfall (Gottfried and others 2002).These high andCoronadoNationalForest,theMalpai intensity precipitation events can accelerate erosion Borderlands Group, the Animas Foundation, theand sedimentation (Hester and others 1997). Good International Arid Lands Consortium, the Natural watershedconditionconsistingofhealthy well- stockedstandsof treesand herbaceous 'Rocky Mountain Research Station, U.S. Forest vegetationisnecessarytoavoidaccelerated Service Flagstaff, Arizona sedimentation and erosion that could impact water 'Rocky Mountain Research Station, U.S. Forest quality (Lopes and Ffolliott 1992). However, due to Service, Phoenix, Arizona increasing fuel accumulations in the oak savannas, there is increased potential for stand replacing 34 wildfire which can have substantial impacts onof the experimental area averaged 597 mm (23.5 watershedprocesses- increasingerosion, inches) between 1981 and 1999 (Gottfried et al. sedimentation and creating hydrophobic soils (Allen 2007). 1995). The parent material in the study area is rhyolite -a fine- grained volcanic rock.The common soils THE CASCABEL WATERSHEDS documented in the 1991 General Ecosystem Survey Twelve small watersheds were selected for of this area of the Peloncillo Mountains (completed study on the east side of the Peloncillo Mountains by the Southwestern Region of the Forest Service) (Figure 1) in southwestern New Mexico to ascertain are generally Typic Haplustalfs, mesic, deep, the impacts of cool season and warm seasongravelly loam compacted or deep very cobbly prescribed burning on oak savanna ecosystems. The sandy -loam gullied (Neary and Gottfried 2004). The watersheds are located north of Whitmire Canyon in soils in the area have been classified as rock land the Coronado National Forest, approximately 50 km with bedrock at 0 -30 cm- however, exposed bedrock (31 mi) south of Animas, New Mexico.The is common. watersheds range in size from 8 -34 ha (20 -83 ac) at The Cascabel watersheds were selected based on elevations between 1,664 to 1,692 m (5,460 -5,550 their potential usefulness for addressing questions ft). Annual precipitation at the Cascabel Ranch east related to multiple ecosystem components of the oak

3?' 3

A Clanton I)raw New Memo) Contour Wert al = 20ft Datum= NAD 191' 12It

Deahuent Bouudaues Control Warm Season Btun 'ool Season Bui n

Figure 1. Watershed location and current experimental design: H, A, K,and G are controls, C, E, J, and M will be treated with a cool season burn and B, I, F, Mwill be treated with a warm season burn. The changing bedload conditionsof watersheds H, A, N, F could result in reassignment of the treatments. 35

savannas such as flora and fauna, hydrologicalaffecting bedloads over time and space (e.g., the processes and soil properties. Usefulness is judged flow conditions and the supply of bed -load material) by multiple logistical and morphological factors. must be addressed. Watershed criteria includes the ecotype of interest Ward and Trimble (2004) define bedload as (oak- savanna),accessibility,distinctchannel sediment transported along a channel bed by a formations (measurability), adequate gauging sitescombination of sliding, rolling, and saltation that for flume installation, and being relatively similar in contributes to building point bars and banks in the size, slope and elevation. The Cascabel Watersheds main channel. Bedload conditions and their are representative of many oak -savannas and open contributingfactorsareassessedhere bya woodland sites of the Southwest borderlands basedcombination of survey methods that assess physical on elevation, annual and seasonal precipitation, andand morphological characteristics. Substrate cover percent of oak cover (Gottfried and others 2002). Six was determined for the reach of each watershed, watersheds were selected on the north and south while channel morphological characteristics of slope sides of a ridge between Whitmire and Walnut and cross -section formation were collected at the Canyons and divided into groups of three. Eachpermanentcross -sections.Channelcoverand group of watersheds have two burn treatments andmorphology data provide insight into the amount of an untreated control. Watersheds A through G areerosion and sediment movement occurring over time on the south -side of the ridge and watersheds H (Ward and Trimble 2004). through N are on the north -side (Figure 1). A channel condition survey was completed in 2003 to determine initial substrate cover of each EVALUATING BEDLOADS channel. The physical condition of the channels, Changing bedload conditions for watersheds A, whether or not they are comprised of fine sediment, H, N and F are evaluated in this paper.These vegetative cover, or long reaches of exposed watersheds are on the eastern (N and F) and westernbedrock, is thought to potentially have a strong (A and H) extremes of the study area. They have influence on hydrologic response (e.g. peak flows, been selected to contrast differences in bedload sediment movement, etc). conditions and total precipitation based on aspect Channel condition was evaluated by completing and east/west orientation. Table 1 outlines the line transect surveys in 100 m increments upstream topographic characteristics such as elevation, area, from the sediment weir through the reach of the watershed and channel slope, and aspect of selectedwatershed until no distinctchannel could be watersheds (A, F, H and N). Bedload dynamics, determined (including side channels). The substrate which are characterized by great temporal andcondition of the channel was measured and averaged spatial variability, are extremely important to aidby watershed (Figures 2a and 2b). understanding of channel evolution (Renault and The second survey method used in the comparison Regüés 2006). To gain a perspective on theof bedload conditions determines cross -sectional changing bedload conditions,the mechanismsformation. This was measured by surveying

Table 1. Topographic characteristics of selected Cascabel watersheds, Peloncillo Mountains. WSAreaMean Mean Channel Channel Aspect Tc (ha) Elevation (m) Slope ( %) Length (m) Slope ( %) degrees (hours) A 13.8 1692 22.8 380 5.6 160 0.13 F 12.5 1665 16.2 429 6.4 190 0.228 H 22.3 1695 18.7 550 5 180 0.126 N 10.9 1672 9 421 3.6 360 0.283 *Tc (time of concentration) is the time it takes for runoff to travel from the most distant point of the watershed to the outlet. (Source: Gottfried GJ, Neary DG, Bemis RJ. 2002.) 36

50.0% 40.0% 30.0% 20.0% 10.0% 0.0% R AF AC VD O R AF AC V \NI3 O 2a 2b Figures 2a and Z. Channel characteristics for western (A and H) and eastern (F and N) oriented. watersheds. Substrate abbreviations used, R -rock, AF- fine alluvium sediment, AC- coarse alluvium sediment, V - vegetation, WD- woody debris, and O- other. elevational changes in channel gradient across another irregularity. Instead of sediment deposition, transects using a stadia rod and surveying level. The some channel incision or degradation were observed cross -section data were then entered into a stream between 2003 and 2006. The causality of this will channel modeling program called Mike 11 (Figures need to be determined after further analysis. 3 -6).Five cross -sections for each watershed of One trend that holds true for the four watersheds interest (A, H, F, and N) were modeled in twoexamined is that the watersheds on the western separate years between 2002 and 2006 to compare portion of the study exhibit more sediment transport changes in bedload conditions. Elevation in meters and accumulation. The topography and size of the is on the y -axis and channel width in meters is on theindividual watersheds, their physical condition (i.e., x -axis. The cross -section highlighted in black is the channel condition), and their orientation on the same in the side by side images and will be used to landscape (east or west and northern or southern determine changing bedload conditions. The otheraspect)arethemajorvariablesimpacting information presented in these figures (i.e. red and morphological changes over time as seen in the blue lines) are not pertinent to this comparison and cross- section analysis. Research on coarse sediment will not be discussed. transport by Reid and others (2007) states that There is a visibly significant change in bedload certain areas of the watershed could act as key conditions on watershed A between 2003 and 2006 sedimentsources. The inconsistentbedload (Figures 3a and 3b). The elevation of allconditions betweentheeasternandwestern cross -sections has risen -implying an increase in watersheds presented here could be impacted by a deposited sediment. Watershed H (Figures 4a and greater number of sediment sources or substrate 4b) shows a similar trend as watershed A, though homogeneity on the western side of the study area. there is a less pronounced increase in elevation byHowever, these are simply conjectures based on individual cross -sections. Both of these watersheds preliminary observations of the calibration data. are situated on the western portion of the study area For this comparison of bedload conditions, and they have similar bedload trends. The changes channelconditionandcross -sectionalsurvey are different in the eastern oriented watershed Ninformation were combined with the data from a (Figures5aand5b). Significantsediment high -intensity precipitation event on August 23, accumulation only occurs upstream, which may be 2005. The objective of this bedload comparison was due to erosion or a rock slide event.Further to determine whether or not storm/ precipitation investigation on the ground and modeling oftrends correspond with the observed changes in cross -sections are needed to ascertain the causalbedload condition. Tables 2 and 3 quantify the peak factor. Watershed F (Figures 6a and 6b) presentsflows following that precipitation event and the 37

Figures 3a. and 3b. Watershed A cross-sectional model for August 2003 and March 2006, respectively.

1 Figures 4a and 4b. Watershed H cross-sectional model for August 2003 and March 2006, respectively. 38

Figures 5a and 5b. Watershed N cross -sectional model for July 2002 and July 2006.

Figures 6a and 6b. Watershed F cross -sectional model for August 2003 and May 2006. 39 amount and duration of precipitation on a perinformation presented here is a relative comparison watershed basis. Only the watersheds of interest are of cross -section changes over time and does not presented. In theinitial experimental design, actually quantify the amount of sediment movement watersheds were blocked into those facing north andthat has occurred. thosefacingsouth and then replicated block treatments were assigned as shown in Figure 1. The CONCLUSIONS rationale guiding this decision was that aspect This preliminary assessment of bedload conditions (northern vs. southern exposure) will have the intheCascabelwatershedshighlightssome

Table 2. Peak storm flows, August 23, 2005

SouthernPeak Flow Peak FlowNorthern Peak Fl ow Peak Fl ow Aspect (cfs) (cm s) Aspect (cf0 (cms) A 67.2 1.903 H 762 2.175 F 24.6 0.697 N 31.3 0.886

Table 3. Precipitation for August 23, 2005

Total Total We other Precipitation Precipitation Start Time End Time Station (inches) (cm) H 2.36 5.99 12:00 1b:00 A-B 2.64 6.71 12:01 15:44 F-G 1.36 3.45 12:09 15:49 N 1.43 3.63 12:21 15:52 greatest impact on total precipitation and peak flows.important factors for consideration regarding this However, the data in Tables 2 and 3 (which reflectresearch project and watershed research as a field. average storm data on the Cascabel watersheds) Further analysis and modeling should be completed show east and west orientation as a more significantprior to the planned treatments. The data presented factor than aspect.Peak discharge and effectivehere support a reassignment of replicated blocks runoff have considerable impacts on bedload and based on an east/west boundary instead of a therefore need to be viewed in tandem (Renault andnorth/south one. However, this requires complete Regüés 2006). Table 3 documents typical stormcross -sectional modeling, some time -series analysis movement that begins in the west and moves and an analysis of covariance to determine the most eastwardresultinginagreateramountofimportant sources of variability between watersheds precipitation and corresponding peak flows in during the calibration period.Utilizing calibration western oriented watersheds (A and H). Althoughdatapre- treatmentandcontinuedmonitoring aspect appears to be a factor there is less variation post- treatment are an attempt toincrease the between watersheds on north and south slopes (Aknowledge of the hydrologic and ecologic processes and H, N and F) as is between watersheds on eastof southwestern oak savannas to improve overall and west sides of the study area (A and F, H and N). land stewardship for a myriad of private and public The changes observed in cross -section analysis stakeholders. match the variation in precipitation data remarkably Observing bedload changes during the calibration well, although they require further analysis. Theperiod further supports an adaptive and flexible 40 approach to the experimental design used in a Hester, J.W., Thurow T.L., Taylor C.A., Jr. 1997. paired -watershedstudytoachievemultiple Hydrologic characteristics of vegetation types as objectivesanddecreasevariabilitybetween affected by prescribed burning. Journal of Range replicates. By examining these data prior to Management 50: 199 -204. prescribed fire treatments, we are able to improveKruse W.H., Gottfried G.J., Bennett D.A., the design of this study, and therefore provide more Mata- Manqueros H. 1996. The role of fire in conclusive data for land managers.The initial Madreanencinaloakandpinyon juniper analysis also provides support for gathering and woodland development. In: Ffolliott P.F., DeBano utilizingsubstantialcalibrationdatapriorto L.F., Baker M.B. Jr., Gottfried G.J., Solis -Garza treatment application. G., Edminster C.B., Neary D.G., Allen L.S., Hamre R.H. technical coordinators. Effects of fire REFERENCES on Madrean Province ecosystems. U.S. Forest Allen, L.S. 1995. Fire management in the sky Service, General Technical Report RM- GTR -289, islands.In:DeBano, L.F.,Ffolliott,P.F., pp. 99 -106. Ortega- Rubio, A., Gottfried, G.J., Hamre, R.H., Lopes, V.L., Ffolliott P.F. 1992. Hydrology and Edminster,C.B.,technicalcoordinators. watershed management of oak woodlands in Biodiversity and management of the Madrean southeastern Arizona. In: Ffolliott P.F., Gottfried Archipelago: the Sky Islands of southwestern G.J., Bennett D.A., Hernandez C., Ortega -Rubio United States and northwestern Mexico. U.S. A., Hambre R.H. technical coordinators. Ecology ForestService, General Technical Report andmanagementofoakandassociated RM- GTR -264, pp. 386 -388. woodlands: Perspectives in the southwestern Barton, A.M. 1999. Pines versus oaks: Effects of fire United States and northern Mexico. U.S. Forest on the composition of Madrean forests in Service,General Technical Report RM -218, Arizona. Forest Ecology and Management 120: pp.71 -77. 143 -156. Neary, D.G., Gottfried G.J. 2004. Geomorphology of Gottfried, G.J., Neary D.G., Bemis R.J. 2001. small watersheds in an oak encinalin the Watershed characteristics of oak savannas in Peloncillo Mountains. Hydrology and Water the southwestern borderlands. Hydrology and Resources in Arizona and the Southwest 34: Water 65 -70. Resources in Arizona and the Southwest 30: Osterkamp, W.R. 1999. Runoff and sediment yield 21 -28. from proxy records: Upper Animas Creek Basin, Gottfried, G.J., Neary D.G., Bemis R.J. 2002. New Mexico. U.S. Forest Service, Research Paper Assessing the impacts of prescribed burning on RMRS- RP -18, 50 p. soil and water resources of oak savannas in the Reid, S., Lane, S.N., Berney, J.M., Holden, J. 2007. southwestern UnitedStates.In Assessing The timing and magnitude of coarse sediment capabilities of soil and water resources in transport events within an upland. Temperate drylands: the role of information retrieval and gravel -bed river. Geomorphology 83: 152 -182. dissemination. Proceedings of the International Renault, N.L., Regüés, D. 2006. Bedload transport AridLandsConsortiumconferenceand underdifferent flow conditions in a workshop. Tucson, Arizona, pp.115 -122. human -disturbed catchment in the Central Spanish Pyrenees. Catena 71(1): 155 -163. Ward, A.D., Trimble S.W. 2004. Environmental hydrology, 2nd ed. Lewis Publishers, New York. CHARACTERISTICS AND BEHAVIOR OF A COOL -SEASON PRESCRIBED FIRE IN THE OAK SAVANNAS OF THE SOUTHWESTERN BORDERLANDS

Karen A. Koestner,' Daniel G. Neary,'Gerald J. Gottfried,' and Ruben Morales'

Oak -savannas and woodlands comprise over 80,000 km2 (31,000 mil) in the mountains and high valleys of the southwestern United States and northern Mexico (Figure 1).Fire, which was once the most important natural disturbance in this system, has been excluded duetoover -grazingandfire suppressionpractices. This hasresultedin ecosystemchangesandfuelaccumulations. Prescribed fire is one management technique to restorenaturalprocesseswithinsouthwestern oak -savannas by reducing woody species density, increasing herbaceous plant production, and creating vegetative mosaics on the landscape. Seasonality of burn treatments and their effects on physical and ecological processes need to be determined prior to broad management application.The Cascabel Watershed Study is a collaborative interdisciplinary Figure 1. Extent of Southwestern oak woodlands study to determine the effects of cool- season and and savannas. (Source: Brown and Lowe 1980.) warm- season prescribed burning on a southwestern oak -savanna ecosystem. Twelve small watersheds in (Quercus emoryi), Arizona white oak (Q. Arizonica), the Peloncillo Mountains of southwestern Newand Tourney oak (Q. toumeyi) are the most prevalent Mexico have been monitored for seven years tospecies in the Peloncillo Mountains (Brown 1982, provide hydrologic data prior to prescribed burningFfolliott and Gottfried 2005). These oaks are treatment application (Figure 2). These watershedstypically small, multi -stemmed, and irregularly aregroupedinfourreplicatedblocks,each formed with variable stand densities (Gottfried et al. consisting of a cool- season treatment, a warm- season 1995). The oak species native to this region are fire treatment and a control watershed. This paper willenduring- experiencing top kill / shoot mortality but discuss the characteristics and behavior of thesprouting following fire (Barton 1999). However, cool -season burn treatment, summarizing burnperiodic fire is required to limit establishment and intensity and severity estimates. growth of woody species in savannas and allow for herbaceouscompetition(VanAuken2000). SOUTHWESTERN OAK SAVANNA Alligator juniper (Juniperus deppeana) is also a VEGETATION common overstory species in this region. Redberry Multiple species of oaks are found in thejuniper ( Juniperus coahuilensis), pinyon (Pinus southwestern oak savannas, however Emory oak cembroides),andvelvetmesquite(Prosopis velutina) are lesser components of the oak -savanna 'Rocky Mountain Research Station, U.S. Forest overstory (Ffolliott and Gottfried 2005). Service, Flagstaff, Arizona The savanna or grassland component of this 'Rocky Mountain Research Station, U.S. Forest ecotype has multiple bunch grass species including Service, Phoenix, Arizona blue (Boutelouagracilis), sideoats (B. curtipendula), 'Douglas Ranger District, Coronado National slender (B. filiformis), and hairy (B. hirsuta) grama, Forest, Douglas, Arizona bull muhly (Muhlenbergia emersleyi),wolftail 42

(lryutou'Draw 'New Mexico Contour Interval = ?Cft Datuni= NAD 1927 12R

Ilea fluent 'Bouuclauì es ....:..;...

...... Control I Warm Season Bum -1 Cool Season Burn Figure 2. Cascabel watershed study area and treatment designations. Watersheds H, C, K, and N were burned in March 2008 for the cool- season treatment component of the Cascabel watershed study.

(Lycuruspheoides),andTexasbluestemAuken 2000, Webster et al. 2001). (Schizachrium cirratum) (Brown 1982, Ffolliott and Gottfried 2005).Some forbs including Fendler FIRE IN THE BORDERLANDS ceanothus(Ceanothusfendleri)and Mexican Fire has been a central component of the cliffrose (Pushia Mexicana) can be found in development of all southeastern Arizona ecosystems southwestern oak savannas. Intermixed shrub (Allen 1995)..Prior to the 1890s wildfires were species include Beargrass (Nolina microcarpa), fairy common in the Southwestern borderlands whenever duster (Calliandra eriophylla), sotol (Dasylirion sufficient fuel loads were present. These were both wheeleri), and manzanita (Arctostaphylos spp.). lightning and anthropogenic fires (Bahre 1991) Yucca (Yucca spp.) and agave (Agave spp.) are alsogenerally occurring in drought years following a wet scattered throughout the oak -savanna ecotype. period when there was large increase in herbaceous Woody species (mostly oaks) have increased in growth (Allen 1995). Fire would typically burn in both density and total cover on the landscape (Hesterthe dry lightning season from May to July prior to et al. 1997, Van Auken 2000). Native woody plant summer monsoons with low severity impacts (oaks) encroachment has also decreased the cover of(Gottfried et al. 2001). The presence and importance perennial grasses (Van Auken 2000, Ottmar et al.of fire on this landscape is further attested to by 2007). Woody encroachment is the result of bothtree -ring fire scars at higher elevations, the rapid environmental conditions and management practices. post -fire recovery and/or enhanced vegetation, Grasses inhibit woody species during early phases ofhistoric records, and the high frequency of summer growth, but are out -competed once the roots oflightning ignitions (Bahre 1991). woody plants are below the root zone of grasses, a Fire exclusion for the past century in this region situation that is maintained by fire exclusion (Vanhas had considerable ecological effects (Kruse et al. 43

1996). The historicfireregimein savannacool- season burn at varying severity across the ecosystems is estimated as four to eight years priordesignated watersheds. Up to 60 -70% of shrubs to European settlement (Kaib et al. 1999). Several within the oak -savannas were to be eliminated as dominant species such as Emory oak and Arizonawell as the top kill of 25-40% of overstory species. white oak are adapted to fire and will sprout afterThe treatment prescription also called for the their tops have been killed off, though severe fire reduction of 1 hr (by 30 -80 %), 10 hr (by 10 -40 %), results in mortality (Ffolliott and Bennett 1996, 100 hr (by 1 -10 %), and 1000 hr (by 1 -20 %) fuels in Barton 1999). Fire exclusion has resulted in fuel the watersheds. The environmental prescription for accumulation and increased stand densities (Hester this burn required that temperatures fall between etal.1997). These conditionsincrease the 50 -105° F (day) and 30 -70 °F (night) with relative probability of stand- replacing high -severity wildfires humidity ranging from 4 - 30 %. (Gottfried et al. 2002). This was the first experimental burn treatment on Prescribed burning is a potential treatmentthe Cascabel watershed study to determine the techniquetorestorepre- Europeansettlement impacts of cool vs. warm- season prescribed fire on conditions within the oak savannas (Gottfried et al.Southwestern oak -savannas. The overall purpose of 2002).Restoring a more "natural" composition the burn is the reduction of hazardous fuels, while includesdensity reduction of woody species,maintaining fire's natural role in the Peloncillo increasing herbaceous plant production, and creating Mountains. Continued monitoring of post -fire mosaics of vegetation on the landscape.Currently response to determine the effects of cool and warm however, there is limited information concerning the season prescribed burning on hydrologic and impacts of prescribed fires on southwestern oakecologic processes will be addressed in future savannas (Barton 1999). The seasonality of fire is studies. an important characteristic because it determines whether or not vegetation will burn during the DETERMINING BURN INTENSITY AND dormant (cool) season in winter /spring or during the SEVERITY summer growing (warm) season. Brockway et al. Burn intensity is the heat produced per unit length (2002) documented grassland response to dormant or of fireline during combustion of a given fuel (Fyne cool season fire with an increase grass cover, forbet al.1996, Neary et al. 2005).Intensity was cover, and species richness. While growing or warm measured by two methods: infrared thermometer season fire reduced graminoids and forbs cover and guns and temperature indicating paints.Infrared increased litter cover.Variable removal ofthermometers were used to take temperatures of vegetation cover and exposure of soil between differentfuelsduring combustion. Table 1 seasonal burn treatments may alter watershed summarizes the range of temperatures exhibited by functioning by increasing erosion, sedimentation and various fuels.The ranges of values displayed in peak flows (Neary et al.2005).Therefore, ifTable 1 (minimum to maximum burn temperatures prescribed fire is going to be adopted as a functional by vegetation type) are extremely variable between management technique in the southwestern border fuel types with litter and light grass combusting at lands, the effects of prescribed fire on the now the lowest average intensities. Yucca, which burned overstocked oak savannas, and the effects of fireat the highest intensity, had minimal cover on the season need to be determined. landscape and was not typically combusted to the soil surface. Whereas bear grass, which burned at COOL SEASON BURN TREATMENT medium to high intensities was typically burned Watersheds N, H, K, and C were burned inright to the soil surface.The range of burn March of 2008 fulfilling the cool -season treatment intensities measured for each vegetation type are due component of the Cascabel watershed study (Figure to the time in the combustion process when the 2). The objective of this treatment was to blackentemperature was recorded, and the wind speed and 50 -75% of the watershed surface area with adirection at that time. The values provided are a 44

Table 1. Burn intensities of various vegetation types as measured with infrared thermometer guns during the cool- season burn treatment, March 2008. AVERAGEMINIMUMMAXIMUMAVERAGEMINIMUMMAXIMUM

FUEL TEMP - °C TEMP - °C TEMP - °C TEMP - °F TEMP - °F TEMP - °F

LITTER 293.0 293.0 293.0 559.4 559.4 559.4

LIGHT GRASS 463.3 150.0 900.0 865.9 302.0 1652.0

AGAVE 481.5 382.0 581.0 898.7 719.6 1077.8

JUNIPER 586.7 216.0 858.0 1088.1 420.8 1576.4

YUCCA 620.7 225.0 997.0 1149.3 437.0 1826.6

HEAVEY GRASS 638.3 292.0 863.0 1181.0 557.6 1585.4

OAK 657.6 237.0 945.0 1215.7 458.6 1733.0

BEAR GRASS 658.6 164.0 887.0 1217.5 327.2 1628.6

MANZANITA 692.6 542.0 828.0 1278.6 1007.6 1522.4

general characterization of burn intensities ofdiscrepancy in interpreting at which temperature the different vegetation types during a cool -seasonliquification occurred. prescribed burn in Southwestern oak -savannas. Fire severity is the effect that fire has on the Ceramic tiles with temperature indicating paints environment -a characterization applied to multiple ranging from 200° F to 2000° F (93° C -1093 ° C) in ecosystem components (Neary et al. 2005, Van intervals of 100° F (38 ° C) were placed under Wagtendonk 2006).For this initial estimate of various fuel types. The full range of temperature cool- seasonburnseverityontheCascabel indicating paints were on each tile (200° F to 2000° watersheds, the severity of the burn at the soil F) and three tiles were placed beneath each fuel surface was considered.Soil burn severity was class. When the paint reachesitsindicatingmeasured under the same vegetation classes as temperature it liquefies.The fuel types /location intensity,at the location where the tiles with used were: light grass (LG), heavy grass (HG), bear temperature indicating paints were placed. This was grass (BG), oak (0), woody debris (WD), and thedone to provide a means to compare burn intensity watershed channel (C).A full range of burn and severity beneath various fuel classes. Using the temperatures were indicated by the paints fromsoil burn severity index outline by DeBano et al. <200° F (no liquification) beneath light grass, (2005) soils were classified as low, moderate or high unconsumed fuels, and in the watershed channel to severity post -burn. The descriptions of each 2000° F beneath woody debris that wasstill characterization are: low severity burn: duff largely smoldering the day after the burn. Interpreting the intact, light ground char, woody debris are partially temperature indicating paints was often obstructedconsumed, mineral soil unchanged. moderate by ash. This was not a useful determinant of burn severity burn: litter consumed, duff charred or intensity due to the fact that the fire often missed the consumed, woody debris mostly consumed (except fuels under which the tiles were placed, and the logs), gray - white ash present. high severity burn: 45

duff is completely consumed, top of mineral soil isseverity data presented here provides managers an reddish or orange, logs consumed or deeply charred expected range of fire behavior and its effects by The cool- season burn was largely low severity. fuel type. However,moderateseveritywascommonly exhibited beneath beargrass clumps, and moderate to REFERENCES high severity beneath consumed or smolderingAllen, L.S. 1995. Fire management in the sky woody- debris. islands. In: DeBano, L.F., Ffolliott, P.F., Another indicator of severity on post -burn soil Ortega- Rubio, A., Gottfried, G.J., Hamre, R.H., properties is water repellency. Soil water repellency Edminster,C.B.,technicalcoordinators. (WR) is a property the can be modified by the Biodiversity and management of the Madrean combustion of litter and organic matter, which can Archipelago: the Sky Islands of southwestern then impact the infiltration of water into the affected United States and northwestern Mexico. U.S. soil (DeBano etal.2005). Determining the ForestService,GeneralTechnicalReport occurrence of WR is important as water -repellant RM- GTR -264, pp. 386 -388. soils can result in increases in runoff and erosion. Bahre, C.J. 1991. A legacy of change: Historic WR was tested using Water -Drop Penetration Time human impacts on vegetation in the Arizona (WDPT), which is the time it takes a drop of water borderlands. University of Arizona Press, Tucson, to completely infiltrate into mineral soil. WDPT Arizona, 231 p. was extremely variable beneath allfuel types Barton, A.M. 1999. Pines versus oaks: Effects of fire ranging from instantaneous to greater than 2 min at on the composition of Madrean forests in Arizona. adjacent test locations at mineral soil.However, Forest Ecology and Management 120: 143 -156. WDPT I cm below the soil surface ranged from Brockway, D.G., Gatewood, R.G., Paris, R.B. 2002. instantaneous to 20 sec. Therefore, the impact of the Restoring fire as an ecological process in cool- seasonburntreatment was minimalto shortgrass prairie ecosystems: initial effects of non -existentonsoil -waterrepellencyacross prescribed burning during the dormant and vegetation types. growing seasons.Journal of Environmental Management 65: 135 -152. CONCLUSIONS Brown, D.E. 1982. Biotic communities of the The results summarized in this paper represent American southwest -United States and Mexico. initialintensity and severityestimatesfora Desert Plants 4: 1 -342. cool- season burn in Cascabel watersheds.The Brown D.E., and C.H. Lowe. 1980. Biotic findingsindicatethatalthoughcombustion communities of the Southwest. USDA Forest temperatures (intensity)are high and variable Service, General Technical Report RM -78. (Map) between fuel types, soil severity and soil -water DeBano, L.F., Neary, D.G., Ffolliott, P.F. 2005. repellency is generally low to moderate following a Soil physical properties. In: Neary, D.G., Ryan, cool- season prescribed burn in oak savannas of the K.C., DeBano, L.F., eds. 2005. Wildland fire in Southwest. High combustion temperatures do not ecosystems: effects of fire on soil and water. U.S. necessarily result in a high severity burn.To ForestService. General Technical Report minimize theimpacts of prescribedfire,the RMRS- GTR- 42- Vol.4, pp. 29 -51. residence time of a prescribed bum needs to be Ffolliott, P.F., Bennett, D.A. 1996. Peak Fire of reduced. 1988:Itseffect on Madrean oak trees.In: Prescribed fire is a management tool that has Ffolliott, P.F., DeBano, L.F., Baker, M.B. Jr., potential to restore "natural" species compositions Gottfried, G.J., Solis- Garza, G., Edminster, C.B., and densities. However, the physical and ecological Neary, D.G., Allen, L.S., Hamre, R.H., tech. effects of prescribed fire need to be determined to coordinators. Effects of fire on Madrean Province support environmental analyses of landscape level Ecosystems.U.S.ForestService,General fire programs. Furthermore, the burn intensity and Technical Report RM- GTR -289, pp 235 -237. 46

Ffolliott, P.F., Gottfried, G.J. 2005. Vegetative Kaib, M., Swetnam, T.W., Baisan, C.H. 1999. Fire characteristicsof oaksavannasinthe history in canyon -oak forests, intervening desert Southwestern United States: A comparative grasslands, and higher elevation mixed conifer analysis with oak woodlands in the region. In: forests ofthe southwest borderlands. In: Gottfried, Gottfried, G.J., Gebow, B.S., Eskew, L.G., G.J., Eskew, L.G., Curtin, C.G., Edminster, C.B., Edminster,C.B.,compilers.Connecting compilers. Toward integrated research,land mountain islands and desert seas: biodiversity management, and ecosystem protection in the and management of the Madrean Archipelago MalpaiBorderlands.U.S.ForestService, II. U.S.ForestService,Proceedings Proceedings, RMRS -P -10, pp. 57 -64. RMRS -P -36, pp. 399 -402. Kruse, W.H., Gottfried, G.J., Bennett, D.A., Gottfried, G.J., Ffolliott, P.F., DeBano, L.F. 1995. Mata- Manqueros H. 1996. The role of fire in Forests and woodlands of the Sky Islands: Madreanencinaloakandpinyon juniper standcharacteristicsandsilvicultural woodland development.In:Ffolliott,P.F., prescriptions. In: DeBano, L.F., Ffolliott, P.F., DeBano, L.F., Baker, M.B. Jr., Gottfried, G.J., Ortega- Rubio, A., Gottfried, G.J., Hamre, R.H., Solis- Garza, G., Edminster, C.B., Neary, D.G., Edminster,C.B.,technicalcoordinators. Allen, L.S., Hamre, R.H., Technical Coordinators. Biodiversity and management of the Madrean Effects of fire on Madrean Province ecosystems. Archipelago: the Sky Islands of southwestern U.S. Forest Service, General Technical Report United States and northwestern Mexico. U.S. RM- GTR -289, pp. 99 -106. Forest Service,General Technical ReportNeary, D.G., Ryan, K.C., DeBano, L.F., eds. 2005. RM- GTR -264, pp. 152 -164. Wildland fire in ecosystems: effects of fire on soil Gottfried, G.J., Neary, D.G., Bemis, R.J. 2001. and water. U.S. Forest Service General Technical Watershed characteristics of oak savannas in Report, RMRS- GTR -42 -Vol. 4. the southwestern borderlands. Hydrology and Ottmar, Roger D.; Vihnanek, Robert E.; Wright, Water Resources in Arizona and the Southwest Clinton S.; Seymour, Geoffrey B. 2007. Stereo 30: 21 -28. photo series for quantifying natural fuels: volume Gottfried, G.J., Neary D.G., Bemis R.J. 2002. IX: Oak/juniper types in southern Arizona and Assessing the impacts of prescribed burning on New Mexico.U.S.ForestService General soil and water resources of oak savannas in the Technical Report, PNW- GTR -714, 41 p. southwestern UnitedStates.In Assessing Pyne, S.J., Andrews, P.L., Laven, R.D. 1996. capabilities of soil and water resources in Introduction to wildland fire. John Wiley & Sons drylands: the role of information retrieval and Inc., New York. dissemination. Proceedings ofthe International Van Auken, O.W. 2000. Shrub invasions of North AridLandsConsortiumconferenceand American semi -arid grasslands. Annual Review workshop. Tucson, Arizona, pp.115 -122. of Ecological Systems 31: 197 -215. Hester, J.W., Thurow, T.L., Taylor, C.A., Jr. 1997. Van Wagtendonk , J.W. 2006. Chapter 3: Fire as a Hydrologic characteristics of vegetation types physical process. In: Sugihara et al. eds. 2006. as affected by prescribed burning. Journal of Fire in California's ecosystems. University of Range Management 50(2): 199 -204. California Press; Berkeley, California, pp. 38 -74. Webster, G.L., Bahre, C.J., eds. 2001. Changing plantlifeof laFrontera:Observations on vegetationintheUniteStates/Mexico borderlands. University of New Mexico Press, Albuquerque, New Mexico. HYDROLOGY AND EROSION IMPACTS OF MINING DERIVED COASTAL SAND DUNES, CHAÑARAL BAY, CHILE

Daniel G. Neary', and Pablo Garcia- Chevesich2

Chile has an economy strongly based on the the sand dunes with multiple row tree shelterbelts exploitation of its natural resources. Copper miningnext to the town of Chañaral. This paper examines represents the main export monetary income,the hydrologic processes which formed the sand employing thousands ofpeople all alongthe country. deposits and the potential remediation program. The Chilean Copper Corporation (CODELCO), El Salvador branch, has been the primary mining STUDY AREA company, but it will be ending most of its activities Copper Mining by 2011 unless copper prices stay at their record Chileisworld'slargest copper producer. levels. Besides the job consequences for the local Cuprous porphyry ore bodies that exist along the population, there are some serious environmentalAndean Cordillera are responsible for Chile's vast issues that must be solved during the shut -down. mineral reserves. Some of the world's largest Nearly 12 km2 of contaminated sand dunes,opencast mines are located at high altitudes and located in the Bay of Chañaral, Chile, are the result harsh cold and arid environments along the Andes of mining operations between 1938 and 1975 that Cordillera. During 2004 Chile' s copper production released contaminated sediments to the bay. Evenreached 5.3 million Mg. Other metallic minerals though the sediment release no longer occurs, themined and smelted in Chile include gold, silver, coastal winds transport the heavy metals attached to molybdenum, zinc, manganese and iron ore. the sand grains over the town of Chañaral. This area Copper mining has been important to Chile's is arid desert, with no more than 10 mm/year ofeconomic development since the arrival of the precipitation. Between 1938 and 1990, more than Spanish conquerors in the 16th century. Between the 300 million Mg of highly contaminated residual 1840s and the 1880s, Chile's share of the world's sediments were deposited in the Pacific Ocean. The copper -mining production rose to about 50 %. Its chemical analyses of the sediments have shown high share decreased to only 5% in the first half of the contents of cupper, iron, arsenic, zinc, cyanide, lead, 20`h Century, but now is expected to rise to 35-40% aluminum, mercury, molybdenum, and other heavy by 2010. Copper production in Chile reached 35% of metals. The toxic metals inside the sediments as well the world market in 2000 (Velasco 2000). During the as the arid nature of this portion of the Atacamalast decade, this activity has accounted for about Desert have affected the establishment of almost any50% of the country's exports and foreign investment, vegetation. about 5 -7% of the gross national product, but less As a consequence of the exposure to the toxic than 2% of the labor force. Copper is the main mining tailings dust produced by the coastal winds, export, followed by gold, silver, molybdenum, iron, there is a high incidence of skin, lung, and eye nitrates, iodine, and lithium. Some of the gold and problems in the local human population, as well as silver and all of the molybdenum are produced as by- a high incidence of cancer tumors. Even though products of copper mining. there have been some attempts at stabilizing the Most of the copper mines in Chile are located in contaminated sand dunes, none have succeeded. The the Andes Mountains, many in the arid to semi -arid most practical solution appears to be to stabilize ofAtacama region of the country. The three most important are: Chuquicamata at 2,680 m above sea level (22 ° 19' South Lat.; 68 ° 56' West Long.). El 'USDA Forest Service, Rocky Mountain Research Salvador at 2,600 m above sea level (26 ° 15' South Station, Flagstaff, Arizona Lat. S.; 69 ° 34' West Long.) and El Teniente at 2Santo Tomas University, Santiago, Chile, and the 2,113 m above sea level (34 ° 04' South Lat.; 70 ° University of Arizona, Tucson, Arizona. 21' West Long.). The environmental impact of metal 48 mining activities has received world wide attentionregion. It is 120 km northwest of Copiapó, and had (Castilla and Nealler, 1978). In two of the largea population 10,000 in 1996.It was once an Chilean copper mines, Chuquicamata and El important shipping center for copper produced at the Teniente, pollution controlis obtained by thePotrerillos, Las Animas, and El Salado mines, but discharge of waste waters into tailing ponds.pro -gradation of the beach with the deposition of However, throughout the history of the E 1 Salvador mine tailings ruined the harbor. mine,alluntreated mining wastes have been .The population depended on copper industry discharged through a semi -artificial canal directly torelated jobs. It declined in the 1990s with the the Pacific Ocean shore. downturn in the copper industry, but may be rebounding with surging world -wide copper prices. El Salvador Mine El Salvador is an open -pit mine, developed by HYDROLOGY block caving techniques (Camus and Dilles 2001). The Rio Salado collects precipitation from the After milling, minerals are concentrated by basicAndes and the Intermediate Depression of the flotation. A copper sulphide concentrate (CuS; Atacama Desert before flowing to the Pacific Ocean CuS2) is produced by the flotation process and (Montti 1973, International Mountain Society 1984). transferred to the molybdenum concentration plantRainfall in the Depression is only 5 mm but it for the separation of molybdenum sulphide by increaseswithaltitude,reaching33 mm at differential flotation. The latter process requires thePotrerillos (elevation 2,850 m). The Rio Salado is use of arsenic and sodium sulphide, cyanide and the only permanent river in the area with an average lime as collectors of molybdenum sulphide and discharge of 10 L /sec. In the vicinity of Llanta, depressantsof coppersulphide.The copper discharge from the El Salvador Mine is added to the concentrate is then sent through a pipe to the current river. refining plant at Potrerillos. Daily use of water in the The majority of water use in the Atacama mining and processing stages at El Salvador is 65 x Region (70 %) is from mining. Most of this water 106 L /day of water (2.5 x 10 3 L of water per ton ofresource is groundwater. The effects of mining ore processed). Water is obtained from Andeanwithdrawals are poorly understood. The total groundwater resources. Excluding the water utilizedwastewater discharge from the El Salvador Mine at the molybdenum concentration plant, 40% of the reaches 39 x 106 L/d containing an average of 25 x water is recovered and recycled. 103 Mg/d suspended solids. Tailing discharges from the copper and molybdenum concentration plants are Chañaral Bay routed through a 25 km long canal which discharges Chañaral Bay is situated in on the Pacific Coast at the locality of Llanta into the bed of Rio Salado of the Atacama Desert. This desert is a virtually (Figure1).At Llanta and Pueblo Hundido, rainless plateau along a 1,000 km strip of land on the additional recovery of copper pyrite is performed by Pacific coast of South America, west of the Andesflotation. Urban waste waters after treatment at the Mountains. The rain shadow on the leeward side of El Salvador sewage treatment plant are canalized 50- the Andes keeps this 20 million -year -old desert 50 55 km downstream before meeting, at Pueblo times drier than California's Death Valley. TheHundido, the tailing discharge stream (Rio Salado). Atacama is the second -driest desert in the world, From then on, a single stream carries both types of after the McMurdo Dry Valleys of Antarctica. It residues to the site of dumping in the coastal area of occupies181,300 km' squarekilometers Chañaral. From 1938 to 1974 tailing and waste (70,000 mie) in northern Chile, and consists ofwaters discharged directly into Chañaral beach mostly of salt basins, sand, and lava flows. Chañaral (Figure. lb). Since February 1975, a new canal Bay averages 20 m in depth, and doesn't deepencollects discharges from Rio Salado, 8 -10 km off much until about 8 km off -shore. Chañaral is the Chaharal beach and carries them to a new dumping capital of the Chañaral Department, in the Atacama site at Caleta Palito. 49

Caleta Palito EL Diversion SALVADOR MINE PACIFIC RIO SALADO,,,_4____0°.' T. OCEAN .. . 411" .' ******1 LLANTA PUEBLO CHAÑARAL o HUNDIDO Potrerillos Refining Plant 40 km

Figure 1. Chañaral Bay area of the Pacific Ocean, Rio Salado, and the El Salvador Mine, Atacama Desert, Chile. The case of the Salado river was different. This In the late 1980s, an environmental group from river received tailings from the mineral- processing Chañaral took CODELCO to court. The court in plantof theElSalvadormine, owned byCopiapó ordered the company to construct a new CODELCO. The tailings completely filled in a tailings dam. In 1989, the Supreme Court ratified the stretch of beach in Chañaral Bay about 1 km wide decision, and CODELCO was forced to build a and 4.5 km long. Minerals in the tailings resulted intailings dam, which has now entered into operation. extensive marine organism mortality (Castilla and This was an important precedent in Chilean law: Nealler, 1978). The El Salvador tailings dam haddespite the outdated environmental laws covering been filled in the original tailings dam, and the liquideffluent,companies,evenstate -owned mining company, which then belonged to a US firm, companies, could be taken to court and forced to allowed the tailings to spill into the Salado River. In deal with environmental problems. This case dealt 1975, the mining operation, which was by that time with the most serious environmental impact of owned by CODELCO, constructed a canal to divertmining on river or sea waters in Chilean history, and the river water, including tailings, to another bay, it illustrates clearly that environmental policy was Caleta Palitos, but this action produced the samenot a priority for CODELCO until very recently. deposition effect there. Overall, the mining operation dumped 330 x 106 Mg of tailings into the river and SEDIMENT DEPOSITS the canal before it was forced to change its tailings - Mine Tailings management practices in 1989. Until 1975, more Mine tailings deposited at Chañaral originate than 150 x106 Mg of untreated tailings werefrom volcanic deposits of the El Salvador -Poterillos continuously discharged, directly affecting moredistrict, located around 120 km east of Chañaral in than 20 km of coastline (Castilla, 1983; Paskoff and the Atacama desert. From 1926 to1959 the Petiot,1990). From 1976 to 1989 the discharge point Poterillos porphyry copper deposit was mined. was moved 10 km north of Chañaral Bay to Caleta Hypogene alterationledtoafeldspar-biotite- Palito (26 °16'S, 70 °41'W), a rocky cove that chlorite- quartz -ankerite- anhydritemineralization received .130 x 106 metric tons of tailings inwith chalcopyrite, pyrite, bornite, molybdenite, 13years. After the construction of an inland enargite,andsphaleriteassulfidemineral sedimentationdamin1990,sediment -free assemblage (Gustafson and Hunt 1975, Camus and wastewaters have been channeled from the dam toDilles 2001). After 1959, the El Salvador deposit Caleta Palito at a flow rate of 200 -250 L/s (Medina was exploited. The primary mineralization of the ore et al., 2004). body in the 41 million- year -old El Salvador deposit is characterized by alkali feldspar -biotite- anhydrite- 50 chalcopyrite- bornite and chalcopyrite- pyrite mineral wind velocities just above the dune surface, assemblages. Supergene enrichment formed an oreproducing even more sediment wind transport. body of 1.5 km diameter and a thickness of 200 m,Besides, with only one row, eddy turbulences are with replacement of chalcopyrite and bornite by created due pressure differentials on both sides of mostly chalcosite and covellite (Gustafson and Hunt, the barrier. This results in sediment accumulation 1975). Tailings deposited at Chañaral were from the and burial of the barrier in newly -forming sand secondary enrichment zone. The alkaline (pH 10.5) dunes. flotation process used for Cu and Mo extraction was Several approaches have been proposed to controlled by lime addition and the resulting tailings stabilize the mine tailings sand dunes. The sediments were the ones ultimately deposited in Chañaral Bay. could be easily stabilized if the wind was gradually forced to rise over permeable barriers. The strong Chañaral Bay coastal winds blowing a few meters above the sand Until 1975, mine tailings were sent into the bay dune surfaces would be less likely to transport sand at Chañaral, covering 4.5 km (Figure 2). Thisgrains. Tree shelterbelts have been proposed as a resulted in a displacement of the shoreline of 1 km mitigation measure for the beach wind erosion. and the accumulation of a 10 -15 m thick layer ofHowever, soil would need to be imported due to the tailings on top of the original beach sediments. Thecontaminated nature of the Chañaral sands, and deposition point at the outlet of the Rio Saladowater would need to be applied via an irrigation migrated northward during the deposition period to system due to the arid nature of the climate ( <10 mm the upper end of the "new" beach (Figure 2). Therainfall per year). tailings originated from the Potrerillos mine from 1926 until 1939, and thereafter from the El Salvador HEALTH IMPACTS mine until completion of the inland sedimentation Mine tailings can cause human disease via dam in 1989. Between 1975 and 1989, tailings were exposure routes of water, dust, and crops. Diseases sent to the sea at Caleta Palito, 8 km north ofcan range from direct toxicity to physical effects, Chañaral, and were exposed to the marine current. silicosis, hypothyroidism, etc. Chemical analyses of The tailings deposition through the natural Rio the Chañaral Bay sediments have shown the Salado (the salty river) ceased in 1989 because ofexistence of high contents of copper, iron, arsenic, the court decision. Today, the tailings are deposited zinc,cyanide,lead,aluminum,mercury, close to the "El Salvador" mine in the "Pampamolybdenum, and other heavy metals. As a Austral" tailings impoundment. Only the coarseconsequence of the exposure to the toxic dust sediment free waters of the settling pond at theproduced by the coastal winds, there is a high Pampa Austral impoundment are mixed with highly incidence of skin, lung, and eye problems, as well as mineralized waters in the Rio Salado riverbed close a variety of cancer tumors among Chañaral's to the El Salado village. These waters are still population (Leon 2001). Concerns about perchlorate discharged to the sea at Caleta Palito. in Chañaral's drinking water affecting thyroid function of children were never validated (Crump et Wind Erosion al. 2000). There are about 12 km2 of tailings-contaminated sand dunes, located on the recent Chañaral beach, as SUMMARY well as the original pre -1938 beach. Even though Nearly 12 km 2 of contaminated sand dunes, new sediment release no longer occurs, the coastallocated in the Bay of Chañaral, Chile, are the result winds transport the heavy metals laden sand grainsof mining operations at the El Salvador Mine over the town of Chañaral. between 1938 and 1975 that released contaminated A single, permeable row of concrete barriers was sediments to the bay via the Rio Salado. About installed next to houses and along the beach with the 150 x 106 Mg of untreated tailings were deposited in intention of "detaining" blowing sand. However, the Chañaral Bay before a diversion was put in place aerodynamic characteristics of the barrier increases 51

PACIFIC OCEAN

BE AC H LINES

C HAÑARAL, CHILE, AN D C OAST AL C OPPER MIN IN G TAILINGS D EPOSITS

Figure 2. Chañaral Bay, Chile (A), showing beach deposits and progradation (B) from 1938 to 1989 (Adapted from Wisskirchen and Dold 2006) Chañaral Bay before a diversion was put in place REFERENCES CITED to Caleta Palito. Another 180 x 106 Mg of tailings Castilla, J.C. 1983. Environmental impacts in was deposited in the Pacific Ocean before a large sandy beaches of copper mine tailing at tailings dam was constructed at the mine. The total Chañaral, Chile, Marine Pollution Bulletin wastewater discharge from the El Salvador Mine 14, pp. 159 -464. reaches 39 x 106 L /d, resulting in a flow in the Rio Castilla, J.C.; Nealler, E. 1978. Marine Salado of 200 -250 L /s. Even though the sediment environmental impact due to mining release no longer occurs because of the diversion activities of El Salvador copper mine, north of Chañaral Bay, the coastal winds transport Chile. Marine Pollution Bulletin 9: 67 -70. the heavy metals attached to sand grains over the Camus, F.; Dilles, J.H. 2001. A Special Issue town of Chañaral. The source of these sands is 4.5 Devoted to Porphyry Copper Deposits of km2 of beach formed since1938.Biological Northern Chile. Economic Geology, stabilization of the sand dunes is complicated by the March 1, 2001; 96(2): 233 -237. arid climate and toxicity of the sands. 52

Crump, C.; Michaud, P.; Tellez, R.; Reyes, C.; Montii,S.1973. Estudio hidrogeologico de la Gonzalez, G.; Montgomery, E.L.; Crump, cuencadelRioSalado.Provincia de K.S.; Lobo, G.; Becerra, C.; Gibbs, J.P. Atacama. Memoria de Prueba. Universidad 2000. Does perchlorate in drinking water de Chile, Santiago. affect thyroid function in newborns or Paskoff, R.; Petiot, R. 1990. Coastal progradation as school -age children? Journalof a by- product of human activity: an example Occupational and Environmental Medicine. from Chañaral Bay, Atacama Desert, Chile, 42(6):603 -612. Journal of Coastal Research 6: 91 -102. Gustapfson, L.B.; Hunt, J.P. 1975. The porphyry Velasco, P. 2000. The mineral industry of Chile. copper deposit at El Salvador, Chile. http: / /minerals.usgs.gov /minerals /pubs /cou Economic Geology 70: 855 -912. ntry/2000 /9506000.pdf. International Mountain Society. 1984. Mining: Warhurst, A. 1998. Mining and the Environment: State of knowledge report on Andean CaseStudiesfromtheAmericas. ecosystems. Pp. 175 -179. In:Mountain International Development Research Center, Research and Development, Vol. 4, No. 2, Ottawa, Canada. 300 p. The Southern Andes and Sierra Pampeanas. Wisskirchen, C.; Dold, B. 2006. The marine shore Lagos, G.; Velasco, P. 1998. Chapter 3: porphyry copper mine tailings deposit at Environmental policies and practices in Chañaral, northern Chile.Pp. 2480 -2489. Chilean mining. Pp. 101- 136.In: Mining and Proceedings of the 7 `h International the Environment: Case Studies from the ConferenceonAcidRockDrainage Americas.InternationalDevelopment (ICARD), March 26 -30, 2006, St. Louis Research Center, Ottawa, Canada. 300 p. MO. Barnhisel, R.I. (ed.) Published by the Leon,I.2001. Contaminación de la bahía de AmericanSocietyofMiningand Chañaral porelmineral ElSalvador. Reclamation, Lexington, KY. Universidad de Santiago. 19 p. Medina, M.; Andrade, S.; Faugeron, S.; Lagos, N.; Mella, D.; Correa, J.A. 2005. Biodiversity of rocky intertidal benthic communities associatedwithcopperminetailing dischargesinnorthernChile.Marine Pollution Bulletin 50: 369 -409. CLIMATE CHANGE IMPACTS ON MUNICIPAL, MINING, AND AGRICULTURAL WATER SUPPLIES IN CHILE

Daniel G. Neary`, and Pablo Garcia-Chevesich2

Agricultural and municipal water supply interests in Mediterranean Central, and cool and wet South Chile rely heavily on streams which flow from the zones. Chile is home to one of the driest deserts in Andes Mountains. The highly productive Copiapó the world, the Atacama Desert, where rainfall agricultural region, on the southern edge of the averages less than 1 mm/year, and one of the wettest Atacama Desert, is a major supplier of fruit and regions on the planet, Isla Chiloé ( >4,000 mm/year) other crops for the Northern American market during (Table 1). The climate is characterized by a wet winter. This region relies entirely on snow and ice - winter (May toAugust) and adry summer melt streams to provide irrigation water. Santiago, (November to March). Mean annual temperatures theChileancapitol,isthecountry'smajor range from 18° C at Arica in the North to 6° C at metropolitan area with a population of 5.5 million. Punta Arenas in the South. Maximums can reach 46° Rainfall that averages 330 mm/year is nearly twice C in the Atacama Desert. that of Phoenix. Santiago is very similar to Phoenix in a number HYDROGEOGRAPHY of ways. It has a rapidly expanding population of 5.5 Water Resources million, and it relies on water supplies derived from Chile's total land area of 756,950 km' is about surrounding or more distant mountain ranges. 1% freshwater, amounting to a total volume of 922 Santiago and Phoenix are located at similar latitudes km3. Of Chiles's total renewable water resource, north and south of the equator (33 °27' South for the groundwater accounts for 15 %, and surface water former and 33° 31' North for the latter). Recent 85% (FAO 2001). About 15% is shared between the changes in the climate in South America aretwo resources. There are no flows into Chile from resultingindecreased snowpacks and glacierother countries and the only outflows are to the volumes in the Andes Mountains. This paper Pacific Ocean. Groundwater withdrawals total 20.3 discusses the current water supply situation in Chile km3 annually, or about 15% of the actual annual in light of its growing demand for water and declines groundwater recharge (140 km 3). Most of these in supply due to climate change. withdrawals are for mining in the North and agriculture in Central Chile. These figures don't tell CLIMATE AND ECOSYSTEMS the whole story since the groundwater withdrawals Chile has one of the widest ranges of climates are in the arid North with low recharge rates. Surface and ecosystems in the world because of its north - waterwithdrawalsarecurrentlymainlyfor ° to 56° south latitude range (4,300 km from 17 agriculture irrigation (64 %). Although domestic uses South Latitude) and large altitudinal of total surface water is low (11%), the presence of gradient (0 to 6,880 m) in a short distance from the most of the population in the Mediterranean climate Pacific Ocean ( <240 km ocean to Andes Mountains of the Central zone places a lot of stress on local crest). It contains arid deserts, semi -arid deserts, water resources. Mediterranean -like areas, humid subtropical regions, Water is a scarce commodity from the Rio Bio- temperate zones, oceanic -dominated climates, sub- Bio northward (Wollman 1968). This is virtually the polar areas, alpine tundra, and ice caps. Thesenorthern half of the country, including the major climates are often combined into the arid North, themetropolitan areas of Santiago, Valparaiso, and Concepcion. Water supplies in the Central and South `USDA Forest Service, Rocky Mountain Research zonesarepredominantlysurfaceinnature, Station, Flagstaff, Arizona originating in snow and glacier melt in the Andes. In 'Santo Tomas University, Santiago, Chile, and thethe North, high evaporation rates of the Atacama University of Arizona, Tucson, Arizona. Desert consume most rainfall and the limited surface 54

Table 1. Average annual precipitation by major city and region, Chile (CIA 2008, FAO 2001).

CITY AVERAGE ANNUALLOCATIONREGION PRECIPITATION mm Arica <1 North Arica Copiapó 21 North Atacama Santiago 330 Central Santiago Concepcion 1320 Central Bio -Bio Valdivia 2535 South Los Rios Puerto Aisén 2973 South Aisén Chiloé National Park >4000 South Los Lagos waters, leaving dry riverbeds or large salt playas. Central and South regions. For example, El Yeso is Groundwater is the most important water source in a reservoir located in the Andes formed by damming this region. the Rio Yeso, a tributary of the Rio Maipo. The reservoir is located in tandem with Laguna Negra at Rivers and Lakes an altitude above 2,600 m. It provides a storage Rivers in Chile are relatively short, flowing from capacity of 255 x 106 m 3, or about 46.3 m3 /person the Andes crest westward to the Pacific Ocean. Few for the Santiago Metropolitan Region. rivers in the North flow to the sea because of high WATER ISSUES AND USES evaporation rates and much diminished precipitation Land use (Table 2). Their main source is Andean snowfall and Most of the issues related to water center around rain. In the North, the Rio Loa is the only river land use and water availability. The major issues between Arica on the northern border with Peru and deal with population concentration in the Santiago the Rio Copiapó (960 km) that flows to the sea. It Metropolitan area (33% of Chile's population), reaches the Pacific with very diminished flows agriculture,mining,damconstructionfor (mean flow 2.4 m3 /sec) despite having a large hydroelectric generation, and climate change. The drainage basin (33,570 km2) and reasonable length latterinteracts with the other land uses and (440 km). South of the Rio Copiapó and the Rio activities. Currently land use in Chile is 3% in arable Huasco in the Atacama Region, the number of rivers crops, 16% in meadows and pastures, 21% in forests that flow continually to the sea increase dramatically and woodlands, and 60% in other categories, mostly (e.g. Elqui, Limari, Maipo, Rapel, Mataquito, Maule, desert and high mountain lands (CIA 2008). The Itata, Bio -Bio, Imperial, Tolten, Palena, Baker etc.). great disparity in water resource abundance between Rivers such as the BioBío are being developed withthe arid North and wet South also aggravates the dams for hydroelectric and water supply purposes. internal water resource situation. Mining in the Most of the lakes in Chile are in the Araucania North and population concentration in the Santiago Region and South (Table 2). There are 17 lakes with Region, although separated by great distances from areas >100 km2. The largest is General CarreraLake the water rich South are affecting decisions on how covering 1,850 km2 with half in Chile and the to use those water resources. remainder in Argentina. Two are in the 500 to 1,000 'Ulf size class, five are 200 to 500 km2 in size, and Agriculture the remainder 100 to 200 km2 category (Wollman Agriculture is a key part of the economic mix of 1968). minerals, food commodities wood products, and There are a number of water supply and power fishing that has been the base of the Chilean generation reservoirs throughout Chile, mainly in the economy. Agriculture accounts for only 5% of the 55

Table 2. Main rivers and lakes by Region of Chile (Wolluran 1968, CIA 2008, FAO 2001).

REGION RIVERSLAKES Arica and Parinacota 4 1 Tarapacá 0 1 Antofagasta 4 3 Atacama 7 1 Coquimbo 6 1 Santiago 6 1 Valparaiso 7 2 O'Higgins 3 1 Maule 20 0 Bio -Bio 22 1 Araucania 26 7 Los Rios 25 7 Los Lagos 10 11 Aisén 10 3 Magallanes 11 10 country Gross Domestic Product but employs 14% constantly supplemented (CIA 2008). For awhile of the national labor force (CIA 2008). Chilean crop Argentine natural gas was supplementing electric products are becoming increasingly important power production until rising prices and a restriction internationally because of their availability during on exporting Argentine natural gas forced a switch northern hemisphere winter and freedom diseases to diesel fuel for thermoelectric power plants, raising affecting other nations. Chile produces 3.9 x 106 Mg costs four -fold in 2008 (Reuters 2008). Chile of fruit, 1.2 x 106 Mg of wheat, 0.5 x 106 Mg ofcurrently supplies about 60% of its electrical energy sugar, and various amounts of vegetables, beef, from water (Business News America 2008). Plans poultry, and wool. However, 96% of the agricultural have been developed to build five hydroelectric lands are irrigated from surface and groundwater dams on the Rios Pascua, Baker, Del Salto, and Bio- supplies ( >12,650 km 2). Water flowing out of the Bio at a cost of $2.4 billion to provide an additional Andes Mountains is critical for Chilean agriculture. 2,430 Mw of power production capacity. New Thus the recent droughts related to climate change transmission lines will be needed to move the are a big issue since increasingly scarce surfacehydroelectric energy over 3,000 km north to the waters have to be reallocated amongst competing population and mining centers. However, the scale economic sectors, including human water supply of this project along with dam placement on demands. ecologically important rivers in the South has created considerable internal political dissent (New Hydroelectric Power York Times 2008). Nevertheless, energy demands Chile's electric power generation capacity is keep rising and unless conservation measures reduce now highly dependent on water resources in the demand, Chile will have to add 300 Mw. South. The national energy grid (SIC) has been fragile because of its dependence on water resources. Mining Chile has not been able to keep up with energy Mining has a major effect on the water resources demands from the public and mining sectors so that of Chile because large quantities are needed for ore its generation capacity of 48 x 109 kWh has to be processing. In addition, smelting is a big consumer 56

of electrical energy. For instance, daily use of water recover normal levels and that means (crop) in the mining and processing stages at El Salvador is irrigation in 2008 -2009 will be difficult." Rainfall 65 x 103 m3 /day (2.5 m3 of water per ton of ore deficits have been in the 35 to 100% range from processed). The source of this water is groundwater Arica in the North all the way south into the Bío -Bío since the mine is in the Atacama Desert. and Aisén regions of the far South. Shallow wells supplying many small towns and farms have dried CLIMATE CHANGE EFFECTS up. The effect of the El Niño /Southern Oscillation Agricultural production in Chile this year has on Chilean weather has been well documentedbeen seriously affected by both the shortage of water ( Nuñez 1992; Haylock et al. 2006). Long -term trends and the high cost of pumping groundwater for investigated by Minetti (1998) documented a steady irrigation (Vargas 2008). Fresh produce production decrease in annual rainfall for a large area west ofhas declined significantly and animal forage is in the Andes Mountains under La Niña conditions. short supply. Drying up of shallow groundwater Rusticucci and Penalba (2000) described a large wells (6 - 10 m) has resulted in significant animal decrease in total annual precipitation at Valdivia in mortality (Martinez 2008). Fruit growers have not the Los Rios Region (Table 1) due to a decrease inbeen impacted yet due to their deep wells ( >50 m). winter precipitation. Winter snowpacks observed byOver 144 municipalities have declared agricultural the authors in mid -August, 2007, at elevations above emergencies because of the current drought (Estrada 2,000 m were shockingly low. North- facing areas at2008). There are indications that the current year- >3,000 m that should have been entirely snow - long La Niña droughts may become due to the covered in August were devoid of snow. These Pacific Decadal Oscillation (Mantua and Hare trends have been evident in extreme southern Chile 2002). So, agricultural impacts could be much where glacial melt rates are some of the fastest in the greater in years to come. world and account for nearly 10% of global sea level Agriculture has already been impacted by power change from mountain glacier melt. Some Andean costs and shortages (Vargas et al. 2008). President glaciers are expected to disappear in 15 -25 years Bachelet ordered voltage reductions during the (Vidal 2006). The effects will be major, reducing Chilean summer of 2008 and extended daylight municipal water supplies and agricultural irrigation savings time. Also ominous is the potential threat to sources. However many effects are occurring now, mining, a mainstay of the Chilean economy and a not in the future. The current La Niña Began in May, large consumer of power. In March, hydroelectric 2007, and is expected to end in August, 2008 reservoirs were dangerously low. Energy supplies (Estrada 2008). It has been atypical in that areaswill remain uncertain until 2010 when more such as Araucania, which normally get heavier hydroelectric production comes on line from current rainfall in a La Niña, have been experiencingdam and power line construction in the southern drought as well. areas of Chile. At a time when prices for minerals In March, 2008, Chile's Public Works Ministry such as copper are rising due to world -wide demand, Undersecretary J.E. Saldivia described the currentprices could rise much higher if Chilean production drought as the worst in 100 years, threatening wateris significantly reduced by the lack of hydroelectric supplies for over 200,000 people (Vargas et al, power. 2008). He stated that the drought was intensifying, and that even if rainfall improves this year, the SUMMARY resultinghydroelectricenergyshortagecould Chile has a tremendous diversity of water become critical. Chile's major hydroelectric reservoir resources as a result of its geographical position in levels are far below their historic levels, and will South America. It contains both the driest desert in need at least one year to rise to normal levels even ifthe world and one of the wettest regions. The this year's rainfall is normal. Mr. Saldivia was country is highly dependent on Andes Mountains quoted as saying that "reservoir levels will notannual snowpacks and glaciers for water used for 57 municipal, agricultural and mining activities. In Haylock,M.R.;Peterson,T.C.;Alves,L.M.; addition, a large portion of the Chilean energy Ambrizzi, T.; Annunciacáo, Y.M.T.; Baez, market so important for cities, agriculture, and J.; Barros, V.R.; Berlato, M.A.; Bidegain, mining comes from hydroelectric sources that are M.; Coronel, G.; Corradi, V.; Garcia, V.J.; also subject to the oscillations in climate. The Grimm, A.M.; Karoly, D; Marengo, J.A.; current La Niña is abnormally dry across the entire Marino, M.B.; Moncunill, D.F.; Nechet, D.; 4,300 km of Chile. This ENSO event may be a Quintana, J.; Rrebello, E.; Rusticci, M.; harbinger of droughts to come as global climate Santos, J.L.; Trebejo, L. Vincent, L.A. changebecomesmorepervasive.Recent 2006. Trends in total and extreme South understanding of the Pacific Decadal Oscillation American rainfall in 1960 -2000 and links (Mantua and Hare 2002) indicates that future with sea surface temperature. Journal of droughts could be on the order of decades -long Climate 19:1490 -1512. durations rather than 1- to 2- years. Impacts on the Mantua, N.J.; Hare, S.R. 2002. The Pacific Decadal major water -using sectors of the Chilean and global Oscillation, Journal of Oceanography 58: economy could be significant. 35-44. Martinez, R. 2008. Chile government hands out REFERENCES water in major drought. Reuters, February Business News America 2008. Chile's power 22, 2008. crunch: Present crisis and future solutions in Minetti, J.L. 1998. Trends and jumps in the annúal the central grid. Business News Americas, precipitation in South America, south of 15 Energy Intelligence Series. March 2008. degrees S. Atmosphera 11:205 -221. http: / /www.bnamericas.com/store /product National Foreign Assessment Center CIA 2008 The s. j sp ?idioma =l §or = l 0 &sku =71 I 10497 World Factbook, Directorate of Intelligence, 394 &periodo =2008. CIA,GovernmentPrintingOffice, Central Intellligence Agency (CIA). 2008. The 2008 Washington, DC. World Factbook. U.S. State Department, New York Times. 2008. Patagonia without dams. Washington, D . C . Editorialpublished04/01/2008. https: / /www.cia. gov /library/publications /t http://nytimes.com/2008/04/01/opinion/. he- world- factbook/index.html. Nuñez, R.H.; O'Brien, J.J.; Striver, J.F. 1992. The César N. Caviedes, C.N. 20005.Contemporary effect of ENSO on rainfall in Chile (1964- geography in Chile: A story of development 1990). Tropical Ocean -Global Atmosphere andcontradictions.TheProfessional ProgramNotes.U.S.Departmentof Geographer Pp. 359 -362. Published Online: Commerce, National Oceanographic and Feb23 2005 12:00AMDOI: Atmospheric Administration, Washington, 10.1111/j.0033-0124.1991.00359.x. D.C. Pp. 4 -7. Estrada, D. 2008. La Niña and climate chaos. Inter. Vargas, M.; Yulkowski, L.; Picinich, J. 2008. Chile Press Service News. March 13,2008. says drought worsening, energy shortage http://www.ipsnews.net/news.asp?idnews critical. Reuters. Friday March 7, 2008. = 41581. http://www.reuters.com/article/latestCrisis Food and Agriculture Organization of the United /idUSN07487682. Nations (FAO). 2001. Statistics on Water Rusticucci, M.; Penalba, O. 2000. Interdecadal ResourcesbyCountryinFAO's changes in the precipitation seasonal cycle AQUASTAT Programme, Water Resources, over Southern South America and their Development and Management Service. relationshipwithsurfacetemperature. October,2001.availableon -lineat Climate Research 16:1 -15. http://www.fao.org/ag/agl/aglw/aquastat/w ater res /index.stm). Rome: FAO.