CC BY
14
——=——— СИСТЕМНОЕ ИЗУЧЕНИЕ АРИДНЫХ ТЕРРИТОРИЙ —
УДК 551.451.8:912(57)
ДРЕВЕСНАЯ РАСТИТЕЛЬНОСТЬ ЭКОТОНА НА РЕКЕ БИЛЛ ВИЛЬЯМС, ЮГО-ЗАПАД США
© 2006 г. П.Б. Шафрот
Геологическая служба США, Научный Центр, Колорадо, 215G center Avenue, Building C, Fort Collins, CO 8G526-8118
В бесплодных полузасушливых и засушливых гористых районах на западе Северной Америки вдоль рек формируются уникальные экотонные ландшафты вода-суша. которые используются в различных целях: Здесь создаются особо благоприятные условия среды для обитания живых организмов, вдоль русел идет миграция животных. Пойменные земли используются в сельском хозяйстве под пашню и для выпаса домашнего скота (Brinson et al., 1981; Naiman et al., 2GG5). Эти ландшафты были известны как содержащие богатую флору (Wolden et al., 19944; Makings, 2GG3), и отмечена их способность поддержать существование более 9G% из позвоночных региона (Knopf et al., 1988). Как в других частях мира, деятельность человека изменила условия этих ландшафтов на обширных территориях. Таким образом, комбинация их уникальных ценностей и сильный антропогенный пресс сделали сохранение прибрежных экосистем высоким приоритетом для менеджеров, управляющих природными ресурсами на западе Северной Америки и во всем мире (Naiman et al., 2GG5).
В Северной Америке, экотоны вода-суша, связанные с реками упоминаются больше всего обычно как «прибрежные» экосистемы. Термин «прибрежный» имеет множество значений (Verry et al., 2GG4). В данной статье он используется а качестве синонима экотона вода-суша, и в качестве прямого значения «прибрежный = bottomland», расположенный вблизи русла. При этом подразумевается, что он включает полный набор растительности, условий среды обитания, и другой биоты, приуроченной к различным флювиальным формам рельефа, которые были в значительной степени сформированы динамикой реки в Голоцене и связаны с водными аллювиальными отложениями. Дифференциация условий среды идет через основные формы рельефа - речные русла, бары, прирусловые валы поймы и террасы. В последнем случае растительность может напоминать зональную растительность прилегающих гор. В бесплодных полузасушливых и засушливых регионах экотоны, обычно имеют резкую границу между приречной растительностью и зональной растительностью.
Зональная растительность территорий, которые пересекает река Билл Виллиамс, описаны в работах Turner (1994a; 1994b). Она представлена кустарниковыми сообществами жарких пустынь умеренного пояса (тип растительности пустыни Мохаве) в верхнем течении. В нижнем течении зональными являются кустарниковые сообщества субтропических пустынь (тип растительности пустыни Сонора). Доминирующие древесные растения имеют очень мелкие листья. Это Prosopis glandulosa, Olneya tesota и Parkinsonia microphyllum; кустарники: Larrea tridentata, Амброзия dumosa, Encelia farinosa, Lycium andersonii, Atriplex spp., Fouquieria splendens, Аcacia gregii и кактусы Carnegiea gigantea, и Opuntia spp.
Растительность пойменного экотона реки Билл Виллиамс представлена несколькими доминирующими древесными видами: Populus fremontii, Salix gooddingii, Tamarix ramosissima, T. chinensis и их гибриды (Gaskin, Schaal, 2GG2), Baccharis salicifolia, и Prosopis
spp. Травяная растительность имеет низкое покрытие в древесных сообществах, но оно резко возрастает на открытых участках вдоль каналов.
В статье поставлена задача дать краткий обзор основных прибрежных сообществ растительности и их пространственного распределения в пределах речных пойм в пустынях Сонора/Мохаве на западе Северной Америки, на реке Билл Виллиамс (рис. 1). Кроме того, обсуждено, как динамика растительности связана с различными аспектами водного режима и его антропогенными изменениями. Предыдущие публикации, материалом для которых также послужили данные, собранные на этой реке, содержат более детальные описания географии, гидрологии, флоры, и фауны (Defining ..., 2006); воздействие регулирования стока на прибрежную растительность реки Билл Виллиамс (Shafroth, 1999; Shafroth et al., 2000; Shafroth et al., 2002); в верхнем бьефе и воздействие зарегулированного речного (или "экологического стока") в нижнем бьефе, на естественную растительность, позвоночных и беспозвоночных животных (Defining ..., 2006; Shafroth et al., 1998).
Ежегодный объем стока на реке Билл Виллиамс между 1941 и 1996 гг. (рис. 2) составлял в среднем приблизительно 1.1*108 м3, изменяясь от 1.8*106 м3 до 8.5*108 м3 (датчик 0942600). После строительства дамбы в районе Аламо в 1968, произошло сокращение пиковых расходов. В течение 28 лет до создания дамбы медианы, средние, и максимальные ежегодные пиковые расходы воды были 416, 223, и 1843 м2/с соответственно, против 44, 18, и 198 м2/с в течение 28 лет после строительства дамбы. В период до зарегулирования реки наводнения обычно происходили в весенние месяцы и в конце лета и осени. Эти более поздние наводнения были фактически устранены регулированием стока. Низкие расходы были изменены также. Они теперь намного выше чем в естественных условиях, так же как периоды с нулевым стоком, который никогда не документировался в естественном режиме.
Первый градиент, структурирующий экотонную систему «вода-суша» и растительность на реках - высотные отметки (положение в рельефе), которые корелируются с такими факторами, как частота и длительность паводкового заливания и глубина грунтовых вод. Экспериментальные наблюдения на реке Билл Вильямс позволили рассчитать частоту встречаемости 19 древесных видов на 227 выбранных участках на реках Санта Марии и Билл Виллиамс (рис. 3), располагающихся вкрест рельефа. Наиболее часто встречающимся (40%) видами оказались тамариксы, имеющие широкую экологию и стречающиеся почти на всех пойменных биотопах. Виды рода Lycium, Hymenoclea monogyra, Populus fremontii, Baccharis salicifolia, Salix gooddingii, Pluchea sericea встречаются с частотой от 10 до 20%. Эти виды иимеют разную экологию и каждый приурочен обычно к своему биотопу. Виды Ambrosia ambrosioides, Parkinsonia microphylla, Nicotiana glauca, Suaeda torreyana, Baccharis sarothroides, Prosopis pubescens, Salix exigua, виды рода Atriplex, Larrea tridentata, ACGR -Acacia gregii, Ziziphus obtusifolia имеют низкую встречаемость - до 5%.
Для того, чтобы понять приуроченность видов к различным участкам экотона, территория поймы была разбита на 13 типов по характеру гипсометрического положения и представленной растительности. Следующим шагом было выявление распределения 13 различных участков с растительностью как функции их положения в рельефе и отношения к урезу воды (рис. 4). Группировки, соответствующие низкому топографическому положению (берега русел) заняты гидрофильной и мезофильной травяной растительностью (Typha spp., Scirpus spp., Polygonum spp., Leptochloa spp.), или проростками древесных видов. Более высокие поверхности на низкой пойме заняты молодыми экземплярами Salix gooddingii и Populus fremontii, и также Baccharis salicifolia. Самые высокие высотные отметки заняты Pluchea sericea и видами рода Prosopis.
Периодические пожары также играют важную роль в формировании современной структуры растительных сообществ экотонной системы побережий на реке Билл Виллиамс.
В результате пожаров тополя выгорают полностью и не восстанавливаются, тогда как Tamarix, Pluchea и в меньшей степени Salix способны повторно вырасти. Tamarix способствует возникновению пожаров из-за большого количества мертвых стеблей. Регулирование стока косвенно способствует пожарам в экотонных экосистемах, потому что без наводнений имеет тенденцию накапливаться отмершая растительность, которая транспортируется водой. На террасах, пожар способствует смене Prosopis spp. ксерофильным кустарником Pluchea sericea.
Река Билл Виллиамс - превосходная "естественная лаборатория" для изучения динамики прибрежной растительности в контексте управления режимом речного стока для экологических целей. Это важная задача управления водными ресурсами для экологических целей, т.к. позволяет в определенной степени способствовать сокращению площадей занятых тамариксами. Все виды тамариксов в США - вселенцы из Азии и Австралии. Они распространились по рекам на обширных территориях, заместив собою естественную растительность. А поскольку они не находят хозяйственного применения у местного населения, с ними ведется борьба, которая отнимает много денег и сил у государства. Сильные дожди зимой 2005 позволили регулировать попуски из водохранилища для того, чтобы способствовать разрастанию Populus и Salix в биотопах, занятых Tamarix. Паводковые расходы в феврале и марте составили 195 м2/с, с постепенным спадом в остальную часть весны и лета. Предварительные оценки растительности показывают, что этот попуск был очень успешен для внедрения древесных видов естественной флоры. Работы по контролю ответа растительности на этот попуск продолжаются. Кроме того, ведутся работы, для того, чтобы лучше понять не только динамику древесной растительности, но также и отклики травяной растительности на режим попусков. В настоящее время объединяются усилия ученых для того, чтобы вести наблюдения за различными группами живых организмов и режимом речного стока.
WOODY ECOTONAL VEGETATION OF THE BILL WILLIAMS RIVER,
SOUTHWESTERN USA
© 2006. P. B. Shafroth
United States Geological Survey Colorado, USA, 2150 center Avenue, Building C Fort Collins CO 80526-8118
In semiarid and arid regions of western North America, aquatic-terrestrial ecotones include unique landscape elements that provide an array of goods and services including provision of wildlife habitat, preferential sites for agriculture, recreation, transportation corridors, and livestock grazing (Brinson et al., 1981; Naiman et al., 2005). These systems have been documented to contain rich floras (Wolden et al., 1994; Makings, 2003), and can support >90% of some vertebrate wildlife groups at some point in their life cycle (Knopf et al., 1988). As in other parts of the world, anthropogenic demands have altered these habitats in many cases. Thus, the combination of their unique values and heavy anthropogenic use has made the conservation of riparian ecosystems a high priority for resource managers in western North America and worldwide (Naiman et al., 2005).
In North America, aquatic-terrestrial ecotones associated with rivers are most commonly referred to as riparian ecosystems. Although riparian has been defined in many ways (Verry et al., 2004), for this paper, I will treat the terms riparian, aquatic-terrestrial ecotone, and bottomland as essentially equivalent. I consider them to include the full suite of vegetation types, physical habitats, and other biota that occur on geomorphic surfaces that have been largely shaped by holocene river dynamics and are connected to the alluvial water table. These surfaces include river channels, bars, floodplains and terraces, and the biotic communities that they support. Unlike more humid regions where floodplain vegetation may resemble the upland (zonal) vegetation, in semiarid and arid region
ecotones, there is usually a distinct boundary between the bottomland vegetation and the upland vegetation.
In this paper, I aim to provide readers from Eurasia with an overview of the principal riparian vegetation communities within a river in the Sonoran/Mojave desert of western North America, the Bill Williams River, including a discussion of how vegetation dynamics is related to various aspects of the streamflow regime and its management. Previous studies on this river contain more detailed descriptions of geography, hydrology, flora, and fauna (Defining ..., 2006); effects of flow regulation on the riparian vegetation of the Bill Williams River (Shafroth, 1999; Shafroth et al., 2000; 2002); and effects of managed streamflow (or "environmental flows") downstream of the dam to favor native vegetation, vertebrates, and invertebrates (Defining ..., 2006; Shafroth et al., 1998).
Рис. 1. Карта-схема бассейна реки Билл Виллиамс, штат Аризона, США. Fig. 1. Map of the Bill Williams River, state of Arizona, USA.
STUDY AREA DESCRIPTION
The Bill Williams River drains more than 13000 km2 of rugged, mountainous terrain in the west-central part of the US state of Arizona. The Bill Williams River begins at the confluence of the Big Sandy and Santa Maria rivers and flows 65 km to the Colorado River confluence at Lake Havasu (Fig. 1). The upper 6.5 km are now impounded behind Alamo Dam, a flood control structure that was completed in 1968 and has a reservoir storage capacity of approximately 1233 m3 x 106. Downstream of the dam, the Bill Williams R. flows 58.5 km with an average gradient of 0.003 (range of 0.001-0.009) to its confluence with the Colorado River (Lake Havasu) at an elevation of 137 m (Fig. 1). The Bill Williams R. passes through canyons interspersed with alluvial valleys. No perennial tributaries enter the Bill Williams R. downstream of Alamo Dam. Channel bed and flood-plain sediments are dominated by coarse particles (81%), primarily sand (67%), and are generally low in electrical conductivity (ca. 1.0 dS/m) and nutrient content (Shafroth, 1999). Climatically, the watershed lies within a transition area between the Sonoran and Mojave deserts. Average annual precipitation in the watershed ranges from approximately 45 cm in the headwaters to 23 cm near Alamo Dam (National Climatic Data Center station Alamo Dam) to 12.2 cm near the Colorado River (National Climatic Data Center station Parker 6NE). Precipitation falls principally from frontal winter rain events, convectional monsoon rain which falls in late summer and early fall, and occasional tropical storms. Land along the Bill Williams River is primarily owned and managed by the United States Government, mostly as "wildlife refuge" and "wilderness." Anthropogenic use is minimal along the Bill Williams R. corridor, but includes small areas of agriculture, cattle grazing, and four-wheel drive vehicle use.
The largest impact to the Bill Williams River is flow regulation from Alamo Dam (Fig. 2). Between 1941 and 1996, annual flow volumes averaged approximately 1.1*108m3 and ranged from 1.8*106m3 to 8.5*108m3 (U.S. Geological Survey Gage #0942600). Several flow attributes have changed since dam closure in 1968, including a dramatic reduction in peak flows. During the 28 years of reliable pre-dam data (1941-1968), median, mean, and maximum annual peak flows were 416, 223, and 1843 m2/s respectively, versus 44, 18, and 198 m2/s during 28 post-dam years (19691996). In the pre-dam era, floods commonly occurred in spring months as well as in late summer and fall. These later floods have been virtually eliminated in the post-dam era. Low flows have been altered as well, with post-dam releases including periods of low flow that are much higher than pre-dam conditions, as well as extended periods with zero flow, which was never documented in the pre-dam record. Total annual discharge is minimally altered by evaporation off of the Alamo Lake surface and occasional inter-year storage, but water is not diverted from the reservoir or river.
DESCRIPTION OF VEGETATION
The upland, zonal vegetation which borders the riparian zone along the Bill Williams River consists of plants typical of the warm-temperate desert Mohave Desertscrub and the Lower Colorado River Valley subdivision of the subtropical desert, Sonoran Desertscrub biomes, described by Turner (1994a; 1994b). The dominant plants are woody, and include small microphyllous trees such as Prosopis glandulosa, Olneya tesota, and Parkinsonia microphyllum; shrubs such as Larrea tridentata, Ambrosia dumosa, Encelia farinosa, Lycium andersonii, Atriplex spp., Fouquieria splendens, Acacia gregii; and cacti such as Carnegiea gigantea, and Opuntia spp.
The riparian vegetation of the Bill Williams River falls within a couple of different classification schemes. At the biome scale, it contains elements of the Warm-temperate Wetland Interior and Californian Riparian Deciduous Forests and Woodlands, Riparian Scrublands, Interior Strands, and Californian Interior Marshlands, and the Tropical-subtropical Sonoran Riparian Deciduous Forest, and Woodlands, Sonoran Riparian Scrublands, and Sonoran Interior Marshlands and Submergent Communites (Minckley, Brown, 1994). More detailed descriptions were provided by Szaro (1989), who describes riparian community types for the US states of Arizona and New Mexico, based on dominant species. Riparian vegetation along the Bill Williams River is dominated by several woody
species common to low elevation southwestern riparian ecosystems, including Populus fremontii, Salix gooddingii, Tamarix ramosissima, T. chinensis and hybrids (Gaskin, Schaal, 2002), Baccharis salicifolia, and Prosopis spp. Herbaceous vegetation cover tends to be low except adjacent to perennial channels where water and light availability are high.
Рис. 2. Расходы воды на р. Билл Виллиамс на гидропосту около Аламо (1941-1968), и ниже Дамбы Аламо (1969-2003; американский Геологический Датчик № 0942600): а) значения максимальных расходов за год; b) значения расходов за год; с) значения летних расходов (с 1 мая по 30 сентября). Fig. 2. Streamflow on the Bill Williams River at the gaging station "near
Alamo" (1941-1968), and "below Alamo Dam" (1969-2003; U.S. Geological Survey Gage №0942600). a) annual flood series; b) mean annual flow; c) mean summer flow (May1-Sept. 30).
In 1996, detailed field sampling of 227 study plots (20 m2 -100 m2) along sixteen permanent transects, eight on the Bill Williams River and eight on the Santa Maria River (a main tributary) revealed patterns of woody species frequency. Combined frequency data from both rivers are presented in Figure 3 to provide a general view of the vegetation across a broader part of the watershed. A total of 19 woody species or "pseudo species" were encountered within the quadrats. We consider pseudo species to be genera that contained more than one species in our plots, but that were not precisely identifiable (Atriplex spp., Lycium spp., Prosopis spp., and Tamarix spp.). Seven species or pseudo species were present in > 10% of all plots, and the non-native (to USA) Tamarix spp. had the highest frequency of all woody plants. Pluchea frequency was notably higher on the Bill Williams R., while Hymenoclea and Lycium spp. frequencies were considerably higher on the Santa Maria R. A few species occurred only in plots on one river or the other: for example, Baccharis sarothroides, Salix exigua, Prosopis pubescens, and Zizyphus obtusifolia only occurred in plots on the Santa Maria River, whereas Atriplex spp., Nicotiana glauca and Suaeda torreyana only occurred in plots on the Bill Williams R. These latter species all occurred at low frequencies (Fig. 3).
Рис. 3. Частота встречаемости 19 древесных видов на 227 выбранных участках на реках Санта Марии и Билл Виллиамс. Сокращения: AMAM - Аmbrosia ambrosioides; PAMI -Parkinsonia microphylla; NIGL - Nicotiana glauca; SUTO - Suaeda torreyana; BASR - Baccharis sarothroides; PRPU - Prosopis pubescens; SAEX - Salix exigua; ATSP - виды рода Atriplex; LATR - Larrea tridentata; ACGR - Аcacia gregii; ZIOB - Ziziphus obtusifolia; LYSP - виды рода Lycium; HYMO - Hymenoclea monogyra; POFR - Populus fremontii; BASA - Baccharis salicifolia; САГО - Salix gooddingii; PLSE - Pluchea sericea; PRSP - виды рода Prosopis; TASP - виды рода Tamarix. Fig. 3. Frequency of occurrence of 19 woody species or pseudo species in 227 sampled vegetation plots on the Bill Williams and Santa Maria rivers. Abbreviations are as follows: AMAM - Ambrosia ambrosioides; PAMI - Parkinsonia microphylla; NIGL - Nicotiana glauca; SUTO - Suaeda torreyana; BASR - Baccharis sarothroides; PRPU - Prosopis pubescens; SAEX - Salix exigua; ATSP - Atriplex species; LATR - Larrea tridentata; ACGR - Acacia gregii; ZIOB - Ziziphus obtusifolia; LYSP - Lycium species; HYMO - Hymenoclea monogyra; POFR -Populus fremontii; BASA - Baccharis salicifolia; SAGO - Salix gooddingii; PRSP - Prosopis spp.; PLSE - Pluchea sericea; TASP - Tamarix species.
The primary gradient structuring vegetation along the study rivers is the relative elevation or topographic position, which is highly correlated with such driving factors as flood frequency, flood magnitude and depth to groundwater. A simple classification grouped the sample plots into different vegetation patch types based on the species with the largest basal area (m2/ha). The three most common floodplain species (Tamarix spp., Salix gooddingii, and Populus fremontii) were further divided into two patch types (large and small), based on the diameter of the largest individual in a plot. Vegetated plots without large woody vegetation were placed into two other patch types: near channel plots dominated by herbaceous vegetation and without woody seedlings, and plots dominated by woody seedlings. The distribution of 13 different patch types as a function of their elevation above the streambed (topographic position) shows Figure 4. General groupings that emerge from figure three include low topographic positions (channel margins, channel bars) dominated by hydric or mesic herbaceous vegetation (Typha spp., Scirpus spp., Polygonum spp., Leptochloa spp.), or by woody seedlings. Slightly higher surfaces (low floodplains) tend to support smaller (younger) Salix gooddingii and Populus fromontii, and also Baccharis salicifolia
"O
U 4E
03
cd
¿3 3 -CO
CD >
О 2 -<
С
О 1 .
4—» 1
аз >
0)
Ш о -
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Vegetation Patch Type
Рис. 4. Среднее положение (+/- стандартное отклонение) биотопа над урезом воды для 13 приречных биотопов на реке Билл Виллиамс, США. Биотопы: 1 - около канала, доминирует травяная растительность; 2 - доминируют проростки древесных видов; 3 - подрост / доминирует Salix gooddingii; 4 - подрост / доминирует молодые экземпляры Populus fremontii; 5 подрост / доминирует Baccharis salicifolia; 6 - взрослые деревья / доминирует S. gooddingii; 7 - взрослые деревья, доминирует P. fremontii; 8 - подрост, доминируют виды рода Tamarix; 9 -взрослые деревья, доминируют виды рода Tamarix; 10 - доминирует Hymenoclea monogyra; 11- домирует Pluchea sericea; 12 - доминируеют виды рода Prosopis; 13 - полидоминантное сообщество взрослых деревьев. Fig. 4. Mean elevation (+/- one standard deviation) above the streambed (thalweg) for 13 riparian patch types, Bill Williams River, USA. Patch type definitions are as follows: 1 - Near channel, herbaceous vegetation dominant; 2 - woody seedlings; 3 - Saplings/small trees Salix gooddingii dominant; 4 - Saplings/small trees Populus fremontii dominant; 5 - Saplings/small trees Baccharis salicifolia dominant; 6 - Large trees S. gooddingii dominant; 7 - Large trees P. fremontii dominant; 8 - Saplings/small trees Tamarix species dominant; 9 - Large trees Tamarix species dominant; 10 - Hymenoclea monogyra dominant; 11 - Pluchea sericea dominant; 12 - Prosopis species dominant; 13 - other woody species dominant.
shrublands. Mid- to high elevation floodplains tend to be dominated by larger Populus and Salix, and all small and large Tamarix spp. Hymenoclea monogyra can form shrublands on coarse particle substrates at relatively high floodplain elevations. Finally, terraces tend to be dominated by Pluchea sericea, Prosopis spp., and various other shrubs and trees, many of which also occur as part of the zonal vegetation (Lycium spp., Acacia gregii, Atriplex spp.).
STREAMFLOW-VEGETATION RELATIONSHIPS ON THE BILL WILLIAMS RIVER
It is widely accepted among riparian ecologists that the composition and dynamics of riparian (streamside) vegetation reflect direct and indirect effects of streamflow (Hughes, 1997; Friedman, Auble, 2000; Stromberg et al., In Press). The flow regime is often the driving variable in these systems, strongly affecting other aspects of the riverine environment such as fluvial processes (e.g., channel widening, meandering) and alluvial groundwater dynamics. These factors, overlain on the geologic and climatic setting, form the physical "stage" on which vegetation dynamics play out.
Research on the Bill Williams River has revealed that periodic floods and summer base flows are key flow components that influence riparian vegetation abundance (Shafroth et al., 2002). Large magnitude floods can remove vegetation and produce bare sites for new growth, and did so on a fairly regular basis before the construction of Alamo Dam. Riparian vegetation growth and survival can be limited by periods of very low flow that do not provide sufficient moisture or replenish alluvial ground water sufficiently for the water-demanding riparian plants, especially during typically hot and dry parts of the growing season (e.g., May-September; Shafroth et al., 2000). Conversely, periods with higher low flows during the growing season, or occasional pulses that replenish soil and ground water can enable relatively high growth and survival of riparian vegetation. Vegetation at a particular point in time or space often reflects the sequence of flow events over previous years and decades, particularly with respect to the magnitude of flood flows and summer flows. Following construction of Alamo Dam, flood magnitudes were greatly reduced and summer flows were generally increased (Fig. 2), resulting in the narrowing of stream channels and associated expansion of dense, woody vegetation into areas of the bottomland that were formerly bare or only sparsely vegetated (Shafroth et al., 2002).
STREAMFLOW MANAGEMENT FOR NATIVE POPULUS AND SALIX REGENERATION
Increasingly, land and water managers worldwide are seeking ways to manage reservoir releases to produce flow regimes that simultaneously meet human needs while also maintaining the health and sustainability of downstream biota (Arthington et al., 2006; Hughes, Rood, 2003; Richter et al., 2003; Rood et al., 2005). On the Bill Williams River, resource managers are interested in controlling streamflow releases from Alamo Dam to favor the establishment of native Populus and Salix forests, while discouraging the spread of Eurasian Tamarix spp., which are viewed as undesirable weeds by many (Shafroth et al., 2005). A substantial body of research has elucidated relationships between streamflow and the germination and establishment of Populus spp. throughout semi-arid and arid western North America (for reviews cf. Braatne et al., 1996; Mahoney, Rood, 1998; Karrenberg et al., 2002). Flood timing, elevation of seedling establishment, and flood recession rates or availability of soil moisture during the first growing season are all important components of Populus recruitment and have been synthesized into a "Recruitment Box" model which uses aspects of the annual hydrograph to estimate areas of potential woody riparian seedling establishment (Mahoney, Rood, 1998).
Shafroth et al. (1998) examined woody riparian seedling establishment along the Bill Williams River in the context of the above hydrologic variables following high flow releases from Alamo Dam in 1993 and 1995. One aspect of this study was a germination model, which combined water surface levels and seed dispersal phenology (timing) to predict which sites supported or did not support seedling establishment. Observed maximum rates of decline where Populus seedlings survived for
both the 1993 (average 1.2-4.4 cm/d) and 1995 cohorts (average 2.8-4.2 cm/d) on the Bill Williams R. that were similar to the 2.5 cm/d generalized rate of Mahoney and Rood (1998). Although precipitation during a period of rapid drawdown may enhance seedling survival, rainfall is typically very sparse along the Bill Williams R. during the months following seed germination.
Removal of germinants or seedlings by high flow events subsequent to germination or establishment can be an important cause of seedling mortality in western riparian ecosystems (Stromberg, Patten, 1991; Friedman, Auble, 2000). On the Bill Williams R., removal of 1993-established seedlings and saplings by high flows in late 1993 (ca. 28 m3/s) and in early 1995 (ca. 188.9 m3/s) was not a primary cause of mortality at the quadrat scale (Shafroth et al., 1998). The apparent inability of flows up to 188.9 m3/s to remove substantial numbers of seedlings and saplings illustrates an impact of Alamo Dam, as unregulated flows would have been much larger and more destructive to young plants. The absence of high flows from 1995-2004 allowed the continued survival and growth of woody vegetation which became established in 1995 at floodplain elevations generally considered too low to allow long-term survival. High flows in fall 2004 and winter 2005 (>150m3/s) removed many of these plants (P.B. Shafroth, personal observation).
Most of the research relating aspects of flood hydrographs to the establishment of riparian vegetation has focused on Populus spp. However, a few studies have looked at applying these models to other species (Shafroth et al., 1998; Horton, Clark, 2001; Amlin, Rood, 2002). It stands to reason that other pioneer plants, which, like Populus, typically require bare, moist substrates for successful seed germination and seedling establishment, would have similar requirements. The most common other woody pioneer taxa on the Bill Williams R. (Salix gooddingii, Tamarix spp., and Baccharis salicifolia) each has a unique period of seed dispersal, though they may be overlapping at times. Salix gooddingii disperses seed later than Populus fremontii on the Bill Williams R., and thus tends to germinate either in response to later floods, or to a later period of the flood recession limb. In the second case, S. gooddingii would generally become established at a lower elevational position within the bottomland than P. fremontii (closer to ground water, but more vulnerable to scour from future high flows). Seed dispersal of non-native Tamarix spp. begins later than P. fremontii on the Bill Williams R. (Shafroth et al., 1998 - though not all rivers in western N. America - e.g., Cooper et al., 1999), and continues throughout the growing season and into fall months. Thus, Tamarix is not nearly as dependent as Populus or Salix on precisely timed floods for establishment.
STREAMFLOW RELATIONSHIPS TO SPECIES WITH WARM SEASON PHENOLOGY
Less is known about the importance of flood flows that occur during the summer or fall months, although flows in these seasons historically comprised approximately one half of the largest magnitude floods in a given year on the Bill Williams River. Some effects that are understood relate to the reproduction of species whose phenology is tied to these later precipitation events. For example, the seeds of mesquite trees (Prosopis spp.) germinate in response to late summer moisture, whether from precipitation or flooding (Stromberg, Patten, 1991). Floods may help promote Prosopis seed germination by scarifying seeds, and by providing moisture. The growth of mesquite trees is also somewhat tied to these summer storms, and floodwaters at this time of year can promote more vigorous growth by providing a supplement of soil moisture and maintaining relatively high water tables. Prosopis growth and productivity are strongly influenced by the availability of groundwater, with maximum growth occurring in bottomland settings where water tables are typically 5-6 m below the surface (Stromberg et al., 1993).
OTHER FACTORS INFLUENCING WOODY VEGETATION DYNAMICS
The maximum streamflow that can be released from Alamo Dam (198 m3/s) is insufficient to be the primary driver vegetation dynamics on all geomorphic surfaces within the bottomland. Two other processes that strongly influence vegetation dynamics are autogenic succession and fire. Flow regulation has been shown to alter succession in other riparian ecosystems (Bravard et al., 1986; Johnson, 1992). Prosopis colonizes the understory of Populus fremontii forests in Arizona (Stromberg et al., 1997; Stromberg, 1998), but other successional trajectories have not been well described. Even less is known of successional pathways in Tamarix stands, largely because Tamarix generally has not occupied sites in North America long enough to senesce. In Grand Canyon, stands of Tamarix that colonized the Colorado R. bottomland following the construction of Glen Canyon Dam are beginning to senesce and are being replaced by a mix of upland and clonal riparian species (Stevens, 1989). On the San Pedro R. in southeastern Arizona, densities of later successional species were similar in Tamarix and Populus stands, but the relationship to stand age differed, with a tendency for higher densities in older Populus stands and no apparent stand age-related pattern in Tamarix stands (Stromberg, 1998). On the Bill Williams R., we speculate that former flood-plain areas that are not reworked by the river will continue to be colonized by typical terrace taxa such as Prosopis, Pluchea, and Atriplex. The rate of colonization of Prosopis may be slowed by the lack of summer overbank floods and associated seed scarification and burial (Stromberg, Patten, 1991); however, our results did not reveal greater recruitment of Prosopis along the unregulated Santa Maria R. We have recently resampled our 227 plots and will be analyzing these data for successional trends.
Periodic fires have also affected riparian forest composition along the Bill Williams R. Fire tends to kill Populus, whereas Tamarix, Pluchea and to a lesser extent Salix are able to resprout (Busch, Smith, 1993; 1995). Tamarix may perpetuate fire disturbance due to the large quantity of standing dead stems it produces and its highly combustible, multiple-stemmed form. Altered disturbance regimes are a product of invasive species in other ecosystems (Mack, D'Antonio, 1998). Flow regulation indirectly promotes fire in riparian ecosystems because dead, combustible vegetation tends to accumulate without floods that transport and export this material and promote its decomposition. On terraces, fire favors the xeric shrub Pluchea sericea over Prosopis spp. (Busch, Smith, 1995).
CURRENT AND FUTURE RESEARCH
The Bill Williams River continues to be an excellent "natural laboratory" for studying the dynamics of riparian vegetation in the context of flow regulation and management of streamflow for ecological benefits. Heavy rains in winter 2005 resulted in another opportunity to manage streamflow to favor Populus and Salix over Tamarix. Scientists, resource managers, land owners, and the operators of Alamo Dam were able to agree upon a flood release of approximately 195 m2/s that was implemented in February and March, followed by a gradual recession through the rest of spring and summer. Preliminary assessments of the vegetation indicate that these controlled flows were very successful at promoting the establishment of native tree species. We are continuing to monitor the vegetation response to these flows. In addition, various scientists are working to better understand not only the woody vegetation dynamics, but also how streamflow regimes are affecting herbaceous vegetation, aquatic organisms, and terrestrial animals that depend on the riparian ecosystem. These scientists are meeting in the coming months to develop an integrated monitoring plan for the river system and to begin to develop linked physical-biological models of the system and its dynamics.
REFERENCES
1. Amlin N.M., Rood S.B. Comparative tolerances of riparian willows and cottonwoods to water-table decline // Wetlands 2002. Vol. 22. pp. 338-346.
2. Arthington A.H., Bunn S.E., Poff N.L., Naiman R.J. The challenge of providing environmental flow rules to sustain river ecosystems // Ecological Applications. 2006. Vol. 16. pp. 1311-1318.
3. Bravard J.-P., Amoros C., Pautou G. Impact of civil engineering works on the successions of communities in a fluvial system // Oikos. 1986. 4792-111.
4. Brinson M.M., Swift B.L., Plantico R.C., Barclay J. S. Riparian ecosystems: their ecology and status. U. S. // Fish and Wildlife Service. Office of Biological Services. Washington, DC, USA. 1981. FWS/OBS-81/17.
5. Busch D.E., Smith S.D. Effects of fire on water and salinity relations of riparian woody taxa. Oecologia. 1993. Vol. 94. pp. 186-194.
6. Busch D.E., Smith S.D. Mechanisms associated with decline of woody species in riparian ecosystems of the southwestern U.S. // Ecological Monographs. 1995. Vol. 65. pp. 347-370.
7. Cooper D.J., Merritt D.M., Andersen D.C., Chimner R.A. Factors controlling the establishment of Fremont cottonwood seedlings on the upper Green River, USA // Regulated Rivers-Research & Management. 1999. Vol. 15. pp. 419-440.
8. Defining ecosystem flow requirements for the Bill Williams River, Arizona. U.S.G.S. Shafroth P.B., Beauchamp V.B., editors. Open-File Report. 2006. (In Press).
9. Friedman J.M., Auble G.T. Floods, flood control, and bottomland vegetation // Inland flood hazards: human, riparian and aquatic communities. E. Wohl, editor. Cambridge University Press, 2000. pp. 219-237.
10. Gaskin J.F., B.A. Schaal. Hybrid Tamarix widespread in U.S. invasion and undetected in native Asian range // Proceedings of the National Academy of Sciences. 2002. Vol. 99. pp. 1125611259.
11. Horton J.L., Clark J.L. Water table decline alters growth and survival of Salix gooddingii and Tamarix chinensis seedlings // Forest Ecology and Management. 2001. Vol. 140. pp. 239-247.
12. Hughes F.M.R. Floodplain biogeomorphology // Progress in Physical Geography. 1997. Vol. 21. pp. 501-529.
13. Hughes F.M.R., Rood S.B. Allocation of river flows for restoration of floodplain forest ecosystems: a review of approaches and their applicability in Europe // Environmental Management. 2003. Vol. 32. pp. 12-33.
14. Johnson W.C. Dams and riparian forests: case study from the upper Missouri River // Rivers. 1992. Vol. 3. pp. 229-242.
15. Knopf F.L., Johnson R.R., Rich T., Samson F.B., Szaro R.C. Conservation of riparian ecosystems in the United States // Wilson Bulletin. 1988. Vol. 100. pp. 272-284.
16. MackM.C., D'Antonio C.M. Impacts of biological invasions on disturbance regimes // Trends in Ecology and Evolution. 1998. Vol. 13. pp. 195-198.
17. Mahoney J.M., Rood S.B. Streamflow, requirements for cottonwood seedling recruitment - An interactive model // Wetlands. 1998. Vol. 18. pp. 634-645.
18. Makings E. Flora of the San Pedro River (Cochise County, Arizona). M.S. Thesis. Arizona State University, Tempe, AZ, USA. 2003.
19. Minckley W.L., Brown D.E. Wetlands // Biotic communities: southwestern United States and northwestern Mexico. D.E. Brown, editor. University of Utah Press, Salt Lake City, 1994. pp. 223-287.
20. Naiman R.J., Decamps H., McClain M.E. Riparia: ecology, conservation, and management of streamside communities. Elsevier Academic Press. 2005.
21. Richter B.D., Mathews R., Harrison D.L., Wigington R. Ecologically sustainable water management: managing river flows for ecological integrity // Ecological Applications. 2003. Vol. 13. pp. 206-224.
22. Rood S.B., Samuelson G.M., Braatne J.H., Gourley C.R., Hughes F.M.R., Mahoney J.M. Managing river flows to restore floodplain forests. Frontiers in Ecology and the Environment. 2005. Vol. 3. pp. 193-201.
23. Shafroth P.B., J.R. Cleverly, T.L. Dudley, J. Stuart, J.P. Taylor, C. van Riper, and E.P. Weeks. 2005. Control of Tamarix in the western U.S.: implications for water salvage, wildlife use, and riparian restoration. Environmental Management 35. pp. 231-246.
24. Shafroth P.B., Stromberg J.C., Patten D.T. Woody riparian vegetation response to different alluvial water table regimes // Western North American Naturalist. 2000. Vol. 60. pp. 66-76.
25. Shafroth P.B., Stromberg J.C., Patten D.T. Riparian vegetation response to altered disturbance and stress regimes // Ecological Applications. 2002. Vol. 12. pp. 107-123.
26. Shafroth P.B. Downstream effects of dams on riparian vegetation: Bill Williams River, Arizona. Dissertation. Arizona State University, Tempe, Arizona, USA. 1999.
27. Shafroth P.B., Auble G.T., Stromberg J.C., Patten D.T. Establishment of woody riparian vegetation in relation to annual patterns of streamflow, Bill Williams River, Arizona // Wetlands. 1998. Vol. 18. pp. 577-590.
28. Stevens L.E. Mechanisms of riparian plant community organization and succession in the Grand Canyon, Arizona. PhD Dissertation. Northern Arizona University, Flagstaff, Arizona, USA. 1989.
29. Stromberg J.C., Beauchamp V.B., Dixon M.D., Lite S.J., Paradzick C. Importance of low-flow and high- flow components of flow regimes to restoration of riparian vegetation in semi-arid southwestern United States // Freshwater Biology. (In press).
30. Stromberg J.C. Functional equivalency of saltcedar (Tamarix chinensis) and Fremont cotonwood (Populusfremontii) along a free-flowing river // Wetlands. 1998. Vol. 18. pp. 675-686.
31. Stromberg J.C., Fry J., Patten D.T. Marsh development after large floods in an alluvial, arid-land river // Wetlands. 1997. Vol. 17. pp. 292-300.
32. Stromberg J.C., Wilkins S.D., Tress J.A. Vegetation-hydrology models: implications for management of Prosopis velutina (velvet mesquite) riparian ecosystems // Ecological Applications. 1993. Vol. 3. pp. 307-314.
33. Stromberg J.C., Patten D.T. Instream flow requirements for cottonwoods at Bishop Creek, Inyo county, California // Rivers. 1991. Vol. 2. pp. 1-11.
34. Szaro R.C. Riparian forest and scrubland community types of Arizona and New Mexico // Desert Plants. 1989. Vol. 9. pp. 69-138.
35. Turner R.M. Warm-temperate desertlands: Mohave desertscrub // Biotic communities: southwestern United States and northwestern Mexico. D.E. Brown, editor. University of Utah Press, Salt Lake City, 1994a. 153.1. pp. 157-168.
36. Turner R.M. Tropical-subtropical desertlands: Sonoran desertscrub // Biotic communities: southwestern United States and northwestern Mexico. D.E. Brown, editor. University of Utah Press, Salt Lake City, 1994b. 154.1. pp. 180-221.
37. Verry E.S., Dolloff C.A., ManningM.E. Riparian ecotone: a functional definition and delineation for resource assessment // Water, Air, and Soil Pollution. 2004. Focus 4.pp. 67-94.
38. Wolden L.G., Stromberg J.C., Patten D.T. Flora and vegetation of the Hassayampa River Preserve // Journal of the Arizona Nevada Academy of Science. 1994. Vol. 28. pp. 76-111.
