Lawrence E. Stevens

PLANNED FLOODING AND COLORADO RIVER RIPARIAN TRADE‐OFFS DOWNSTREAM FROM GLEN CANYON DAM, ARIZONA

Regulated river restoration through planned flooding involves trade-offs be- tween aquatic and terrestrial components, between relict pre-dam and novel post-dam re- sources and processes, and between management of individual resources and ecosystem characteristics. We review the terrestrial (wetland and riparian) impacts of a 1274 m 3 /s test flood conducted by the U.S. Bureau of Reclamation in March/April 1996, which was designed to improve understanding of sediment transport and management downstream from Glen Canyon Dam in the Colorado River ecosystem. The test flood successfully restored sandbars throughout the river corridor and was timed to prevent direct impacts to species of concern. A total of 1275 endangered Kanab ambersnail (Oxyloma haydeni kan- abensis) were translocated above the flood zone at Vaseys Paradise spring, and an estimated 10.7% of the total snail habitat and 7.7% of the total snail population were lost to the flood. The test flood scoured channel margin wetlands, including potential foraging habitats of endangered Southwestern Willow Flycatcher (Empidonax traillii extimus). It also buried ground-covering riparian vegetation under .1 m of fine sand but only slightly altered woody sandbar vegetation and some return-current channel marshes. Pre-flood control ef- forts and appropriate flood timing limited recruitment of four common nonnative perennial plant species. Slight impacts on ethnobotanical resources were detected .430 km down- stream, but those plant assemblages recovered rapidly. Careful design of planned flood hydrograph shape and seasonal timing is required to mitigate terrestrial impacts during efforts to restore essential fluvial geomorphic and aquatic habitats in regulated river eco- systems.

THE 1996 CONTROLLED FLOOD IN GRAND CANYON: FLOW, SEDIMENT TRANSPORT, AND GEOMORPHIC CHANGE

The 1996 controlled flood released from Glen Canyon Dam into the Colorado River was a small magnitude, short duration event compared to pre-dam floods. The controlled flood was of lesser magnitude than a 1.25-yr recurrence, and only 10% of the pre-dam spring snowmelt floods during the period 1922–1962 were of lower magnitude. The flood occurred unusually early: 36–38 d prior to any previous annual flood since 1922. The stage difference between the floods peak and the recessional baseflow was smaller than in those pre-dam years of similar magnitude or annual volume. However, the controlled flood was large from the perspective of the post-dam flood regime. The flood had a recurrence of 5.1 yr for the period between 1963 and 1999 and a similar magnitude flood had not occurred in 10 yr. The sediment flux of the flood was small in relation to pre-dam floods, and the suspended sand concentration was within the historical variance for flows of similar magnitude. This flood reworked fine-grained deposits that are primarily composed of sand, but the flood caused much less reworking of coarser grained deposits. Scour primarily occurred in the offshore parts of eddies, in many eddy return-current channels, and in some parts of the main channel. Return-current channels constitute important nursery habitats for the native fishery when baseflows are low, because these channels become areas of stagnant and warmer water. The number and area of these backwaters increased greatly after the flood. Fluvial marshes were extensively scoured because these habitats occur in the low elevation centers of eddies where velocities during the flood were large. Riparian shrubs that were inundated along the banks were not scoured, however, because these shrubs occur where flood velocities were very low and where deposition of suspended sediment occurred. Some physical changes persisted for several years, but other changes, such as the area of newly formed backwaters decreased quickly. Thus, the lasting effect of this flood varied among different small-scale fluvial environments.

The influence of floods and precipitation on Tamarix establishment in Grand Canyon, Arizona: consequences for flow regime restoration

Decoupling of climate and hydrology combined with introduction of non-native species creates novel abiotic and biotic conditions along highly regulated rivers. Tamarix, a non-native shrub, dominates riparian assemblages along many waterways in the American Southwest, including the Colorado River through Grand Canyon. We conducted a tree-ring study to determine the relative influences of climate and hydrology on Tamarix establishment in Grand Canyon. Riparian vegetation was sparse and annually scoured by large floods until completion of Glen Canyon Dam, which allowed pioneer species, including Tamarix, to expand. Post-dam floods in the mid-1980s were associated with high Tamarix mortality but also initiated a large establishment event. Subsequent establishment has been low but continuous with some exceptions. From 1984 to 2006 establishment increased during years of high, late-summer flows followed by years of low precipitation. This combination provided moist surfaces for Tamarix establishment and may have caused reduced erosion of seedlings or reduced competition from native plants. Attempts to mimic pre-dam floods for ecosystem restoration through planned flood releases also have affected Tamarix establishment. Early (March 1996) and late (November 2004) restoration floods limited establishment, but a small restoration flood in May 2000 followed by steady summer flows permitted widespread establishment. Flood restoration is not expected to prevent Tamarix spread in this system because historic flood timing in May–July coincides with seed release. To decrease future Tamarix establishment, river managers should avoid floods during peak Tamarix seed release, which encompasses the historic spring and early summer flooding period. Tamarix dominance may be reduced by early spring floods that initiate asexual reproduction of clonal shrubs (e.g., Salix exigua, Pluchea sericea).

Large mixed-species dispersal flights for predatory and scavenging aquatic Heteroptera and Coleoptera, northern Arizona, USA

Abstract We report 4 incidents of large (hundreds to hundreds of thousands of individuals) mixed-species flights of predatory and scavenging aquatic Heteroptera and Coleoptera. The events occurred on normal (mostly clear, calm) autumn 2005 and spring 2006 late afternoons near Flagstaff, Arizona. Flight days were either near the full moon or in advanced waning lunar periods. At least 18 species were involved in the flights, with as many as 16 species (7 species of Heteroptera in 3 families, 9 species of Coleoptera in 2 families) in a single flight. Heteroptera (especially from the family Corixidae) were 2–4 orders of magnitude more numerous than Coleoptera. The diasporas plummeted, with much direct mortality, onto green metal roofs, and the largest event lasted more than 2 hours. Even after 2 days, uninjured individuals failed to resume their flights. The literature suggests that such flights occur for autumn movement to winter habitats or for location of springtime habitats for reproduction. The relative proportions of species in flocks were similar neither to the relative proportions in the region nor to those in nearby livestock watering tanks, indicating that the flights consisted of nonrandom assemblages of species. These events are rare or unique observations of coordinated movement of mixed predatory and scavenging invertebrate species.

Odonata biogeography in the Grand Canyon Ecoregion, Southwestern USA.

ABSTRACT The Odonata fauna of the Grand Canyon ecoregion (GCE) on the southern Colorado Plateau includes 89 species (35 genera, seven families), including 49 Anisoptera species (25 genera, four families) and 40 Zygoptera species (10 genera, three families), and with 58 Odonata species in Grand Canyon (GC; 24 genera, seven families). Three biogeographic hypotheses account for this relatively high regional species richness: faunal affinity (origin), elevation effects on range, and landform impacts across spatial scale. The GCE Odonata assemblage is the result of mixing of taxa from adjacent Neotropical and Nearctic regions. Allochthonous taxa include 34.8% tropical (Mexican, Caribbean, Neotropical, or Pantropical) and 21.3% boreal (Nearctic or Holarctic) species. Autochthonous species (43.8%) are range-centered in North American, neither clearly Nearctic nor Neotropical, with a strong Pacific Coast influence. Area-adjusted species richness is negatively linearly related to elevation. Tropical species have lower elevation ranges than do boreal species, whereas the elevation ranges of both allochthonous groups overlap those of autochthonous species. Odonata generally overcome landform-based range constraints at coarse spatial scales, but barrier/filter and corridor effects predominate over refuge and null biogeographic effects in GC. Anisoptera and Zygoptera biogeographic patterns are similar, except that 9-fold more Zygoptera species exist in refugia in GC compared with Anisoptera. Although no GCE Odonata previously have been considered rare or at risk, 15 (16.9%) species are restricted to three or fewer localities, four (4.5%) of species have been detected at only a single locality, and four high-elevation Nearctic species may be at risk of extirpation though climate change impacts on their habitats.

Biogeography Of Aquatic And Semiaquatic Heteroptera In the Grand Canyon Ecoregion, Southwestern USA

Abstract We examined the biogeography of aquatic and semiaquatic Heteroptera (ASH) in the Grand Canyon (GC) ecoregion (GCE) on and adjacent to the southern Colorado Plateau. We report 89 ASH taxa in 86 species, 37 genera, and 14 families in the GCE, including 54 ASH taxa detected within or on the rims of GC and its major tributaries, a fauna 3.8-fold greater than previously reported. We tested 2 groups of biogeographic hypotheses to account for this high level of diversity, demonstrating an underlying pattern of mixed biogeographic affinity and strong landform-climate effects. Equal numbers of ASH taxa were derived from allochthonous (neotropical and nearctic) sources and autochthonous (range-centered) sources. A negative linear relationship existed between area-adjusted ASH taxon density and elevation, with more Mexican/neotropical taxa at low elevations and more nearctic taxa at higher elevations. While species richness was positively scale dependent, biogeographic landform impacts were unrelated or negatively related to spatial scale. The uplifted southern margin of the Colorado Plateau along the Mogollon Rim supported elevated ASH diversity as a function of ecotone effects and interprovincial basin connectivity. Barrier/filter effects were stronger than null, or refuge effects, and little endemism was detected in the GCE. Colonization history varied across elevation and in relation to landscape evolution. No reported GCE taxa have been extirpated, but 52.8% of the fauna occurred at 3 or fewer localities (primarily springs), sites that may be threatened by habitat alteration and climate change.

Odonata of Ash Meadows National Wildlife Refuge, Southern Nevada, USA

Abstract The Odonata of Ash Meadows National Wildlife Refuge (AMNWR) in southern Nevada were studied bimonthly in 2004 and 2005, revealing 32 species, a moderately high level of diversity for this relatively small, semi-isolated southern Nevada valley. Enallagma civile (Coenagrionidae) was the most regularly encountered species, followed by Rhionaeschna multicolor (Aeshnidae), Argia sedula (Coenagrionidae), and Pachydiplax longipennis (Libellulidae). Fourteen species were detected at three or fewer sites. The assemblage was co-dominated by taxa with ranges centered in North America and western North America, and 25% of the fauna were Mexican-neotropical. We report Macrodiplax balteata as new to Nevadas Odonata list, and six other new Nye County records. Odonata larval density/m2 and overall species richness (but not Shannon-Weiner diversity) were highest in the largest AMNWR wetlands, regardless of whether they were natural or anthropogenic, and were greater in two restored springs. Several of the most regularly detected larval Anisoptera (i.e., Erpetogomphus compositus and Erythemis collocata) were benthic ooze dwellers that have a flattened body morphology, which may allow them to avoid predation by non-native Procambarus clarki crayfish. Geomorphic restoration of springs may increase Odonata production, while augmentation of habitat area may increase species richness.

Springs ecosystem distribution and density for improving stewardship

Springs support some of the most diverse and unique ecosystems on Earth, but their stewardship has been hindered by the lack of knowledge of the distribution and density of springs across landscapes. Death Valley National Park (DEVA) and the State of Arizona in the USA are 2 landscapes for which significant knowledge exists about the distribution and density of springs. We used data on springs in DEVA to test the application of accumulation curves for estimating spring density. We used a spring-specific database in Arizona as an example of how to compile geospatial information for a large landscape. In both landscapes, springs are nonrandomly distributed because they emerge in topographically and geologically complex terrain and in clusters of multiple sources. Thus, estimates of their density depend on the spatial scale of inquiry and the extent to which sources are considered independent. For example, based on the current inventory, density in DEVA is estimated to be 0.033 to 0.074 springs/km2 depending on whether springs are defined as individual orifices or as complexes (groups of related spring orifices). The best data for springs as individual orifices yield an estimated 0.035 springs/km2 in Arizona. These densities are based on current data sets, and an unknown number of springs remain unmapped in both landscapes. To predict the total number of springs in DEVA, we used a modified density accumulation curve, involving the number of springs detected in surveys over the past century. The analysis indicated that undocumented springs may exist across the landscape. Knowledge of the distribution and density of the springs can help land and resource managers develop unbiased prioritizations of spring ecosystems for stewardship actions. Management actions could benefit further from an understanding of the emergence environment of a complex of springs, instead of each emergence point of a spring in a complex.

Tamarisk Reproductive Phenology and Colorado River Hydrography, Southwestern USA

ABSTRACT Non-native tamarisk (Tamarix spp.) widely colonized riparian habitats and reservoir shorelines across elevation throughout the western United States during the 20th century. To improve understanding of tamarisk reproductive plasticity, we compiled elevation, date, and phenology data from specimens in Southwestern herbaria, and we conducted field studies in Glen and Grand Canyons from 1984–2009. We modeled tamarisk reproductive seed-release phenology across elevation in relation to hydrography and flow management in the Colorado River Basin. We compared the potential for tamarisk recruitment in the pre- and post-dam Colorado River mainstream with that in Lake Mead and Lake Powell reservoirs and in tributaries with low- or high-elevation headwaters. Flooding timed with seed release was likely to result in tamarisk recruitment. Conversely, planned floods from Glen Canyon Dam that specifically avoided the May-June peak tamarisk seed-release period resulted in little tamarisk recruitment downstream in Grand Canyon. Failing recruitment in the post-dam Colorado River mainstream in Grand Canyon has occurred because: (1) the spring-summer hydrograph is generally unsuitable for tamarisk seedling establishment, and (2) post-dam flooding has coarsened grain size, resulting in less stable soils with lower nutrient concentrations. Hydrograph management can result in reduced tamarisk recruitment, while poorly timed floods, simulated natural flow regimes, and unregulated tributary flows permit tamarisk establishment. We discuss tamarisk life history in relation to ongoing environmental changes in biological controls and habitat management.

A New Springsnail (Hydrobiidae: Pyrgulopsis) from the Lower Colorado River Basin, Northwestern Arizona

Abstract. We describe a new springsnail species, Pyrgulopsis hualapaiensis, from the Lower Colorado River basin (northwestern Arizona) that has an ovate- to narrow-conic shell and narrow penis ornamented with a small gland on the distal edge of the lobe. This new species differs from closely similar congeners from the Lower Colorado River basin in several details of female reproductive anatomy and in its mtCOI haplotype (3.0%–5.0% mean sequence divergence). Bayesian, maximum parsimony, and distance-based phylogenetic analyses of COI data congruently resolved P. hualapaiensis as sister to a divergent lineage of Pyrgulopsis thompsoni in the middle Gila River watershed (southeastern Arizona), although this relationship was not well supported. Pyrgulopsis hualapaiensis is endemic to a spring complex in the Hualapai Indian Reservation that is a culturally sensitive site for the tribe. The small population of these snails appears to be robust despite recent habitat modifications (trenching of outflow and construction of a spring box) and disturbance from road traffic. Future conservation measures could include monitoring of the population and augmentation of the gravel habitat used by these snails.