SPATIO-TEMPORAL DYNAMICS OF EDGE HABITAT IN NATURAL AND ALTERED SPRING SNOWMELT RECESSION FLOW REGIMES, SIERRA NEVADA, CALIFORNIA
CAROLINE WALLIS
Institute of Science and the Environment, University of Worcester, Henwick Grove,
Worcester, WR2 6AJ, United Kingdom
IAN MADDOCK
Institute of Science and the Environment, University of Worcester, Henwick Grove,
Worcester, WR2 6AJ, United Kingdom
SARAH YARNELL
Centre for Watershed Science, University of California, 1 Shields Avenue,
Davis, 95616, California, United States
Edge habitat, defined as the relatively shallow, slow zones in channel margins, provides important flow refugia and nursery habitat for juvenile fish, amphibians and riparian species [1,2,3,4]. The Mediterranean-Montane rivers of the Sierra Nevada foothills, California, have a spring snowmelt flow regime. The timing and rate of change of the recession limb has been shown to shape local ecological processes, such as reproductive cues for the threatened foothill yellow-legged frog (Rana boylii) [4,5,6,7]. It was hypothesised that flow regulation would alter the availability, location and spatial dynamics of edge habitat during the recession limb. A hydraulic patch dynamics framework, outlined in [8], was applied to a regulated (Middle Fork) and unregulated reach (North Fork) of the American River, California to evaluate potential differences in edge habitat dynamics. Repeat surveys of the location and hydraulic characteristics (depth, mean column streamwise and lateral velocity and velocity variability), of the edge habitat zone, delineated in the field by relative differences in depth, velocity and/or surface flow type conditions from the main channel, were conducted throughout the spring snowmelt recession limb (April–August 2011) using a 2D Acoustic Doppler Velocimeter. This paper presents the results of the patch dynamics analysis, highlighting differences in the number of hydraulic patches and their associated hydraulic characteristics as well as the change in location and extent of edge habitat throughout the recession limb. The ecological implications of flow regime alterations on edge habitat dynamics and suggestions for how these could be minimised will also be discussed.
1. INTRODUCTION
The adverse impacts of dams and river regulation have been widely documented and continue to be researched [9,10,11]. In the 1990s the importance of natural flow variability in maintaining the ecological integrity of rivers gained widespread acceptance and has become a guiding paradigm for river restoration and environmentally sensitive flow management [12,13]. Scientists and water resource managers, however, must continue to work closely to ensure appropriate quantitative evidence of the physical and ecological effects of altered flow regimes is collected and incorporated into flow management plans [14,15,16].
Recent research into the ecology of the spring snowmelt recession flow regime, characteristic of rivers in the Californian Mediterranean-Montane climate, has linked the timing, magnitude and rate of change of flows to the success of native species [6,7]. The timing of peak snowmelt flows provides the reproductive cue for cottonwood (Populus spp.) and the threatened foothill yellow-legged frog (Rana boylii) [17,18]. Peak flows scour riparian zones, resetting succession and carrying an influx of nutrients which aid the growth of new seedlings [19]. Foothill yellow-legged frogs only breed after the risk of potentially scouring peak flows has passed. Females then lay their eggs on substrate or vegetation in well oxygenated but relatively sheltered water near the channel margin. Crucial to the success of both species, whose life cycles have adapted to the flow regime in this region, is the availability, persistence and predictable shift in location of edge habitat during the subsequent flow recession. Edge habitat, variously defined as the relatively shallow, slow waters near the channel margins [1,2,20], provides ideal spawning and nursery habitat. If flows recede too quickly after peak spring runoff, egg masses and seeds are prone to dessication and tadpoles/juvenile fish to stranding.
The ecological importance of edge habitat is recognised however relatively little research has investigated how flow regulation effects the hydraulic characteristics, availability, and spatio-temporal dynamics of edge habitat during the spring snowmelt recession. A limited number of hydrodynamic modelling studies suggest that flow regulation can severely reduces edge habitat persistence and disturb the seasonal pattern of edge habitat availability [1,2]. Further research could provide important information to dam operators seeking to minimise ecological impact whilst balancing multiple water resource needs.
This research was part of a wider project investigating the relationship between hydraulics, ecology and geomorphology during the spring snowmelt recession at six sites across two catchments in the Sierra Nevada foothills, California. The specific aims of research presented in this paper were to (a) characterise and delineate hydraulic patches in the edge habitat zone of an unimpaired and impaired reach, and (b) to explore the effects of flow regulation on the location, availability and spatio-temporal dynamics of edge habitat and its constituent patches. It was hypothesised that whilst the total extent of edge habitat might be relatively similar at each site, the hydraulic characteristics and change in location of hydraulic patches within the edge habitat zone would vary between sites, reflecting the differences in bar morphology and flow regime.
2. METHODS
2.1 Study area
The research was conducted in the American River watershed which drains a 5,568km2 area of the north western slopes of the Sierra Nevada mountains in northern California. The American river is heavily modified in two of its three main forks; flows are regulated to produce hydropower and provide suitable conditions for white water rafting during summer. The Mediterranean-montane rivers in this region are characterised by a spring snowmelt recession hydrograph. Peak flows typically occur between March-May when air temperatures are sufficient to melt the high elevation snowpack. Discharge recedes throughout the summer drought months, reaching baseflow in August. During this study, owing to the unusually cold, wet spring of 2011, peak snowmelt occurred in mid to late June. The falling limb of the recession was steeper and of shorter duration than average and the tail of the recession limb extended to early September 2011.
Study sites were selected in the North and Middle Forks. The North Fork (NF_AMR) flows through a steep, narrow, wooded valley and is unimpaired. Mean daily flow during 2011 water year was 43.6m3s-1. The Middle Fork (MF_AMR) is heavily regulated by a series of dams licensed by the Federal Energy Regulatory Committee. In 2011 near-natural patterns of flow variability were maintained until mid July, after which a hydropeaking regime was operated resulting in diurnal fluctuations of 24 m3s-1 (10-34m3s-1). Mean daily flow during 2011 water year was 50.6m3s-1.
The head of a lateral bar was selected as the focal point of each study reach. Lateral bars experience considerable wetting and drying as flows vary and so provide an ideal location for studying the dynamics of edge habitat. The MF_AMR reach extended 120m and contained a pool tail out with a large recirculation zone upstream of the lateral bar (left bank) and a large vegetated mid-channel bar which split the channel adjacent to the head of the lateral bar. The bar was dominated by cobble-gravel substrate and covered with a layer of silt at the head. It varied in width from 50-75m. The profile of the bar shelved gently at its head but a sharp drop-off was evident in the cross-channel profile where the bar was adjacent to the split channel. Dense, established willow vegetation was present throughout the reach. The NF_AMR reach, scaled to the channel width, extended 96m, and incorporated a pool tail out at the head the lateral bar, a run and a rapid. The lateral bar extended the full length of the reach on the right bank, shelving gently from a bedrock outcrop at the channel margins into the main channel. The bar was 15-25m wide, composed of coarse cobble substrate and sparsely vegetated with willow saplings. Inter-site differences were inevitable but minimised where possible.
2.2 Data collection
Six surveys of the spatial extent and hydraulic characteristics of the edge habitat zone were carried out in each reach between April-August 2011 at a range of discharges representative of (a) peak snowmelt, (b) high flow during the declining limb of the recession, (c) moderate flow at mid-recession and (d) the hydropeaking regime in the impaired reach and the tail of the recession limb in the unimpaired reach. Edge habitat was delineated in the field based on relative differences in depth, velocity and/or surface flow type from the main channel. Water depth and mean column velocity (streamwise and cross-stream) at 0.6depth, were measured at 0.8m/1m cross-stream and 4m/5m streamwise intervals at the NF_AMR and MF_AMR sites respectively using a Sontek FlowTracker 2D Acoustic Doppler Velocimeter.
2.3 Data analysis
A hydraulic patch dynamics framework presented in [8] was applied to evaluate differences in edge habitat dynamics. Hydraulic data were combined by site and standardised (z-scores) to account for different scales of measurement. Each combined dataset was clustered using the fuzzy c-means [21] and Gath-Geva [22] algorithms in the Fuzzy Clustering and Data Analysis Toolbox for MATLAB [23] to delineate relatively homogeneous hydraulic patches and the transitional zones between them. Seven partitions of each combined dataset were derived, each with a unique number of clusters (c) in the range 2≤c≤8. The fuzziness value (m) was held constant at m=2. Fuzzy clusters were defuzzified to define crisp hydraulic patches using a 0.7 α-cut threshold and Confusion Index (CI>0.6) rule [24,25].
Hydraulic patches and transition zones were mapped at each flow using ArcGIS v.10 (ESRI, 2010) and exported to FRAGSTATS v.3.3 [26]. Thirteen spatial metrics quantifying patch diversity, patch geometry and patch configuration were calculated within the edge habitat zone at each discharge. Permutation-based resampling techniques were used to test for statistical differences in hydraulic patch diversity in the edge habitat zone between discharges and sites. Patch configuration was explored using non-metric multidimensional scaling (MDS) and multivariate statistics in PRIMER v.6.1 [27]. GIS overlays were also used to calculate the percentage change in location of the boundary and constituent hydraulic patches of the edge habitat zone with every increase in discharge [28].
REFERENCES
[1] Bowen Z., Bovee K. and Waddle T., “Effects of flow regulation on shallow-water habitat dynamics and floodplain connectivity”, Transactions of the American Fisheries Society, Vol. 132 (2003), pp 809-823.
[2] Freeman M., Bowen Z, Bovee K. and Irwin E., “Flow and habitat effects on juvenile fish abundance in natural and altered flow regimes”, Ecological Applications, Vol. 11, No. 1, (2001), pp 179-190.
[3] Merritt D. and Cooper D., “Riparian vegetation and channel change in response to river regulation: a comparative study of regulated and unregulated streams in the Green River basin, USA”, Regulated Rivers: Research and Management, Vol. 16, (2000), pp 543-564.
[4] Kupferberg S., “Hydrologic and geomorphic factors affecting conservation of a river-breeding frog (Rana boylii)”, Ecological Applications, Vol. 6, No. 4, (1996), pp 1332-1344.
[5] Lind A., Welsh A and Wilson, R., “The effects of a dam on breeding habitat and egg survival of the foothill yellow-legged frog (Rana boylii) in northwestern California”, Herpetological Review, Vol. 27, No. 62, (1996), pp 67.
[6] Yarnell S., Lind A. and Mount J., “Dynamic flow modelling of riverine amphibian habitat with application to regulated flow management”, River Research and Applications, Vol. 28, No. 2, (2012), pp 177-191.
[7] Gasith A and Resh V., “Streams in Mediterranean climate regions: abiotic influences and biotic responses to predictable seasonal events”, Annual Review of Ecology and Systematics, Vol. 30, (1999), pp 51-81.
[8] Wallis C., Maddock I., Visser F. and Acreman, M., “A framework for evaluating the spatial configuration and temporal dynamics of hydraulic patches”, River Research and Applications, in press.
[9] Bunn S. and Arthington A., “Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity”, Environmental Management, Vol. 30, No. 4, (2002), pp 492-507.
[10] Petts G. and Gurnell A., “Dams and geomorphology: research progress and future directions”, Geomorphology, Vol. 71, No. 1-2, (2005), pp 27-47.
[11] Ma Y., Huang, H., Nanson G., Li Y. and Yao W., “Channel adjustments in response to the operation of large dams: the upper reach of the lower Yellow River”, Geomorphology, Vol. 147-148, (2012), pp 35-48.
[12] Poff N., Allan J., Bain M., Karr J., Prestegaard K., Richter B., Sparks R. and Stromberg J., “The natural flow regime”. BioScience, Vol. 47, No. 11, (1997), pp769-784.
[13] Postel S. and Richter B., “Rivers for life: managing water for people and nature”, Island Press, (2003).
[14] Poff N. and Zimmerman J., “Ecological responses to altered flow regimes: a literature review to inform the science and management of environmental flows”, Freshwater Biology, Vol. 55, (2010), pp 194-205.
[15] Acreman M. and Ferguson J., “Environmental flows and the European Water Framework Directive”, Freshwater Biology, Vol. 55, (2010), pp 32-48.
[16] Richter B., Matthews R., Harrison D. and Wigington R., “Ecologically sustainable water management: managing river flows for ecological integrity”, Ecological Applications, Vol . 13, No. 1, (2003), pp 206-224.
[17] Yarnell S., Viers J. and Mount J., “Ecology and management of the spring snowmelt recession”, BioScience, Vol. 60, No. 2, (2010), pp 114-127.
[18] Stella J., Battles, J., Orr B and McBride J., “Synchrony of seed dispersal, hydrology and local climate in a semi-arid river reach in California”, Ecosystems, Vol. 9, (2006), pp 1200-1214.
[20] Naiman R., Latterall J., Pettit N. and Olden J., “Flow variability and the biophysical vitality of river systems”, Comptes Redus Geoscience, Vol. 340, (2008), pp 629-643.
[21] Papanicolaou A., Elhakeem M., Dermiss D. and Young N., “Evaluation of the Missouri shallow water habitat using a 2D-hydrodynamic model”, River Research and Applications, Vol. 27, (2011), pp 157-167.
[22] Bezdek J., Ehrlich R. and Full W. “FCM: the fuzzy c-means clustering algorithm”, Computers & Geosciences, Vol. 10, No. 2-3, (1984), pp 191-203.
[23] Gath I. and Geva, A., “Unsupervised optimal fuzzy clustering” IEEE Transactions on Pattern Recognition and Machine Intelligence, Vol. 11, No. 7, (1989), pp 773-781.
[24] Balasko B., Abonyi J. and Feil B., “Fuzzy clustering and data analysis toolbox for MATLAB”, (2001), Online at: http://www.mathworks.com/matlabcentral/fileexch ange/7486.
[25] Cheng T., Molenaar M. and Lin H., “Formalizing fuzzy objects from uncertain classification results”, International Journal of Geographical Information Science, Vol 15, (2001), pp 27-42.