Drivers of thermal and biological diversity within rivers


6th Creek, a typical salmon stream in Bristol Bay.

I am interested in understanding the processes that regulate  thermal regimes in streams and rivers and the subsequent effects on aquatic species. Part of my dissertation research quantified how watershed topography generates rich thermal variation in tributaries across river basins in SW Alaska. Watershed slope, elevation, drainage area and presence of lakes are all simple characteristics of watersheds that together can explain about 72% of the variation in average summer temperature among streams.

My research also suggests that snowmelt buffers summer stream temperatures to warmer weather. Using stable isotopes of oxygen and hydrogen in water, I estimated the contribution of snow relative to rain in stream discharge during the summer months. Snow dominated streams draining steeper catchments are 5 to 8 times less sensitive to variation in summer air-temperature compared to rain dominated watersheds draining flatter, lower elevation basins. However, this work also suggests that this network thermal heterogeneity may be lost with reduced snowpack and increased ratios of rain to snowmelt in stream discharge with climate change.

I also study how lakes influence the thermal regimes of downstream rivers. Lakes are well-known geomorphic features in boreal river systems that typically stabilize thermal regimes of downstream rivers. This work does not often account for the potential destabilizing effect that wind has on surface temperatures and therefore on outflowing rivers. I show how wind action on large upstream lakes can create temperature excursions in the river by 5-10°C in few hours. In fact, rivers draining larger lakes are some of most variable environments that we found in our surveys of stream temperature in Alaska.

Lisi PJ, DE Schindler, KT Bentley, and GR Pess. 2013. Association between geomorphic attributes of watersheds, water temperature, and salmon spawn timing in Alaskan streams Geomorphology 185: 78-86 

Lisi, P. J., D. E. Schindler, T. J. Cline, M. D. Scheuerell, and P. B. Walsh.  Topography and snowmelt control stream thermal response to air temperature. Geophysical Research Letters (in Review).

Lisi, P. J. and D.E. Schindler. Destabilization of river thermal regimes by wind action on upstream lakes. Limnology and Oceanography (In Press).


These diverse expressions of climate are an important attribute of river system and often an under appreciated dimension of ecosystem complexity. I study how the thermal heterogeneity across a river basin is important to salmon spawning habitat and the terrestrial species that rely upon them. In the rivers in coastal Alaska, variation in the phenology of salmon spawn-timing is determined by variation in water temperature among streams due to the geomorphic complexity of watersheds (Lisi et al. 2013). My work shows how this variation in salmon residence can be indirectly associated to the bloom timing of riparian carrots that are pollinated by carrion flies that depend on salmon carcasses (Lisi and Schindler 2011). Further research shows that the phenologic and spatial variation in salmon spawning can triple the foraging season for predators and scavengers such as gulls and bears (Schindler et al. 2013).


From left to right: Carrion blowflies visiting kneeling angelica on a salmon stream.  Coastal brown bears cruising Lake Nerka’s beaches at night.  Glaucous gulls scavenging salmon carcasses on Pick Creek. 


Lisi PJ  and DE Schindler. 2011. Spatial variation in timing of marine subsidies influences riparian phenologythrough a plant-pollinator mutualism. Ecosphere. 2(9):1-14.

Schindler et al. 2013. Riding the crimson tide: mobile terrestrial consumers track phenological variation in spawning of an anadromous fish. Biology Letters 9:1–4.

Conservation of endemic native fishes in the Hawaiian Archipelago


A waterfall in Puerto Rico (I haven’t been to Hawaii yet!)

As part of my postdoctoral research at the University of Wisconsin, I am exploring how the hydrologic variation among Hawaiian streams informs the life history diversity of endemic freshwater gobies. Streams of Hawaii are characterized by flashy flows and neighboring watersheds can have widely different daily streams flows due to strong precipitation gradients between leeward and windward sides of the islands. Gobies that live in these streams, (Awaous spp.) have an amphidromous life history- meaning that adult gobies spawn in freshwater streams, their larvae immediately flush out to sea, and months later, juvenile gobies return to freshwater streams for the rest of their lives.  However, results from laser microchemistry of goby otoliths reveals that 60 to 80% of gobies choose an entirely freshwater lifestyle, never reaching the sea. I am finding that the hydrologic variability of the watersheds helps shapes the proportion of the stream population that goes to sea: the more variable the stream, the more likely for the stream to contain sea-going individuals. What the trade-offs are for the individual and the conservation of the species? Why are some streams more variable in their stream flow than others? A drier climate, water diversions for tourism/agriculture, and aquatic invasive species create serious concern for goby habitat in Hawaii. This research is supported by the Department on Defense (SERDP), with Co PI’s at Tulane University (Michael Blum), University of Wisconsin (Peter McIntyre), NC State (Jim Gilliam) and Texas A&M (Derek Hogan).


Climate research on lakes.

Lakes are lowland areas that integrate the upland effects of the watersheds they reside in. The physical and chemical characteristics provide unique record of biological processes as they change through time, something I became interested in from paleo-limnological research at the University of Washington. Through collaborative efforts with lead PI’s Daniel Schindler, Peter Leavitt, Bruce Finney, Irene Gregory-Eaves, and several others, we reconstructed the 500yr history of a number of Pacific Salmon nursery lakes in US and Canada. To do this, we use carbon and nitrogen stable isotopes and pigment records from annual layers in lake sediment cores. This research shows that 1) commercial fishing significantly reduced the amount of marine derived nitrogen across the majority of salmon nursery lakes and 2) prior to commercial fishing, salmon populations showed asynchronous 30 to 80 year fluctuations across nursery lakes (Rogers et al. 2013).  In addition, sediment cores from non-salmon lakes revealed a coherent anthropogenic signature of atmospheric nitrogen, a signal starting around industrial revolution (Holtgrieve et al. 2013).

Rogers, et al. 2013. Centennial-scale fluctuations and regional complexity characterize Pacific salmon population dynamics over the past five centuries. PNAS 110:1750–1755.

Holtgrieve, et al. 2011. A coherent signature of anthropogenic nitrogen deposition to remote watersheds of the Northern Hemisphere. Science 334: 1545-1548

With much shorter historical data sets, I am exploring how lake transparency responds to variation in hydrology over the last 30 years in NW Wisconsin lakes. Local citizens are concerned with the environmental drivers that impact the water transparency, ecology, and recreational use of local lakes. I hypothesize that water transparency should fluctuate similarly among neighboring lakes in response to local variation in the precipitation and nutrient inputs.Here, I take advantage of statistical methods used in econometrics to examine how water transparency is changing in group of 25 neighboring lakes and if periods of drought have a detectable effect on water transparency and if there are any shared temporal trends between lakes responding to the same climate signal.  In fact, many lakes have water transparencies that are clearer during drought and much darker with flooding. However against conventional wisdom, eutrophic lakes show the inverse response through time, with clearer water during flooding. I am testing how the basin characteristics, redox conditions, and restoration activities mediate hydrological effects on water transparency. This work and methods may be appropriate for examining time series in streams, reservoirs and lakes elsewhere

secchi disk image

Z-scored secchi depth time series of Silver lake and Mud lake in NW Wisconsin. Temporal dynamics captured by a lake level record


Fish ecology

I am amazed at what fish eat and how variable diet contents can be from between fish, across space, and through time. We have discovered a number of patterns in the diet contents from rainbow trout, Arctic char, and Arctic grayling that live in streams and lakes in Alaska.  These resident fish tend to follow the spawning aggregation of sockeye salmon, targeting their energy-rich eggs that drift downstream during spawning events.  In fact, resident fish can put on most of their summer growth from just a few week of eating salmon eggs.  However, salmon populations are not always plentiful.  Resident fish need to flexible enough to survive on much lower quality insect prey when salmon are not around – and not every stream has the same basal productivity for supporting fish.  Compared to Arctic grayling, rainbow trout seem to be much better at finding salmon eggs when salmon populations are much lower whereas grayling tend to feed on terrestrial and aquatic insects (Bentley et al. 2012).

Bentley, KT, DE Schindler, JB Armstrong, CP Ruff, and PJ  Lisi.2012. Inter-annual variation in a pulsed resource subsidy mediates the foraging and growth response of stream-dwelling salmonids. Ecosphere

In some cases, larger resident fishes opportunistically feed on small land mammals.  The occasional shrew in the stomach contents of large tout sparked my curiosity. I was interested in how often this actually happens and looked back at 13 years of diet content data.  Every few years, about 25% of Arctic grayling and rainbow trout (those greater than 12 inches in length) eat shrews in peak years.  But not all 12-inch grayling or trout get to eat shrews.  Often the largest fish had many more shrews then the next largest fish, sometimes excluding their smaller peers access to shrews.  We believe the periods where fish eat shrews represent an abundant shrew population on land. When shrew populations peak, it might create a situation in which shrews take greater risks to find their insect prey – hunting closer water’s edge or entering the water to find aquatic insects or salmon eggs.

Lisi, P. J., K. T. Bentley, J. B. Armstrong, and D. E. Schindler. 2013. Episodic predation of mammals by stream fishes in a boreal river basin. Ecology of Freshwater Fish:1–9.


Arctic grayling with 8 shrews in several stages of digestion photo by Jonny Armstrong



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s