Upper Mississippi River Restoration Program

Upper Mississippi River Restoration Program

Long Term Resource Monitoring

 

Upper Mississippi River Restoration Program

Long Term Resource Monitoring element

Fisheries

Upper Mississippi River Restoration - Environmental Management Program U. S. Army Corps of Enineers

Management-relevant presumptive fish habitat models in a large river system

email Brian Ickes Brian S. Ickes
U.S. Geological Survey
Upper Midwest Environmental Sciences Center
2630 Fanta Reed Road
La Crosse, WI 54603

Guidance for Crediting UMRR and its Long Term Resource Monitoring element

Introduction

Habitat rehabilitation efforts on the Upper Mississippi River System (UMRS) are designed and planned in partnership ventures between Federal and State science and management agencies under the auspices of a Congressionally defined program known as the Upper Mississippi River Restoration Program (UMRR). Since 1989, more than 41,000 ha of habitat have been rehabilitated in the UMRS valley by the UMRR, representing ~ 4% of total rehabilitation potential. 

As rehabilitation practices have evolved, increasingly cost-benefit analysis and pre and post project assessments have become a major part of rehabilitation efforts, both to ensure cost-benefit efficacy and to adaptively learn best practices that can be carried forward into future habitat rehabilitation efforts. Data driven and science informed approaches to habitat rehabilitation are now the norm in habitat rehabilitation planning and design efforts on the UMRS. Correspondingly, new technical methods increasingly support habitat rehabilitation efforts in the UMRS basin, and this is an active area of work by technical and scientific assets in the UMRS basin. Examples include Status and Trends assessments of natural resources within the UMRS basin (USGS 1999; Johnson and Hagerty, eds. 2008), Decision Support Systems that help managers bring volumes of data and maps into the project planning forums, and spatially-explicit models that seek to define and predict important ecological responses so that managers can forecast prospective project benefits (Ickes et al. 2014a).

This project page describes and provides access to resources resulting from recent efforts to generate spatially-explicit habitat models for 28 fish species (Ickes et al. 2014a) over 1900 km of river, using unprecedented data assets available through the UMRR’s Long Term Resource Monitoring (LTRM) element. Included are links to the data used to create the habitat models, descriptions of the methods used, publications generated by these efforts, and spatially-explicit presumptive fish habitat maps for many species. These resources stand in support of habitat planning and design efforts across the UMRS.

Data

The data derived from the UMRR LTRM and were gained through public portals to the raw data. Data represented all sample site scale day electrofishing observations 1993-2014, from each of the six study reaches monitored by the LTRM (Figure 1). From these data, 28 fish species representing lentic, lotic, and generalist habitat guilds (O’Hara et al. 2007) were selected for modeling. All catch data were transformed from abundance (continuous) data to presence/absence (binary) data. In addition, 17 sample site-specific environmental variables observed synoptically with the catch data were retained for modeling as predictors of species occurrence at each sampling site. All LTRM data are collected using highly standardized sampling methods and using a randomized statistical design (Ratcliff et al. 2014; Ickes et al. 2014b). The data represent a 22 year random sample of each of the six study reaches. This randomization in sample design assures unbiased (representative) observations of both fish observations and environmental attributes associated with each fish observation. The 28 species that were modeled, including their habitat guild designations, are presented in Table 1. The environmental variables and their data attributes are presented in Table 2.
Figure 1. Locations of the six UMRR LTRM study reaches.
Figure 1. Locations of the six UMRR LTRM study reaches. The six reaches average 50 km in length and 2.5 km in width.

Table 1. Fish species selected for modeling.

Species Scientific name Guild
Black crappie Pomoxis nigromaculatus Lentic
White crappie Pomoxis annularis Lentic
Bluegill Lepomis macrochirus Lentic
Largemouth bass Micropterus salmoides Lentic
Warmouth Lepomis gulosus Lentic
Northern pike Esox lucius Lentic
Yellow perch Perca flavescens Lentic
Blue catfish Ictalurus furcatus Lotic
Flathead catfish Pylodictis olivaris Lotic
Rock bass Ambloplites rupestris Lotic
Skipjack herring Alosa chrysochloris Lotic
Blue sucker Cycleptus elongates Lotic
Shovelnose sturgeon Scaphirhynchus platorynchus Lotic
Sauger Sander canadense Lotic
Golden redhorse Moxostoma erythrurum Lotic
Shorthead redhorse Moxostoma macrolepidotum Lotic
Channel catfish Ictalurus punctatus Generalist
Red shiner Cyprinella lutrensis Generalist
Logperch Percina caprodes Generalist
Brook silverside Labidesthes sicculus Generalist
Freshwater drum Aplodinotus grunniens Generalist
Emerald shiner Notropis atherinoides Generalist
Blackstripe topminnow Fundulus notatus Generalist
Smallmouth bass Micropterus dolomieu Generalist
Longnose gar Lepisosteus osseus Generalist
White bass Morone chrysops Generalist
Smallmouth buffalo Ictiobus bubalus Generalist
Walleye Sander vitreum Generalist

Table 2.  Environmental variables observed synoptically with Upper Mississippi River Restoration Program Long Term Resource Monitoring element’s fish component sampling in the Upper Mississippi River System. Methods associated with recording environmental observations, in highly standardized ways, are detailed in Gutreuter et al. (1995) and Ratcliff et al. (2014).

Variable name Abbreviation Variable type Unit(s)
Environmental variables
Secchi Secchi Continuous cm (nearest 1)
Conductivity SpecCond Continuous µS/cm (nearest 1)
Water velocity Watervel Continuous m/s (nearest 0.1)
Water temperature Temp Continuous °C (nearest 0.1)
Water depth Depth Continuous m (nearest 0.1)
Dissolved Oxygen DO Continuous mg/L (nearest 0.1)
% emergent submersed vegetation AqVeg Categorical (4 categories) %
Vegetation Density VegDens Categorical (2 categories) scaleless
Predominant substrate Substrate Categorical (4 categories) descriptive
Other structures
  Woody debris Woody Binary Presence absence
  Tributary mouth Trib Binary Presence absence
  Inlet/outlet channel InOut Binary Presence absence
  Flooded terrestrial FloodTer Binary Presence absence
  Wing dam/dyke WingDam Binary Presence absence
  Revetment Revetment Binary Presence absence
  Low-head dam, closing dam, weir LowHead Binary Presence absence
Diagnostic variables
Field station Fstation Diagnostic Numeric ordinal label
Period Period Diagnostic Numeric ordinal label

Methods

Given intentions to model the probability of occurrence for 28 fish species within each of the six LTRM study reaches (Figure 1), we chose Multiple Logistic Regression (MLR) with binary responses (SAS version 9.3; Proc Logistic) as our modeling framework. For each species, the probability of occurrence is modeled as a function of 17 environmental variables measured synoptically with fish observations (Table 2). 

Initially, we intended to use data from all six LTRMP stations and develop “systemic models” for each species, resulting in 28 species-specific models. However, it proved largely impossible to get good fits for the systemic models. Thus, we divided the UMRS into regions, and developed an “Upper UMRS” regional model for each species using LTRM study reaches Pools 4, 8, and 13, and a “Lower UMRS” model for each species using LTRM study reaches Pool 26, Open River, and La Grange. This resulted in 56 attempted models. To fit this many models, and gain predictions, we had to simplify the approach and apply it uniformly across all intended models, given constrained resources.

Although we used species presence/absence data, we modeled only presences, or positive observations, and not absences, which may derive from a species either actually being absent, or simply not detected. The LTRM fish component does not collect information to adjust for non-detects in the determination of absences, and other available methods for dealing with this issue required adopting additional assumptions we could not test. Importantly, the goal here is to develop models to predict relative differences in habitat suitability based upon observed presences and their association with observed synoptic environmental data sources, not necessarily an adjusted and more accurate estimate of site occupancy. As such, estimates and predictions arising from these models should be viewed as conservative under-estimates because occurrence probabilities would resolve higher if we could adjust for non-detection. The important results are the relative comparisons and differences among model predictions in space, useful for identifying suitable or unsuitable conditions and considering habitat rehabilitation project siting.

Habitat Maps

Spatially-explicit predictions of the probability of occurrence for each fish species and modeling region were provided by the models reported in Ickes et al. (2014a). These predictions are georeferenced, and are therefore able to be mapped. Work in 2015 used the predictions of Ickes et al. (2014a) to generate maps of the probability of occurrence for each modeled fish species within each of the six LTRM study reaches. Predictions are mapped and provided as (1) point estimates; and (2) interpolated prediction surfaces. The interpolated prediction surfaces were generated using spatial interpolation methods, specifically, the splines with barriers method. This method constrains interpolations from crossing land/water boundaries and was the most conservative method available, as well as the method that introduced the fewest additional assumptions into the predicted mapped surfaces. Maps can be gained for any well-fit model, species, and LTRM study reach using the tables below.

Viewable PDF versions of habitat maps
  Upper Reach Models Lower Reach Models
Species Scientific name Pool 4 Pool 8 Pool 13 Pool 26 Open River La Grange
Black crappie Pomoxis nigromaculatus p04_bkcp.pdf p08_bkcp.pdf p13_bkcp.pdf      
White crappie Pomoxis annularis            
Bluegill Lepomis macrochirus p04_blgl.pdf p08_blgl.pdf p13_blgl.pdf      
Largemouth bass Micropterus salmoides p04_lmbs.pdf p08_lmbs.pdf p13_lmbs.pdf      
Warmouth Lepomis gulosus p04_wrmh.pdf p08_wrmh.pdf p13_wrmh.pdf      
Northern pike Esox lucius p04_ntpk.pdf p08_ntpk.pdf p13_ntpk.pdf      
Yellow perch Perca flavescens p04_ywph.pdf p08_ywph.pdf p13_ywph.pdf      
Blue catfish Ictalurus furcatus       p26_blcf.pdf or_blcf.pdf lg_blcf.pdf
Flathead catfish Pylodictis olivaris p04_fhcf.pdf p08_fhcf.pdf p13_fhcf.pdf p26_fhcf.pdf or_fhcf.pdf lg_fhcf.pdf
Rock bass Ambloplites rupestris p04_rkbs.pdf p08_rkbs.pdf p13_rkbs.pdf      
Skipjack herring Alosa chrysochloris            
Blue sucker Cycleptus elongates p04_busk.pdf p08_busk.pdf p13_busk.pdf p26_busk.pdf or_busk.pdf lg_busk.pdf
Shovelnose sturgeon Scaphirhynchus platorynchus            
Sauger Sander canadense p04_sger.pdf p08_sger.pdf p13_sger.pdf p26_sger.pdf or_sger.pdf lg_sger.pdf
Golden redhorse Moxostoma erythrurum       p26_gdrh.pdf or_gdrh.pdf lg_gdrh.pdf
Shorthead redhorse Moxostoma macrolepidotum p04_shrh.pdf p08_shrh.pdf p13_shrh.pdf p26_shrh.pdf or_shrh.pdf lg_shrh.pdf
Channel catfish
Ictalurus punctatus

         
Red shiner Cyprinella lutrensis            
Logperch Percina caprodes p04_lgph.pdf p08_lgph.pdf p13_lgph.pdf p26_lgph.pdf or_lgph.pdf lg_lgph.pdf
Brook silverside Labidesthes sicculus p04_bkss.pdf p08_bkss.pdf p13_bkss.pdf p26_bkss.pdf or_bkss.pdf lg_bkss.pdf
Freshwater drum Aplodinotus grunniens p04_fwdm.pdf p08_fwdm.pdf p13_fwdm.pdf p26_fwdm.pdf or_fwdm.pdf lg_fwdm.pdf
Emerald shiner Notropis atherinoides p04_ersn.pdf p08_ersn.pdf p13_ersn.pdf p26_ersn.pdf or_ersn.pdf lg_ersn.pdf
Blackstripe topminnow Fundulus notatus       p26_bttm.pdf or_bttm.pdf lg_bttm.pdf
Smallmouth bass Micropterus dolomieu            
Longnose gar Lepisosteus osseus p04_lngr.pdf p08_lngr.pdf p13_lngr.pdf p26_lngr.pdf or_lngr.pdf lg_lngr.pdf
White bass Morone chrysops p04_wtbs.pdf p08_wtbs.pdf p13_wtbs.pdf p26_wtbs.pdf or_wtbs.pdf lg_wtbs.pdf
Smallmouth buffalo Ictiobus bubalus p04_smbf.pdf p08_smbf.pdf p13_smbf.pdf      
Walleye Sander vitreum       p26_wlye.pdf or_wlye.pdf lg_wlye.pdf


GIS IMG files and XML Metadata
  Upper Reach Models Lower Reach Models
Species Scientific name Pool 4 Pool 8 Pool 13 Pool 26 Open River La Grange
Black crappie Pomoxis nigromaculatus p04_bkcp.exe p08_bkcp.exe p13_bkcp.exe      
White crappie Pomoxis annularis            
Bluegill Lepomis macrochirus p04_blgl.exe p08_blgl.exe p13_blgl.exe      
Largemouth bass Micropterus salmoides p04_lmbs.exe p08_lmbs.exe p13_lmbs.exe      
Warmouth Lepomis gulosus p04_wrmh.exe p08_wrmh.exe p13_wrmh.exe      
Northern pike Esox lucius p04_ntpk.exe p08_ntpk.exe p13_ntpk.exe      
Yellow perch Perca flavescens p04_ywph.exe p08_ywph.exe p13_ywph.exe      
Blue catfish Ictalurus furcatus       p26_blcf.exe or_blcf.exe lg_blcf.exe
Flathead catfish Pylodictis olivaris p04_fhcf.exe p08_fhcf.exe p13_fhcf.exe p26_fhcf.exe or_fhcf.exe lg_fhcf.exe
Rock bass Ambloplites rupestris p04_rkbs.exe p08_rkbs.exe p13_rkbs.exe      
Skipjack herring Alosa chrysochloris            
Blue sucker Cycleptus elongates p04_busk.exe p08_busk.exe p13_busk.exe p26_busk.exe or_busk.exe lg_busk.exe
Shovelnose sturgeon Scaphirhynchus platorynchus            
Sauger Sander canadense p04_sger.exe p08_sger.exe p13_sger.exe p26_sger.exe or_sger.exe lg_sger.exe
Golden redhorse Moxostoma erythrurum       p26_gdrh.exe or_grdh.exe lg_grdh.exe
Shorthead redhorse Moxostoma macrolepidotum p04_shrh.exe p08_shrh.exe p13_shrh.exe p26_shrh.exe or_shrh.exe lg_shrh.exe
Channel catfish
Ictalurus punctatus

         
Red shiner Cyprinella lutrensis            
Logperch Percina caprodes p04_lgph.exe p08_lgph.exe p13_lgph.exe p26_lgph.exe or_lgph.exe lg_lgph.exe
Brook silverside Labidesthes sicculus p04_bkss.exe p08_bkss.exe p13_bkss.exe p26_bkss.exe or_bkss.exe lg_bkss.exe
Freshwater drum Aplodinotus grunniens p04_fwdm.exe p08_fwdm.exe p13_fwdm.exe p26_fwdm.exe or_fwdm.exe lg_fwdm.exe
Emerald shiner Notropis atherinoides p04_ersn.exe p08_ersn.exe p13_ersn.exe p26_ersn.exe or_ersn.exe lg_ersn.exe
Blackstripe topminnow Fundulus notatus       p26_bttm.exe or_bttm.exe lg_bttm.exe
Smallmouth bass Micropterus dolomieu            
Longnose gar Lepisosteus osseus p04_lngr.exe p08_lngr.exe p13_lngr.exe p26_lngr.exe or_lngr.exe lg_lngr.exe
White bass Morone chrysops p04_wtbs.exe p08_wtbs.exe p13_wtbs.exe p26_wtbs.exe or_wtbs.exe lg_wtbs.exe
Smallmouth buffalo Ictiobus bubalus p04_smbf.exe p08_smbf.exe p13_smbf.exe      
Walleye Sander vitreum       p26_wlye.exe or_wlye.exe lg_wlye.exe

This study was conducted by the U.S. Army Corps of Engineers’ Upper Mississippi River Restoration Program Long Term Resource Monitoring (LTRM) element. The LTRM is a cooperative effort between the U.S. Army Corps of Engineers, U.S. Geological Survey, U.S. Fish and Wildlife Service, and the states of Illinois, Iowa, Minnesota, Missouri, and Wisconsin. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Literature cited

Gutreuter, S., R. Burkhardt, and K. Lubinski. 1995. Long Term Resource Monitoring Program procedures: Fish monitoring. National Biological Service, Environmental Management Technical Center, Onalaska, Wisconsin, July 1995. LTRMP 95-P002-1. 42 pp. + Appendixes A–J.

Ickes, B. S., J. S. Sauer, N. Richards, M. Bowler, and B. Schlifer.  2014a.  Spatially-explicit habitat models for 28 fishes from the Upper Mississippi River System (AHAG 2.0). A Program Report submitted to the U.S. Army Corps of Engineers’ Upper Mississippi River Restoration-Environmental Management Program from the U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, Wisconsin. January 2014. LTRMP Program Report 2014-P001.  26 pp. + Appendixes A–B.

Ickes, B.S., Sauer, J.S., and Rogala, J.T., 2014b, Monitoring rationale, strategy, issues, and methods: UMRR-EMP LTRMP Fish Component. A program report submitted to the U.S. Army Corps of Engineers’ Upper Mississippi River Restoration-Environmental Management Program, Program Report LTRMP 2014–P001a, 29 p., http://pubs.usgs.gov/mis/ltrmp2014-p001a/

Johnson, B. L., and K. H. Hagerty, editors. 2008. Status and trends of selected resources of the Upper Mississippi River System. U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, Wisconsin, December 2008.  LTRMP Technical Report 2008-T002. 102 pp. + Appendixes A–B. Report available online at http://pubs.usgs.gov/mis/LTRMP2008-T002/

O’Hara, M., B. S. Ickes, E. Gittinger, S. DeLain, T. Dukerschein, M. Pegg, and J. Kalas 2007. Development of a life history database for Upper Mississippi River fishes. U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, Wisconsin. LTRMP 2007-T001. 10 pp. + Appendixes A–B. (NTIS ADA470170)

Ratcliff, E. N., E. J. Gittinger, T. M. O’Hara, and B. S. Ickes. 2014. Long Term Resource Monitoring Program procedures: Fish monitoring, 2nd edition. A Program Report submitted to the U.S. Army Corps of Engineers’ Upper Mississippi River Restoration-Environmental Management Program from the U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, Wisconsin. February 2014. LTRMP Program Report 2013-P001 91 pp. including Appendixes A–G.

U.S. Geological Survey (USGS).  1999. Ecological status and trends of the Upper Mississippi River system 1998. A report of the Long Term Resource Monitoring Program. U.S. Geological Survey, Upper Midwest Environmental Sciences Center, La Crosse, Wisconsin. LTRMP 99-T001. 236 pp.

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USA.gov logo U.S. Department of the Interior | U.S. Geological Survey


Page Last Modified: December 11, 2015 US Army Corps of Engineers USGS Upper Midwest Environmental Sciences Center US Fish and Wildlife Service U.S. Environmental Protection Agency U.S. Department of Agriculture Natural Resources Conservation Service Minnesota DNR Wisconsin DNR Iowa DNR Illinois Natural History Survey Missouri DC