Grande Ronde Sub-Basin Wild Spring Chinook

Introduction

The fish in this study are wild yearling chinook that were tagged in their rearing grounds in the Grande Ronde River basin in northeast Oregon. The fish were seined and tagged in their natal river during the autumn of the year previous to their outmigration. Thus the tagged fish overwintered before initiating downstream migration. Achord et al. (1996) provide details of the fish capture and tagging techniques. We combined stocks from several tributaries in close proximity (see map below) to generate sample sizes large enough for the analysis.

Since the actual date of migration initiation is not known, observed travel times from the release site to the first observation site (Lower Granite) are not available for these fish. Beginning in 1993, though, many of the detected PIT tagged fish were subsequently diverted back into the river (instead of being tranported to below Bonneveille Dam). Thus many individuals were detected at both Lower Granite Dam and one of the downstream observation sites. These "double detections" result in observed travel times of the wild fish through portions of the mainstem rivers.

Map of the Grande Ronde Basin and migration route

Data

We have examined wild chinook pittag releases from the Grande Ronde Sub-Basin, (hydrological units 17060104, 17060105, and 17060106), during 1993-1997. Individual fish observed at Lower Granite and at least one additional site downstream were identified and grouped into cohorts based on their date of initial observation at Lower Granite. Travel times were calculated to subsequent sites of observation. Only three PITTAG sites yielded enough observations for anaylsis: Little Goose, Lower Monumental, and McNary. Tagged fish were observed at Lower Granite from mid April to the beginning of June.

Models

We applied six nested models of varying complexity to the data:

Model details are provided by Zabel, et al. (1997)*.

• Model 1 uses a constant migration rate.
  
• Model 2 incorporates a linear flow relationship.
  
• Model 3 adds a seasonal/flow interaction - fish use more of the flow for migration as the season progresses.
• Model 4 adds an experience factor - flow independent migration speed increases the longer the fish have been in the river.
• Model 6 removes the seasonal component from model 4 and becomes
  
• Model 7 adds a linear length dependent term to model 1:
  
• Model 8 adds a linear length dependent term to model 2:
  
  where is the slope of the length-migration rate relationship and is the average length of cohort i

* In Model 4, a sixth parameter,(₂), describing the slope of flow-independent term at its inflection point, is allowed to vary.

Graphical Results

The plots below show observed average travel times versus modeled average travel times to each of the three observation sites for all six models.

corresponds to observations at Little Goose Dam.

corresponds to observations at Lower Monumental Dam.

corresponds to observations at McNary Dam.

Tabular Results

Table 1 . Travel-time parameters for spring chinook cohorts constructed from fish initially observed at Lower Granite Dam. 25 cohorts with a minimun of 15 fish were examined from 1993-1997.

model	parameter estimates (std. error)								resid. ss	mult. R2
	MIN	LGTH	MAX	FLOW		TINFL	2	2
1	10.6217 (0.4054)	-	-	-	-	-	-	78.54	719.361	-
2	2.1551 (0.8063)	-	-	0.9984 (0.1109)	-	-	-	78.54	480.283	.3323
3	2.4444 (0.5785)	-	-	0.5180 (0.0395)	0.2470 (0.1062)	103.519 (1.6445)	-	78.54	392.555	.4543
4	2.8252 (0.3832)	-	16.3170 (2.0341)	0.9313 (0.3584)	0.0113 (0.0027)	252.00 (37.789)	3.6786 (438.17)	78.54	228.167	.6828
6	2.3350 (0.5592)	-	14.9373 (1.8660)	0.4393 (0.0868)	-	-	3.6900 (76.891)	78.54	240.472	.6657
7	9.8783 (0.4571)	0.0088 (0.0024)	-	-	-	-	-	78.54	721.091	-0.002

---

Discussion

Based on model 1, the average migration rate from Lower Granite to McNary was 10.6 miles per day. The linear flow model offers a moderate improvement over model 1 with an R-squared of 0.332. The seasonal flow model (model 3) slightly improves upon the linear flow model. With this model and the fitted parameters, the fish exhibit a slight tendency to use less of the river flow for migration early in the season. Model 4, which add a downstream acceleration term improves the R-squared to 0.683. A reduced form of this model (model 6), which removes the seasonal flow term, only slightly decreases the R-squared while reducing the parameter standard errors. This is the recommended model for CRiSP model runs for this stock. With model 7, migration rate is linearly related to fish length. This model conferred no improvement over the simple model (model 1), and so we conclude that length is not an important factor in terms of migration rate for these fish.

References

Achord, S., G.M. Mathews, O.W. Johnson, and D.M. Marsh. 1996. Use of passive integrated transponder (PIT) tags to monitor migration timing of Snake River chinook salmon smolts. North American Journal of Fisheries Management 16: 302-313

Zabel, R.W., and J.J. Anderson. 1997. A model of the travel time of migrating juvenile salmon, with an application to Snake River spring chinook. North American Journal of Fisheries Management 17:93-100

Zabel, R.W., J.J. Anderson, and P.A. Shaw. 1997. A multiple reach model to describe the migratory behavior of Snake River yearling chinook salmon (Oncorhynchus tshawytscha). accepted by Canadian Journal of Fisheries and Aquatic Sciences