Airsurveys of breeding and wintering Steller's sea eagle at Kamchatka: successes, failures and perspectives

By Alexander LADYGUIN

Science Park of Moscow State University
Russian Federation
Contents: (brief version)
Introduction.
Specific of air survey at Kamchatka
Brief historical review: scope of the areas, covered by helicopter surveys
Technique
methods,
tools and techniques,
strategyes
disadvantages)
Results:
 account of the flight hours, quantity of birds, nests,
distribution of the surveys by seasons
breeding:
- status
- distribution
wintering:
- status
- distribution
Perspective of future monitoring of a populations



Summary

The regular airsurveys of large birds of prey have begun on Kamchatka from the beginning of 80th. For the surveys the various types of air crafts and various techniques and strategies were used, the disadvantages and advantages of which are discussed in the presentation. Now more than 520 SSE and more than 20 WSE nests are mapped, inspected more than 120 ooo square kilometres, the data about preferred by raptors habitats are systematized. The most full inspected are the following areas: a valley of the river of Kamchatka, east coast of a peninsula from cape Lopatka up to Ust-Kamchatsk, the river Elovka basin. If it is possible to recognize results of the surveys of the breeding birds as satisfactory and representative, almost northing known about the status and distribution of wintering birds, that now is the most dangerous blank in our knowledge of raptors ecology. Therefore most urgent actions on monitoring of the SSE population at Kamchatka is the organization of the global winter surveys.
Introduction.

Population of Steller's eagle in Kamtchatka is expected relatively stability, but habitat use by eagle as well as the total account of population is poorly understood . Estimated density of population has increased twice after each more or less seriuos survey, like it has been done by Lobkov at 1986. A better knowledge of habitat use by these eagle , is vital to determine the influence of current and future land-use practices on population trends of this species. Our objective was to made new surveys of SSE in the areas which were not cover by previuos air surveys and to compare habitat characteristics of areas receiving high and low use by Stellers'eagle in Kamtchatka. General goal is to understand the reasons determine the eagle habitat selection and does human impact in Kamtchatka landscapes has effected on eagle population?

Specific of air survey at Kamchatka

The survey of large birds of prey at Kamchatka has some specific which strictly determine methods, technique and schedule of surveys.

  1. First, we have at Kamchatka very bad developed road net: Kamchatka is about 350 000 squere kilometres and the total extension of the roads is about 1300 kilometres, among which just 120 kilometres is covered by asphalt.
  2. Low density of human population which are concentrated in a small local settelments which are very far one from another. It is means that helicopter for survey is actually available just in one point - main city Petropavlovsk and we have to spent a lot of flight hours to reach survey locations.
  3. Landscape specifics: strongly billowy landscape, short mountain rivers, rocky sea coast, which makes impossible another methods of survey exept aerial.
  4. Features of SSE nestling on Kamchatka, where their nests are authentic less, than on a continental part of breeding area, and allocate in a dense part of crown, that make quite difficalt their recognition both from the helicopter, and from the ground.
  5. Climate conditions: long winter, very short spring determined that SSE nestling begins at the end of winter.
    These last ones make the period for airsurvey quite short - just twi-three weeks during first week of hatching till the foliage appearence.
 Fig1. Location of the helicopter bases and main people settlements in Kamchatka

Fig4. Percent of recognised eagle nests at the model area in a different seasons		 The indicated features with necessity dictate: usage of the helicopter for the surveys, unacceptability of alternate methods of the surveys, a very short and strictly limited time assigned for the surveys - prior to the beginning of the appearance of foliage on trees, which, as have shown special researches, not only hides contents of the nest, but also increases mistakes of the survey at this time of the year up to 60%.

Productivities of the winter surveys (I mean the time of the year with snow cover, which still present at Kamchatka at the beginning of June) comes nearer to 100%, summer and even the spring surveys made on the model splots have shown, that their efficiency approximately have 65%. High percent of mistakes during summer surveys depends on the difficultes of recognition nest withing canopy.

Brief historical review
The surveys of large birds of prey at Kamchatka have begun from a beginning of 60th years. First estimation of the SSE was published by Ostroumov and was based on the detailed observations of about 100 river basins during salmon spowning period in July-September 1960-1966. SSE in this study were counted incidentally because of the main task of this project was to count salmon at a spowing sites.
The special surveys of birds of prey by the ornithologists were conducted at the end of 70th by Kronotskiy State Reserve scientists. At the end of 80th annual surveys became regular, the survey area was expanded far behind the Reserve territory - especially in a basin of the river Kamchatka (Fig 1.). At this period first chek-list of Steller's eagle nests was composed. At the 1989 first issue of the annual check-list contains about 320 nest, include 220 living ones. But more then 40% of these "known" nests were recorded by post perquisition of hunters, fishers and some other forest and wildlife service staff members . So, these data were collected from just amateur observers and there is absent any detail descriptions of habitat range and eagle nest itself.
The surveys at eastern Kamchatka were most full conducted because of its relative availability, as the city Petropavlovsk-Kamchatskiy is located here. For period from the 1984 to 1994 by different scientists about 800 flight survey’s hours were made.
The extent of the surveys paths has made about 100 000 square kms. In result, about 350 SSE, WSE and osprey nests were taken into account (that, in my opinion, is an extremely low productivity of the airsurveys during ten years), totally about 8300 meetings with raptors were noticed.

Nevertheless, some areas, due to their extreme inaccessibility, has appeared almost not covered by these surveys. This is the area of western Kamchatka from the river Vorovskaya to the north and north-eastern part of Kamchatka from the line between the river Tigil and river Ozernaja.
Valley of Kamchatka river (special surveys were conducted here during 3 years 1985-1987) and the east coast from Lopatka cape up to Ust-Kamchatsk were the most detail investigated.
At the beginning of the Gorbachev’s perestroika the budget of any govermental offices was disastrously truncated, as a result, also any regular surveys were completely stopped. Nevertheless, due to the efforts of several Russian scientists special surveys were supported in 1995, 1996 and 1998. The most successefull surveys (due to the good choice of the time of the year, technique and selected areas) were ones, supported by WWF in 1996. The total budget allocated of Kamchatka’s part of this project has made 6000 thousands of US dollars. If to take in mind that one hour of small helicopter was cost about 600 dollars, it would be about 10 hours of flight. It is enought for flight from Petropavlovst to Kronotskiy Reserve and back three times, but it is northing if you would like to survey eagles.
However, due to collaboration with another WWF team, which surveing bears, totally we flight about more than 50 hours. In result 196 new nests of SSE were found and the area of more than 12 000 square km which have been not covered with the surveys of all previous years was inspected.

Fig2. The map of summer air surveys have done at Kamchatka during 1984-1998
I'd like to notice, that all this results have respect to breeding, summer status of SSE.
What about winter SSE allocation, the special International survey was made by Dr. Lobkov and his Japanese colleagues at 1985 both at Kamchatka, Sachalin, Magadan and Japan. At Kamchatka totally about 900 birds were surveyed. Also, as I said about summer survey, the large part of information about wintering ranges was obtained from amateur observers. But since this survey nobody undertake something similar during 14 years.

Technique
During last decade at Kamchatka a different methods and different strategy of surveys were used. As was shown, the land and motor boat surveys are extremely ineffective on Kamchatka due to their exclusive complexity. In the beginning of 80th years in Kronotskyi reserve were applied double during a season land surveys of a model population in territory about 12 000 square kilometers. However, was remarked, that the visiting of the nests in spring period, that is in the beginning of period hatching has an extremely negatively effect for final success of reproduction of birds, resulting to increased of eaglets dearth. Therefore in subsequent it was necessary to refuse from land survey of the nests in the spring and made it from helicopter. Then, the result of breeding success in the nests examined in the spring, was checked in August by land survey. Such strategy, first, save our time and force, because we know which one nest is living before land visiting (which is difficalt in Kamchatka landscape, of course), second, save money for second helicopter survey, third, we have opportunity to check eaglets, describe the nest, food, remains, ets.
The application of such strategy looks the most productive in conditions of the limited financing for the helicopter survey.
Actually, the technique of the air survey did not differed from common practice, unless by that we had no possibility to use neither GPS tools, nor any navigating systems, nor computer simulation of the processes occurring in a population. Normally, we used small helicopter Mi-2 and fly with speed about 150 km/hour 50 m above the land.
All nests were mapped at the mape 1km in 1 cm scale and describing.

Results
Breeding status

Totally in different years more then 525 nests of SSE are accounted. More than 60 characteristics both the nests and surrounded habitats were used for nests' description. At the moment these data are in processing, but we hope, that in the nearest future we can use the database for understanding the features of habitats which determine the choice by birds for nestling and other preferences.

The main surprise of 1996 survey was the extremely high nest density at the two plots: Elovka river and at the north-west coast of the Peninsula.
Living nests were concentrated exectly along narrow river basin and were sometimes just 200 m from each other. Totally at Elovka were estimated 58 living nest per 100 km and 48 living nest per 100 km were account at North-west (absolutely it was 29 nest at Elovka and 39 nest at north-west). In compared, at east coast we estimated 15 living nest per 100 km.
Kamchatka river basin has a low density of eagle population due to high level of different kinds of human activity. Totally, about 45 pairs of eagles (both SSE and WSE) nesting here.
Distribution during nestling.

The types of habitats structure were determined on the base of the following characteristics, which were used for computer analisys of habitat's preferences:

A. Used landscape type (macrohabitat pattern) [investigated area in a square km] 1. low river valley [751] 2. high river valley [45] 3. top of the river valley (platou) with birch forest [347] 4. tundra with certain forest islands [81] 5. flat tundra [93] 6. sea shoreline [218 km] 7. old mixed coniferous taiga [541] 8. birch forest outside river basin [219] 9. mursh land [421] 10. mountain tundra [387] 11. rocky mountain [372] 12. agricultural land [291] 13. alpine zone with low bush 14. large old volcanic-glacier lakes(Kuril,Kronotskoe)[125] 15. old voulcano caldera with forest and water reservouir within [24]

B. shore line landscape:

1. Rocks and clifs used by sea birds
2. High latitude rocky coast with short litoral and covered by the bush on the platou, usually with numerous small river
3. Rocky coast without woodland and river valley
4. The sloping mountainside covered by bush
5. Flat tundra or woody coast with bank
6. Flat grassland or mursh with(or without) certain small wood island
7. Shallowed gulf and lagoon:
- with wooded coast
- with mursh coast
8. Grassland and mursh in basin of large river with the net of small streams , lakes and creeks.

C. litoral bottom characteristics

D. vegetation on the shoreline within 300 meter strip

- single species, mixed
- age
- continuosness: continuos forest,
fragmented, certain forest lands

E. type of the water reservoir: sea, lake, river, salt lagoon

F. feeding site:

- distance from the nest site,
- presence/absence of tidal currents,
- flood pattern
- transparency of water
- presence/absence of underground springs
- available prey list
- presence/absence salmon spowning grounds
- food availability and abundance
- period when food is available
- avarage size,
- depth,
- landscape of bottom,
- ice condition
- abundance of perch and place for roosting
- protection against strong whether conditions

Fig. Nest distribution of SSE at Kamchatka
(click on the map
to expand it)

G. type of nestling river valley:

- deep wide valley covered by old forest
- deep narrow valley covered by old forest
- flat murshland valley with certain wood island
- flat tundra and mursh land without any vegetation
- flat tundra valley with poplar forest band by the
stream

H. nest location:

1. on the tree
tree characteristics:
- species,
- class: dominant, subdominant,
- age: old, middle, young
- quality: live, dead, crashed
- distance from the forest boundary
- distance from the water boundary
- distance from the feeding site
2. on the rock

I. stream pattern

1. salt or fresh
2.absence/presence(number) of salmon spowning grounds,
3. physical characteristics
- depth
- width
- stream speed
- transparensness
- open/wood covered
- presence/absence of gravel bars, small open islands
- bottom feartures
- temperature

L. Coast pattern


- flat or breaked coast line,
- absence/presence of small islands,
- absence/presence lagoon,
- steep-walled coast,
- wooded/bare

Habitat characteristics were compared using single-classification analysis-of-variance and Mann-Whitney rank sums test. Stepwise logistic regression was used to determine which habitat characteristics best distinguishied between used and unused areas (and high and low used areas) for all revealed habitats combined and for each habitats separetely. A characteristics were used in the regression model if comparisons between used and unused areas produced an F-statistic of P<0.01. To reduce redundancy with other characteristics , only one pair of highly correlated (P<0.005) characteristics was used in the model.



Use of landscape.

The first step of the research was to determine the number and the features of eagle distribution in variable Kamtchatka landscapes The strong selected (i.e. habitat selection occurs) both nesting sites and wintering locations were showed (table 1).


Used landscape type:
1. low river valley
2. high river valley
3. top of the river valley (platou) with birch forest
4. tundra with certain forest islands
5. flat tundra
6. sea shoreline
7. old mixed coniferous taiga
8. birch forest outside river basin
9. mursh land
10. mountain tundra
11. rocky mountain
12. agricultural land
13. alpine zone with low bush
14. large old volcanic-glacier lakes(Kuril,Kronotskoe)
15. old voulcano caldera with forest and water reservouir within
We found, that unusually for other large eagle Steller's tend to form "collective" nestling and form winter aggregation in a local habitats. At Kamchatka Steller's eagle is usually strong "communal" bird without any trends to dispersal distribution.
The most preferable both for nesting and wintering are low river valley landscape type and rocky sea shoreline. Additionally for wintering the large old volcanic lakes is the most important.

The next step of the research was to determine the reasons of non dispersal distribution of eagles. There are two common hypothesys:
1. Mosaic distribution is the result of strong demand either to the nest sites or to the food base location.
2. The next one reason is the differentation in the level of human activity in a different habitats.

Nest site use.

Within landscape type eagle shows a high variation in nest site use. We estimated both all nest sites characteristics combined and separately within each type of landscapes.
This analysis shows that eagles use for nesting any sites which are available, i.e. there is no any selection of nest site withing selected landscape type.


Nets tree selection

Surprisently, nest tree selection was not found in Stellers eagle, like it is normall for another raptors. There is no some preference neither in tree size, nor age, nor tree species, nor tree construction.

Other nest site characteristics include nest mycrohabitats feartures and some preference were found here. The following characteristics were significant in selection mycrohabitats:
- distance from the forest boundary
- distance from the feeding site ( but this characteristic was more variable)
- presence of salmon spowning grounds

When shoreline was selected as nesting site, the significant preference of coast with gulf and lagoon was showed. There was no significant preference in any other type of shoreline.
There are no some significant preference in C, D, E characteristics (litoral bottom, vegetation on the shoreline, type of the water reservoir), i.e. Steller's eagle demonstrait a very high diversity in used habitats and their characteristics.

So, we can conclude, that inspite of strictly landscape selection ocurrs, eagles select home ranges in a very different types of habitats with high diversity of physical characteristics.

Patterns of prey abundance does not effected spacing below the landscape level. Spasing between pairs varied unpredictable and without some correlation with food resource and habitat structure. The matter is Stellers use feeding sites communally.
The reasons for communal feeding are the local food dispertion and high weight of salmon, which eagle are not able to carry to the nest and have to consumpting at feeding site in conjunction with others.

Certainly, there is still a lot of whitespots, but I can resume, that the SSE breeding status and distribution we understand full rather. I believe, that in the nearest future we hardly should obtain any data, which could change this picture essentially. Of course, we can supplay more or less new information to the database or conduct the survey at new areas, add in a map some more tens slots, but it is basicly nothing will change in our principal understanding of SSE distribution.


Winter status

What about winter status of SSE at Kamchatka, the last more or less representative winter survey was conducted at 1986, 13 years ago. Since this survey we northing know any news about winter distribution and status af SSE in peninsula. Also, I have to say, that this survey covered just 5 percent of territory potentionally attractable for eagles.
My researchers were conducted just at the Kuril lake basin and nobody knows something about other some millions of square kilometres of Peninsula.
Main part of winter allocation of SSE is assembled in area of Kuril lake, where annually concentrates about 500 birds. It is characteristically, that this probably biggest SSE concentration in the world was opened only in 1985, though the first information about it are available in the beginning of this century! Summarise, we northing know about status and the character of residence in winter of other then 2,5-3 thousand birds.
Meanwhile, some facts can intrigue us. So, it is known, that in winter SSE forms temporary, fast breaking up aggregations in a local spots of temporary appearance of a plentyful forage, for example, winter spowning of smelt or Salmo salvelinus. These temporary agregations break up as fast, as well as arised, but can include up to 200-300 individuals.

Where are these thousand birds arised from, how they find out the appearance of new food locations and where they disappearing after food source run down is an enigma for us.

The absence of the information of winter eagles' allocation is extremely alarm fact, the main blank spot in our knowlege of the SSE ecology.



Fig3. The map of winter air survey made at Kamchatka during 1985-1998 (click on the map to expand it)
Winter distribution

Wintering and nestling locations in Steller's eagle coinside with nestling ares only in part. Most of adult and subadult eagles undertake the wide seasonal migration in fall and spring in searching available food sites. Migration schedule is unregular being depends on season food abundance and distribution and fluctuate year by year.
This is means that the same postnestling foraging sites changes their attractive feartures in a different years, and as a result, their preference by eagles is changing permanently. Eagles winter preference depends on food available, which flucluate seasonally, which, as we postulated, is mostly result of within, preyself population process, but not depends on physical characterisrics of habitats, which are usually does not change.
So, we focused investigation of wintering habitat distribution on that feartures, which can characterize habitats as potentially attractive for eagle.
Number of eagle fluctuate at the same habitats in dependence of concret food abundance in current year. In this case the annually survey would show that eagle use the same habitats in different mode although the physical characteristics of habitat were not changed. So, such method could not answer what feartures of a concret habitat exactly attract eagle. More other, using such approach it is not possible to understand what characteristics determine the availability of food. Probably the last characteristic is the most important because of even food is abundance, but not available, eagle would not occupy such habitat.
So, to understand trends in eagles distribution during the concret year, which is permanently is changing, we have to understand their prey dictribution peculiar properties.

Meanwise, the complecated correlations with and within wintering feeding sites characteristics were found. The following characteristics are significant in distingushing preferable feeding sites (inportance is decresing during this rank):
- salmon spowning grounds presence
- period when food is available
- transparency of water
- underground springs presence
- average size
- depth
- ice condition (non frost during winter)

Perspective of future monitoring of the population
The most important sort of data we need for population monitoring, is the status of birds at protected area, to control population we have to know how much, where and when birds are located. Now I can say, we are controlling breeding status, but we are not able to control about 2000-3000 (and even does not know how much exactly !) of eagles, which are ramblering somewhere during 7 winter month.

Studying during the last ten years the various sides of biology of this bird - behaviour, ecology, energetic, morphology, - I was convinced, that the most critical periods of life circle of this bird is the wintering. The condition in wich birds start reproduction period is the most limiting factor both for eaglets surviving and for the ability of adult to catch enought food (to be in a good condition) for them. So, I gess, when planning the future monitoring plans, we have to take into consideration these points.



If the main our objective is to find the real and working method to control eagle population, we have to run new strategy of our researches.
I support the idea to organise the long-term all year round investigations of the factors limiting SSE surviving. It must be new level of investigation, comprehensive and complexly.

The following acts I am consider as the high-priority for future monitoring of SSE population at Kamchatka:

First, It is impossible to grasp immense, Kamchatka is too large to cover all the territory by helicopter surveys both at winter and summer. So, we have to use the model research base station for long-term research. The Kronotskiy Reserve territory looks the best for this goal, first of all, because of there is population which is under control during last 20 years here and we have the detail information about the life of more then 60 nests. Also, it is good known area with developed research station and wide net of local cabins. Also, at this territory a long term research of Pacific salmon are made by Salmon Institution.

What kinds of reserach this project must include:

I gess, we have to focus our efforts at the free kinds of problems:
  1. Detail research of SSE life cycle and relationships with the main prey life cycles;
  2. Detail research the living requirements for the habitats, nests, hunting range (I gess, it is almost clear for us);
  3. Understanding the eagle's living strategy (I mean first of all the enigmatical notunderstandable winter distribution.
I gess, that the most important for SSE population at Kamchatka monitoring would be the following:
1. The new global winter airsurveys.
2. Mass marking of eaglets.
3. Marking the adult birds on wintering sites both at Kamchatka and Japan.
4. I am not support the idea to continue summer airsurvey.