Northern Prairie Wildlife Research Center
Approximately 80% of the sandhill cranes in North America, including most of the midcontinent population, stop along the Platte and North Platte Rivers each spring enroute to their breeding grounds in Canada, Alaska, and the Soviet Union. In recent times, loss of river channel and meadow habitat (Frith 1974; Lewis 1977) has raised concern for the well-being of this population.
Previous studies have documented several aspects of sandhill crane biology during spring in the Platte River Valley (Wheeler and Lewis 1972; Frith 1974; Lewis 1975) but there has been a paucity of information on several key aspects important to fully understanding the role of this staging area in preparing the population for migration and reproduction. Therefore, present investigations have centered on quantitatively describing physiological changes in birds associated with the spring stopover period and monitoring nutrient storage and utilization patterns during migration and reproduction. In addition, studies have focused on defining the actual areas used by the population in the Platte and North Platte River Valleys, on obtaining refined estimates of current population size, and based on a synthesis of numerous types of information, estimating the habitat base necessary to sustain the crane population.
The midcontinent sandhill crane population occupies three major staging areas during its annual spring stopover along the Platte and North Platte Rivers (Fig. 6) plus a small area along the upper reaches of Lake McConaughy near Lewellen. From east to west, the three sites are designated the Kearney - Grand Island Area (staging area 1), the Overton - Elm Creek Area (staging area 2), and the Sutherland - North Platte Area (staging area 3) and encompass 578, 257, and 184 km2 (223, 99, and 71 mi2), respectively; an 18 km2 (7 mi2) area was utilized near Lewellen.
The distribution of sandhill cranes along the Platte and North Platte Rivers is restricted to a few reaches of river and the remainder of the river is uninhabited by the species (Fig. 7). Among the 406 0.8 km (0.5 mile) segments of river channel within the study area, 137 (34%) were utilized as roosts in 1979; 105 of 299 segments of the Platte River and 32 of 107 segments of the North Platte River were occupied by roosting cranes. Crane densities at 18% of the sites along the Platte River exceeded 5000 birds/segment. Highest densities occurred on staging area 1 in the channel area bordering Mormon and Shoemaker Islands. The crane density exceeded 10,000 cranes/segment at two sites.
Sandhill cranes spent the daylight hours on lands relatively close to their roosting sites on the Platte and North Platte Rivers. Photographic surveys of the Kearney - Grand Island and Overton - Elm Creek staging areas indicated that the cranes typically were distributed 1.5-3.5 km (0.9-2.2 miles) from the river channel with more than 90% of them within 4.5 km (2.8 miles). Cranes were not recorded beyond 8.3 km (5.1 miles) from the river channel. Analysis of movements of radio-marked cranes verified the patterns observed from aerial photography. The maximum distance that radio-marked birds moved from the channel averaged 4.2 km (2.6 miles) during the spring staging periods in March and April 1978 and 1979. The farthest perpendicular movement from the river channel of a radio-marked bird was 5.6 km (3.5 miles).
Estimates of the maximum size of the sandhill crane population along the Platte and North Platte Rivers ranged from 306,000 to 541,000 in 1979 and 1980, respectively (Table 2). Population surveys suggest that the staging crane population reaches maximum size in early April and that the lower estimate in 1979 was partially due to the early timing of the census. Estimates for both years are based on photographic surveys and are not directly comparable with ocular estimates made during the past 23 years. Tests to evaluate the accuracy of visual aerial estimates were made during 1978 and 1979; findings suggested that observers tended to underestimate the crane population by approximately 30%. Consequently, the larger population estimate appears due primarily to recent application of more accurate censusing methodology and to later timing of the surveys rather than indicating a sizable increase in the crane population.
Approximately two-thirds of the population is located in the Kearney- Grand Island Area. Most of the remaining cranes used the Overton - Elm Creek and Sutherland - North Platte Areas (Table 2); approximately 4000 cranes occurred on the Lewellen area. Cranes occasionally stop along the Platte River during fall migration but in low numbers in comparison to spring.
Most of the sandhill cranes staging along the Platte River are of 2 races; lesser sandhill cranes occur principally west from Grand Island and individuals of the Canadian race stage principally between Kearney and Grand Island.
Sandhill cranes typically inhabit the Platte River Valley from late February to early May, although most of the midcontinent population is present for only part of this period. The size of the population during March 1979, estimated visually by a team of ground observers during each of four weekly counts on all staging areas, was as follows: 5 March, 82,813; 12 March, 116,943; 20 March, 258,112; and 26 March, 245,715. These data indicate a rapid population increase until about 20 March followed by a leveling off late in the month. The small decline indicated on the 26 March count is probably due to a bias in surveys on one staging area on that date rather than an actual decline in the population. Cranes rapidly depart from the Platte Valley in April when weather conditions are favorable for migration.
Based on information collected from 20 radio-marked cranes in 1978 and 1979, sandhill cranes arriving in late February and early March remain in the North Platte and Platte River Valleys into April along with later arrivals. In 1978 and 1979, a total of 20 cranes were radio-marked and followed throughout their stay at the Platte. In 1978, six of these cranes spent a minimum of 26.2 days on staging area 1 whereas 14 cranes marked in 1979 stayed an average of 31.7 days. The difference in length of stay between the 2 years probably was due to varying weather conditions. In 1979, adverse weather conditions caused the crane population to remain along the Platte until mid-April, a delay of 7 to 10 days from the previous year.
A home range is defined as the total area encompassed by movements of a sandhill crane during the staging interval (Baker 1978). Home range size was estimated using a modification of the computer fill method described by Siniff and Tester (1965).
Sandhill cranes occupied relatively large home ranges during the spring staging period in the Platte River Valley. Home ranges of 20 radio-marked individuals monitored on a 272 km2 (105 mi2) study area near Kearney averaged approximately 36 km2 (14 mi2) and ranged from 12 km2 to 67 km2 (4.5-26 mi2) during early spring of 1978 and 1979 (Table 3).
The cumulative home range represents the entire area utilized by a crane in meeting its needs during the spring staging period. Approximately 39% of the cumulative home range was utilized in any 5 days. Several factors probably contributed to the marked variation in home range size. The distribution of suitable roosting and foraging habitat, the degree of disturbance, and the associated need to shift activity centers are among the most significant factors influencing observed variations in home range size. Habitat characteristics of home ranges used by radio-marked sandhill cranes indicate cranes select areas containing diverse habitats including cropland, native grassland, and hayland (Table 3`). Cranes occupied home ranges that averaged 44% cropland, 20% native grassland, and 10% hayland. Riverine habitat (i.e., former and present channel area) averaged approximately 18% of the home range.
Sandhill cranes exhibited a high degree of fidelity to a certain section of river channel after their arrival on the staging area. Only 2 of the 20 monitored cranes moved out of the Kearney to Shelton reach; 1 moved 18 km (11 miles) to the vicinity of Shoemaker and Mormon Island and the second bird moved westward approximately 23 km (14 miles) to the vicinity of Elm Creek. The average reach of river occupied by individual cranes was 11.8 km (7.3 miles). Cranes frequently shifted roost location from one night to the next; the average daily shift was nearly 1.6 km (1 mile).
The mean distance traveled daily by an individual radio-marked crane was approximately 9.7 km (6 miles). The cranes traveled an average of 2.8 km (1.7 miles) from their nocturnal roost site to their diurnal locations. An average of 5.4 moves was made before returning to the evening roost and each movement averaged 1.8 km (1.1 miles). Diurnal movements decreased during midday, when cranes gathered in large flocks to rest, and increased toward evening. The average distance moved per day decreased as the season progressed, implying a tendency by the cranes to occupy sites that contain the desired combination of habitat needs in close proximity.
During daylight hours, sandhill cranes in the Platte Valley occurred principally in either cropland, native grassland, or alfalfa fields (Table 4). Ninety-seven percent of 2972 diurnal radio-locations of 20 marked sandhill cranes on the Kearney - Grand Island staging area in 1978 and 1979 were in these three habitat types; more than half the locations were in cornfields, primarily grazed corn stubble. Plowed fields, winter wheat, and milo accounted for the remaining cropland use. Crane use of croplands increased from 44% during 1978 to 61% in 1979. Crane use of alfalfa dropped from 27% to 8% over the 2 years whereas use of grassland remained nearly constant. Ninety-four percent of the native grassland locations were in grazed pastures and most of the hayland usage was in mowed alfalfa fields. A typical habitat-use pattern of a radio-marked crane over the entire staging period in 1979 is shown in Fig. 8.
Crane habitat use varied with time of day and over the course of the staging period. Crane use of grasslands was high in the early morning, during midday, and just before dusk, but low in the mid-morning and early evening. Telemetry data supported previous observations of cranes utilizing large grassland tracts located within 0.8 km (0.5 mile) of the river roost sites at dawn and dusk; these sites have been termed secondary roosts by Wheeler and Lewis (1972). Cropland use was highest during mid-morning and mid- and late afternoon whereas intensity of use of alfalfa remained relatively constant throughout the day. Cranes' use of cropland increased later in the staging period and time spent in alfalfa diminished; use of grassland remained fairly constant throughout the season.
Examination of activity patterns on various habitats revealed that while on alfalfa fields, cranes spent approximately half of their time feeding; cranes fed only 39% and 36% of the time on cropland and native meadows. Cranes tend to gather on native grassland to rest during midday. Cranes on grassland spent 24% of their time resting whereas only 12% and 11% was spent in this activity on alfalfa and cropland, respectively.
The diet of sandhill cranes varied among habitats (Fig. 9). In cornfields, cranes ate almost exclusively waste corn; ground beetles and corn borer larvae were present in some samples but only in trace amounts. In contrast, cranes foraging in alfalfa fields and grasslands consumed primarily invertebrates. The proportions of animal food eaten by cranes in grasslands, alfalfa fields, and cornfields were 99%, 82%, and <0.5%, respectively.
Cranes obtained the bulk of their total diet from cornfields. The average dry weights of food samples from cranes collected while feeding in cornfields in 1978 and 1979 were 27 and 19 g; samples from cranes feeding in alfalfa fields and grasslands averaged less than 1 g during each year.
The cranes feeding in cornfields ate principally waste corn, whereas the birds foraging in grasslands consumed earthworms, snails, crickets, grasshoppers, sowbugs, spiders, and adult and larval beetles. The only plant foods consumed in grasslands were grass leaves. Earthworms were the predominant animal food item on alfalfa fields, and alfalfa leaves and stems were the primary plant foods.
Cornfields provide cranes with large quantities of readily digested food of high energy content whereas the grasslands and alfalfa fields are better sources of protein and minerals, in particular calcium. Time-activity and telemetry data were integrated with food habits information to develop a better understanding of the effects of nutritional requirements on crane feeding behavior. These analyses indicate that cranes feeding in cornfields gained energy at 10-20 times the rate of expenditure. On grasslands and alfalfa, however, the cranes used more energy than was ingested and thus temporarily functioned at a net energy loss in order to increase the protein and mineral content of their diet. Therefore, cranes sacrificed time and energy to satisfy nutrient requirements.
Three other lines of evidence suggest cranes cannot obtain all their mineral requirements from corn. First, corn contains very low levels of calcium, i.e., 0.02% (Morrison 1957). Second, more than half of the birds collected from grassland habitats had ingested snail shells which contain 30-40% calcium. Third, some cranes regularly visited a pasture 8 km (5 miles) southeast of Grand Island to eat lime concretions which they dug from the prairie sod with their bills. Five of the cranes which were collected on the staging area during the spring of 1979 had ingested one or more of these lime concretions. The foods from grasslands and alfalfa fields increased the calcium intake by more than 500%. Cranes cannot feed enough hours in cornfields to obtained an equivalent amount of calcium.
Crane foraging patterns are influenced by the diverse nutritional needs satisfied by foods available from different habitats. Analyses of the composite daily diet of cranes indicated that 96% of their food, by dry weight, was obtained from cornfields and 4% from alfalfa fields and grasslands. The 4% of the diet obtained from grasslands and alfalfa fields increased the protein content of the composite diet by 10-20%. The cranes could have obtained an equivalent quantity of protein by increasing their foraging time in cornfields by 10-20%, and avoided the other habitats. However, corn is deficient in certain amino acids, particularly tryptophan and lysine, whereas invertebrate foods contain markedly higher levels. The utilization of grasslands and alfalfa fields for foraging, therefore, suggests that protein quality is an important consideration.
Sandhill cranes deposit substantial fat reserves during stopover periods on their annual spring migration across the Great Plains. The fat content in carcasses of adult lesser sandhill cranes increased from approximately 8% of body weight at departure from wintering grounds in west Texas in mid-February to 23% at departure from staging grounds in central Saskatchewan in late April. Carcasses of adult cranes departing for the breeding grounds from the Last Mountain Lake area of central Saskatchewan in late April and early May contained an average of approximately 900 and 800 g of fat in males and females, respectively.
The Platte River Valley serves as a major conditioning area during spring migration. Adult male and female cranes deposited fat at a daily rate of 13.2 and 9.1 g, respectively (Fig. 10). Assuming a 30-day stopover interval, adult males and females deposited approximately 400 and 275 g, respectively, or 44% and 34% of the total fat reserve acquired on staging areas in the Great Plains Region. The actual magnitude of fat stored by individuals during their stay at the Platte varies directly with the duration of time spent there.
Although the pace of migration varies among years, as cranes adjust to existing weather conditions in the northern Great Plains during April, fat deposition is not slowed by a delayed departure from the Platte Valley. For example, during 1979 most of the cranes remained about a week longer at the Platte than during the previous year and fat deposition continued at a pace similar to the previous month. These data suggest that food resources were ample to allow the cranes to delay their departure date without adversely affecting physical condition and deposition of fat reserves.
When leaving staging areas in the Platte Valley, sandhill cranes migrate northward across the Dakotas but seldom stop for prolonged periods until reaching central and western Saskatchewan and eastern Alberta. The cranes remain in central Canada until early May. The birds are more dispersed on these staging areas than they are along the Platte and wheat replaces corn as the principal energy source. There appears to be limited utilization of fat reserves during the movement from Nebraska to central Saskatchewan because of ample food resources and the slow pace of migration across the plains. However, a substantial part of the fat reserve is utilized after departure from central Saskatchewan and before arrival on the breeding grounds. Based on carcass analyses of cranes collected both at departure from Saskatchewan and arrival on the breeding grounds in Alaska, an estimated 40% of the fat reserve acquired on staging areas in the Great Plains is expended before arrival on breeding grounds in the Yukon-Kuskokwim Delta of western Alaska.
The fat reserves of adult females upon arrival at the breeding grounds averaged approximately 480 g. These reserves were largely utilized during the nesting period (Fig. 10) and were expended at a rate of approximately 6.7 g/day. Cranes were observed feeding during the prelaying and laying periods. Time budget observations indicated that cranes spent 30% and 20% of each day feeding during the prelaying and laying phases, respectively. The presence of substantial fat reserves allows the female to search for foods containing high quality protein needed for egg formation although such a feeding strategy may be inefficient energetically. Males did not undergo a similar decline in fat content during the nesting period.
Modeling food and energy requirements
The food and energy requirements of the crane population were estimated
with a mathematical model of the relationships shown in Fig.
11. First, the existence energy requirements of individual cranes were
calculated with an equation from Kendeigh et al. (1977:143):
The daily corn requirements of the population were calculated from the individual
energy requirements, and estimates of population size and composition (Fig.
11). Daily population size was assumed to increase slowly from 1 March
to a peak on 31 March according to the equation:
where date increases from 1 to 15. Based on the crane census data (cf. Table 2), 69%, 13%, and 18% of the birds were assigned to staging areas 1, 2, and 3, respectively. Measurements of the cranes collected during the feeding ecology research suggest that approximately 50%, 95%, and 100% of the populations in staging areas 1, 2, and 3 are lesser sandhill cranes. Juveniles constitute 11.6% of the population in the model (Buller 1979), and juvenile and adult sex ratios are 50:50 and 55:45 males:females (cf. Lewis 1974:33-34).
Food requirements and availability
The predicted daily corn requirements of individual cranes varied from a low of 0.08 kg/bird for immature female lesser sandhill cranes to a high of 0.15 kg/bird for adult males of the Canadian race. The daily corn requirements for hypothetical maximum populations of 350,000 and 450,000 cranes (Fig. 12) increased from 0.3 metric tons (1 metric ton of corn = 103 kg = 40 bushels) on 1 March to 44.6 and 57.3 metric tons, respectively, on 31 March. The cumulative requirements for populations of 350,000 and 450,000 cranes were 1023 and 1315 metric tons of corn, including 711 and 914, 131 and 169, and 181 and 232 metric tons for staging areas 1, 2, and 3.
The availability of waste corn in the Platte River Valley was sampled on randomly-selected cornfields at the time of harvest, before crane arrival, and after crane departure (Table 5). On each field, all the corn on ears was collected from three random 40.5-m2 quadrants, and all the individual kernels from three 4.05-m2 quadrants within the larger plots. The mean density of waste corn on the fields following fall harvest was nearly 400 kg/ha (Table 5). Interviews with farm owners and managers indicated that yields on the same fields averaged 6214 kg/ha (~ 101 bushels/acre) for a 6-7% rate of loss during harvest operations. The acreage of corn harvested within the study area was taken from our analysis of the habitat characteristics of staging areas.
Statistically significant changes in total waste corn availability occurred between fall harvest and crane arrival, and between crane arrival and departure (Table 5). Nearly half of the available corn was removed during winter when 26 of the 30 fields sampled were grazed by livestock. Changes in corn availability during winter were probably due to livestock feeding because the loss was primarily of ear corn; kernel corn losses were not statistically significant (P>0.05). Conversely, changes in corn availability that occurred during the crane staging period were kernel corn losses; there was no significant decrease in ear corn at that time.
Waste corn availability at the time of crane arrival was 205 kg/ha, and at the time of departure 128 kg/ha (Table 5). Data from the study of mallard winter ecology indicate that 0-25% of the fields in the study area are tilled in the fall. This level of fall tillage would leave a total weight of waste corn present at crane arrival of between 6529 and 8424 metric tons, and at crane departure, between 4077 and 5260 metric tons. The loss of corn during the staging period was, therefore, between 2452 and 3164 metric tons. Using the estimates of population food requirements (1023-1315 metric tons), the cranes accounted for nearly half of the total corn use. Among the other factors affecting corn availability would be continued livestock feeding, and use by ducks, geese, and other wildlife. The data and analyses indicate the cranes used only 10-20% of the corn available at arrival, and left behind a quantity equal to 3-5 times their total requirement.