BIRD SURVEY TECHNIQUES By Asad R Rahmani Based

BIRD SURVEY TECHNIQUES By Asad R Rahmani Based on the manual prepared by Salim Javed & Rahul Kaul Indian Bird Conservation Network (IBCN) Bombay Natural History Society

WWhy Count Birds? Fundamental questions Elemental arithmetic tendency to count Necessary to devise a strategy and new plans Doing something in a systematic manner Interesting and necessary to know more about birds

Where to count? Any area – garden, orchard, park, forest, wetlands, grassland, desert Unstudied area Important staging area Threatened area Sanctuary or a national park Research site(s)

When to count birds? Breeding season Winter counts Seasonality and timing of count is very important During migration

How to plan a survey Permission from forest department/ relevant authority Make local contacts Logistics i. e. places of stay, routes etc Necessary funds Area maps: Toposheets, vegetation maps, aerial photos Literature (checklist of birds and plants, reports, research papers, working plan) Data sheets, pen, field note book, pad Equipment: Binocular, camera, compass, altimeter, GPS Fieldguides, reference books and papers

Which site(s) to select We have to ask the following questions: What is the purpose of census? IBA or EBA Sanctuary or National Park Size: logistics, time frame, budget and the size of the team

Strategies A. Non-sampling strategy Total count in an area e. g. Siberian cranes in Keoladeo Nest count e. g. Greater Adjutant stork colony Territory mapping e. g. Bengal florican display grounds

B. Sampling strategy Counting a small representative population and D. then extrapolating about the total population e. g. random, stratified or systematic random C.

Important points to remember Do not unnecessary collect large data Maximize your efforts Determine sample size Collect right type of data

To determine effective sampling efforts, first answer Two major questions 1. How many samples to take? 2. What should be the plot size so that most species are covered ? To answer these questions, we need to plot Species Discovery Curve. Species Discovery Curve is plotting of number of species detected or discovered per unit of sampling efforts (length of transect, time spent walking a transect or standing on a point).

Species discovery curve showing a cumulative total of species seen in riparian habitat in Dudwa National Park, India over a 30 day period

What is distance sampling? Distance sampling involves collection of data where distances of objects are estimated or measured. e. g. Line transects Point Counts Cue counts (calls, territory)

What is Line Transect ? Line transect is based on theory of walking along a predetermined route at a regular interval to record the Birds on or near the line.

Methodology or Study Design A. Site Selection Random or systematic Stratified : covering different habitats within the study area Stratified random: randomly choosing areas in different habitats

B. Where can one monitor Line Transects? Best in open, flat habitat Also conducted in hilly areas Homogeneous habitat preferred C. Where to place Line Transects accessibility and terrain straight line, not zig-zag avoid along roads, streams or contour of hills well spaced out (minimum 200 m apart) random, stratified or linear (hilly terrain)

D. Permanent or Temporary transect If permanent - mark with stones or colour posts (trees) divide the transects in to 40 -50 m segments E. What should be Transect Length? It varies according to species, habitat and aim of study single species study community study rare or common species diversity of habitat How to determine transect length? preliminary checklist few test runs species area curve

What is Species Area Curve ? Species area curve is drawn by plotting the sighting frequency of birds with increasing transect length. After a point, species discovery curve tends to stabilise, which means that with subsequent increase in the line of the transect there is little or no corresponding increase in the new species being added. The point at which curve flattens out (asymptote) can then be considered as adequate for sampling birds.

35 S P E C I E S 5 100 200 300 400 500 600 700 101 Transect Length in meters Species area curve for obtaining the length of the transect

E. What should be the speed of travel? walk in standard pace (about 8 -10 m/minute) time duration should not vary more than 10% among transects F. What time of the day? preferably morning 15 -25 minutes after sunrise continue for 2 -3 hours G. In which weather condition? sunny weather avoid raining or windy days keep tract of weather condition of study area

H. How often (periodicity) ? weekly or fortnightly or monthly (intensive study) seasonal (long term study, 5 -10 years) once or twice a year (very long term study, 20 -25 years or more) I. How many replicates ? Minimum two replicates of each transects Minimum of 6 monitoring in each season J. Open width or fixed width ? Open width: All birds are noted irrespective of their sighting distance Fixed width or Belt transect: Birds seen up to a certain distance (width on either side of transect length) are only noted

Biases in census counts Observer bias Effect of habitat Bird behaviour Weather How to do analyses of the collected data? A. Simple method : n D= -------- 2 L x Y n= total number of individuals; L=length of the transect; Y= mean perpendicular distance B. Computer programmes: 1. TRANSECTS-II 2. DISTANCE

How to determine the width distance? It depends upon the birds being censused and type of habitat. Data collection What to record ? number of individuals of a species perpendicular distance sighting angle and sighting distance sex (male, female if possible) age (adult, juvenile) activity (singing, foraging, flying, etc) substratum (ground, bush, tree etc) If the bird is not seen but heard, records its call and try to judge the distance.

15 I ND I V I D U AL S 2 Perpendicular Distance Frequency histogram of perpendicular distances (number of individual detected decrease with increasing distance from the line)

Question: What is perpendicular distance? Answer: Perpendicular distance is the distance of the bird from the transect line. Should we record the exact distance? Recording exact distance is difficult (e. g. 11 m, 16 m) Therefore, record in group intervals (0 -5 m, 5 -10 m, 10 -15 m).

PD= Perpendicular Distance SD= Sighting Distance O= Observer PD SD L= Transect Line L = Object Recording of perpendicular distances and sighting angles in line transect sampling

The Fourier series (FS) estimator used in the analysis is the expansion of probability density function (pdf), f (x). Fourier series estimator is a robust non-parametric procedure in which the difference in detectability between different habitats is taken care of by pooling robustness of FS estimator and its estimation efficiency. n F (0) D = --‑‑‑‑‑‑‑----- 2 L n = Total number of bird groups seen; L = Length of the transect F (0) = Probability density function

N= No of objects L= Length of transect Y= Mean perpendicular Distances 10, 12, 15, 5, 10, 25 N = 6 L = 500 Y = 12. 8 D = n / 2 LY = 6 / 2 x 500 x 12. 8 = (6 / 12800) x 10000 = 4. 6 birds / ha A simple way of calculating density from ungrouped perpendicular distances

Advantages of Line Transect More economical Greater species turnover · Larger area is covered in relatively shorter time Applicable throughout the year Permanent transects can be monitored for a longer period of time Can be used in most of the habitat types (except wetlands) With little care can also be used for hilly terrain · Disadvantages of Line Transect · Distances are not correctly measured · Movement of observer may disturb the birds · Chances of missing skulking or shy birds are great

Assumptions of Line Transect No bird is missed Transect is a straight line Every detection is independent Points/objects are fixed at initial sightings and they do not move before being detected

Precautions to be taken: Try to keep line as straight as possible Make special effort to find those birds that are close to the transect Distances should be measured correctly Transect should be long enough to allow detection of at least 40 individuals Transect should be representative of a habitat Observer must be interested, competent and trained

Point Count Method Point Counts can be imagined as transects of zero length conducted at zero speed. Types of Point Count: 1. Plot Counts (fixed radii for all species) 2. Point Counts with variable radii, by species 3. Point Counts with unlimited radii (Total Counts)

What are the assumptions? birds do not approach the observer or flee. all the birds are detected at the point of the observer. birds do not move much during the count period birds behave independently of one another. distance estimates are accurate birds are fully and correctly identified.

Advantages of Point Counts less time consuming duration can be controlled total attention to detect birds small homogenous habitats can also be studied Disadvantages of Point Counts generating bird list is slower many species are missed area sampled in one unit is small

How to select points for counting ? Points to be counted are to be laid out systematically or selected randomly in the study area. Points should ideally be 200 meters apart to avoid double counting. In a small area, lesser points should be monitored to avoid duplication In small patches of habitat, inclusion of points near the edges should be done with caution depending on the objectives of the study.

What should be the Count duration? It can vary from 2 -20 minutes Record only these birds seen during Count duration. Do not include for analysis those birds seen while walking between two points. How far Point counts should be done ? Not very far (> 200 m) Not very close (<20 m) Ideally 50 to 100 m apart

At what time it should be done ? 15 -20 minutes after sunrise Evening counts can be done How to record distance ? Record bird’s distance from the observer (yourself) Record distance in categories (0 -25 m, 25 -50 m and >50 m)

Data recording 1. Number of individuals of each species detected within a 20 -25 m radius surrounding the observer. The radius will vary depending upon the habitat type. • 2. Number of individuals of each species detected beyond the 25 m radius but still within the same habitat. • 3. All individuals detected while the observer walked between count points are recorded but the data are not used in the analysis. Species recorded during this period will contribute to the completeness of species list for the site. • 4. Birds that originally were detected outside 20 -25 m radius boundary but later move within 25 m of the observer are recorded as occurring within the fixed radius circle. This facilitates comparison among vegetatively different habitats. Objective is to count each individual bird once and only once.

Analysis density, species diversity, richness and species composition can be calculated enter large data in EXCEL and LOTUS sample data matrix can be generated using SPECDIVRS. BAS. follow the instructions given in statistical books and/or take help of a good bio-statistician

POINT COUNT Date Locality Altitude Habitat Observer Salim Lat Long Slope Weather Sheet # Time Start Time End Visibility Aspect Sr. No. Bird Species Time No. Sex Age Distance Height Activity Plant spp. Hgt. Remarks 1. Blue Whistling Thrush 0710 2 - 10 m Foraging - 2. Minivet 0810 4 - -- 5 m 1 m Foraging - 3. 4. Example of a data sheet for point count used in Ranikhet, Kumaon Himalayas

Comparison of transect and point count census methods __________________________________________________ S. No. Items Transects Point Counts Duration ______________ Short Long __________________________________________________ 1. Saturation of observer with cues from bird 2. Birds moving into and out of the range 3. Speed of generating species list 4. Birds missed by flushing 5. Skulking birds missed 6. Total count/unit line 7. Attention divided 8. Area sampled in one unit 9. Bias 10. Precision Little Few Fast No Yes Same Yes Large Small High Some Medium Yes No Same No Small High Much Many Slow Yes No Same No Small High __________________________________________________ Desirable method= Small bias and high precision

Species Richness Methods Species richness methods are simple methods of counting or generating species list and then making useful interpretation from them. Depending upon the area, type of survey, whether single species or multi-species, a species richness method can be conducted. Types of species richness methods 1. Encounter Rates 2. Mc. Kinnon’s Species Richness Method 3. Timed Species Count 4. Mist netting

1. Encounter Rates species seen per unit efforts (time, distance etc) useful for single species or multiple-species surveys data gives only relative abundance not density It gives number of birds/unit area or number of birds/unit time that can be compared with other habitats/areas/seasons.

Data sheet for collecting encounter rate data DATA SHEET _________________________________________ Date Locality Habitat Data Sheet Time Start Time End Weather Visibility Aspect Slope Altitude Coordinates Observer SJ Transect/Trail Length Transect # Sheet # _________________________________________ Bird Species Time Flock Size Habitat Sex Age Activity _________________________________________ Roseringed Parakeet 0645 2 MF M A Perching Redvented Bulbul 0650 1 MF SA Feeding Green Bee- eater 0700 4 MF A Perching

A common method of presenting encounter rate data ________________________________________________________________ Sites No. of birds No. of hours Birds/hour Transect Length Birds/km ____________________________________________ Site 1 2 5 2/5 3 km 2/3=0. 6/km Site 2 3 2 3/2 2 km 3/2=1. 25/km Site 3 5 3 5/3 2 km 5/2 = 2. 5/km ____________________________________________

How to present multi-species survey data? Calculate encounter rate of each species Enter data in EXCEL or LOTUS Present data in descending order or classification-wise Categories the encounter rate (very common 20 -25 sightings; common 15 -20 sightings; etc)

Encounter rates from a multi-species survey from a standard one hour walk at each site ________________________________________________________________ Species Site 1 Abundance Site 2 Abundance _______________________________________ Rose-ringed Parakeet 5 5/hr 2 2/hr Ring Dove 2 2/hr 1 1/hr Green Bee-eater 0 0/hr 2 0/hr Emerald Dove 4 4/hr 1 4/hr Red breasted Flycatcher 1 1/hr 0 0/hr ___________________________________________________________ Common: 5 -10; Very Common: 10 -25; Uncommon: 1 -5

Advantage of Encounter rate data quick and easy method, especially in hilly terrain easy calculation data comparable across sites/seasons/habitats diversity and richness can be calculated

2. Mc. Kinnon's Species Richness Method developed by Mc. Kinnon & Philip (1993) in Indonesia. quick and easy to know richness useful for rapid surveys in difficult habitats useful for multi-site surveys

What is the methodology? walk in an area till a given number of species are recorded the number of species could be 10, 15, 20, 25 or 30 but it should be constant for all the sites once a given number of species(e. g. 20) have been seen, go to another area and record the same number of species (i. e. 20). prepare 10 to 15 such lists from different parts of the study area common species will get recorded in several lists. in species rich areas, listing will be very fast note starting and ending times, habitat condition, weather etc.

Analysis By plotting the cumulative total of species detected against the number of lists, species discovery curves can be produced. This species discovery curve for each site is a measure of species diversity and can be plotted to compare several sites If additionally numbers of individuals are also recorded these values can be used to compare species diversity based on discovery curve but also indices of richness and diversity. Number of time a species reappears in subsequent lists can be converted into frequency of occurrence, which gives some idea about relative abundance of that species when compared with other species. An index of relative abundance can also be generated by dividing the number of lists a species appears in by the total number of lists. An index between 0 -1 is produced for each species.

Advantages of Mc. Kinnon’s Species Richness method The method is simple, quick and does not require any special observer skills and equipment. A very useful method for multi-species survey. Can be used in difficult terrain. Can be used for different habitats. Precautions Each list should be independent Lists should be made away from each other so same individuals are not counted again. For different habitats, separate lists should be prepared. Number of lists should be according to available habitat.

Data sheet for recording species __________________________________________________ DATA FORMAT FOR Mc. Kinnon’s LIST __________________________________________________ Sr. N. List 1 List 2 List 3 ________________________________________________ Species No Species No Species No __________________________________________________ 1. Roseringed Parakeet 1 Collared Bushchat 1 Small Blue Kingfisher 1 ………………. …. . 2. Ring Dove 2 Rubythroat 1 ………………. …. . 3. Little Brown Dove 1 Black Drongo 1 ………………. …. . 4. Green Bee Eater 5 Jungle Crow 1 ………………. …. . 5. Collared Bushchat 1 Greyheaded Flycatcher 1 ………………. …. . 6. Pied Myna 2 Shikra 1 ………………. …. . 7. Purple Sunbird 1 House Crow 2 ………………. …. . 8. Jungle Babbler 2 Purple Sunbird 2 ………………. …. . 9. Crow Pheasant 1 Pariah Kite 1 ………………. …. . 10. Purple Sunbird 1 Greyheaded Flycatcher 1 ………………. …. . 11. Greyheaded Flycatcher 1 Grey Tit 2 ………………. …. . ………………. …. . 12. Redbreasted Flycatcher 1 Common Myna 1 ………………. …. . ………………. …. . 13. Chiffchaf 1 Golden Oriole 1 ………………. …. . ………………. …. . 14. Black Drongo 1 Redwattled Lapwing 1 ………………. …. . ………………. …. . 15. House Sparrow 2 Brahminy Myna 2 ………………. …. . ………………. …. . 16. Golden Oriole 1 Indian Roller 1 ………………. …. . ………………. …. . 17. Brahminy Myna 2 White Eye 3 ………………. …. . ………………. …. . 18. White Eye 5 Whitebreasted Kingfisher 1 ………………. …. . ………………. …. . 19. Yellowlegged Green Pigeon 8 Tailor Bird 1 ………………. …. . ………………. …. . 20. Tailor Bird 1 Roseringed Parakeet 3 ………………. …. . ………………. …. . __________________________________________________

3. Timed Species Counts (TCS) Principle Common species are generally detected earlier than the rarer forms, therefore the time taken to detect a bird forms a measure of its abundance. Methodology Take observations for a certain time, say one hour. Divide one hour in six 10 minute time periods All the species seen in the first 10 -minute time period are recorded. Then go to another site and record species not recorded earlier. Repeat this six times, 10 minute period each Never record a species twice. Make a minimum of 10 -15 visits to the study area at different seasons/months. During each visit, try to cover at least 1 km 2 area

Analysis All species recorded are ranked according to their time period. Thus, species recorded in the first 10 minute interval are ranked 6 followed by 5 for the species recorded in the second 10 minute interval and so on. Unrecorded species are ranked 0. An index of relative abundance of species is calculated as the mean score for each species across all survey visits to the site. Therefore scores between a maximum value of six and a minimum value of 1/n (n is the number of repeated surveys) are obtained.

Advantages simple, easy and quick amateur birdwatcher can do it Disadvantages provides only crude relative indices of abundance. comparisons of different species within area or between areas can be made but these need to be interpreted with caution because of differential detectability of species in different habitats/areas. flocking species may end up with lower indices compared to the more widely dispersed ones because flocked species may not be recorded in subsequent time-periods once the flock has been recorded

Hypothetical data set to demonstrate the use of TSC method 0 -10 mins 10 -20 mins 20 -30 mins 30 -40 mins 40 -50 mins 50 -60 mins Yellowbellied Fantail Flycatcher White cheeked Bulbul Kestrel Himalayan tree creeper White Crested Kalij Beautiful Nuthatch Jungle Crow Long-tailed Minivet Strong footed bush warbler Yellow rumped leaf warbler Streaked Laughing Thrush Collared Bush chat Grey Tit Verditer Flycatcher Large Hawk Cuckoo Slaty headed parakeet Himalayan Griffon Black Drongo Tailor bird Green backed Tit

Ranking of the above dataset using Timed Species Counts ________________________________________________________________ Species Visit 1 Visit 2 Visit 3 Total rank score Mean rank score Species rank ________________________________________________________________ Yellowbellied Fantail Flycatcher 6 3 2 11 3. 66 10 Jungle crow 6 6 18 6. 0 1 Collared Bush chat 6 4 1 11 3. 66 10 Slaty headed Parakeet 5 4 3 12 4. 0 9 Tailor Bird 5 4 1 10 3. 33 14 White-cheeked Bulbul 5 4 14 4. 66 5 Long-tailed Minivet 5 4 4 13 4. 33 7 Grey tit 4 5 4 13 4. 33 7 Hmalayan Griffon Vulture 4 3 4 11 3. 66 10 Green backed Tit 4 5 5 14 4. 66 5 Kestrel 4 3 2 9 3. 0 16 Strong footed Bush warbler 4 5 6 15 5. 0 3 Verditer Flycatcher 4 6 6 16 5. 33 2 Himalayan tree creeper 3 5 3 11 3. 66 10 Yellow rumped leaf warbler 3 0 6 2. 0 17 Large Hawk cuckoo 3 6 6 15 5. 0 3 White crested Kalij 2 0 3 8 1. 66 18 Beautiful Nuthatch 1 0 0 1 0. 33 20 Streaked Laughing Thrush 2 1 4 1. 33 19 Black Drongo 3 3 4 10 3. 33 14 ____________________________________________

4. Mist netting Principle By trapping birds in a standardized way, bird populations or communities of different sites can be compared. Methodology a fixed number of nets are operated for a fixed period of time at different sites. the birds caught by these nets are recorded, measured, ringed and released data are recorded in a format results are expressed as birds caught/net/hour

Analysis many ways of analysing data variations in population size and structure over years or between sites can be detected. species diversity and similarity indices can be generated to study the bird communities at various sites survival rates and productivity can also be examined using this technique by calculating the rate of adults to juveniles caught after the breeding period Beside this, we also get longevity, survival, site-fidelity, moult, breeding and migration data.

Advantages good method for detecting shy, skulking birds very rich source of information confirmed identification photographic evidence available very good learning process Disadvantages labour intensive exercise problems of permissions costly and time consuming ringing training and licence required not good for short rapid surveys birds become net shy Biases Results can be influenced by the way the nets are laid and the location where they are laid and therefore comparisons between sites may be difficult unless sites selected are truly random.

Why count during breeding season? birds become comparatively conspicuous most birds sing or display most birds have territories during breeding season movement become restricted so easy to count many birds come back every year to same area for breeding some birds nest in colonies hence easy to count WWhat are the methods of counting breeding birds 1. Territory mapping 2. Call counts. Nest counts

1. Territory mapping most birds, especially males defend territories song and display can be easily detected and pinpointed territorial males can counted and mapped density of singing males per hectare/sq. km can be calculated D A G J A B H G A B A H G F E C B C C C Territory mapping for Painted Francolin from calls

Advantages provides good data calculation very easy can be graphically illustrated repeatable every year/season provide information about bird-habitat association Disadvantages time consuming and expensive needs expertise and patience can be done in breeding season only could disturb the nesting birds Assumptions Birds live in pairs in non-overlapping territories

2. Call Count many bird species call during the breeding season. good for Galliformes species call can be counted from a strategic place can be repeated every season can be compared across habitats and sites

Methodology most species have distinct calls species can be identified by calls observers sit 300 -500 m apart and note every call direction of call and time are also noted before starting, observers synchronise watches observations starts early morning when most birds call counts should be stopped 15 -20 minutes after the first call is heard

Analysis observers sit together and compare call locations duplicate records are deleted after the records have been pruned, the minimum number of birds calling is noted density of calling birds is estimated from the total study area

Advantages, disadvantages and biases excellent method for vocal but skulking birds non-invasive method and does no disturb the birds simple and does not require much equipment easy to monitor trend for a number of years birds can be stimulated to call by call play-back fairly accurate estimates of abundance indices

Disadvantages difficult to identify call and distance by inexperienced observers counts have to be repeated 4 -5 successive mornings to account for the large variation in calling of birds may also be over-estimated if call counts are conducted for long durations birds tend to shift their locations after initial calling and there is a chance that a bird might be counted twice if call counts are prolonged. day to day variation in number of calling males density estimation of calling males or pairs only, not of non ‘breeding individuals

N Observer K 2 T 1 K 3 K 1 T 2 Call count sheet commonly used for counting pheasants

3. Nest Counts over one-eighth birds nest in colonies nests or pairs of a colony can be counted or estimated nest site fidelity is seen so birds come to same area every year breeding population can be estimated easily comparative data can be obtained across years/sites photography can be used for estimation/counting in small colonies, total nest can be counted in large colonies, sample surveys can be done

Advantages easy and interesting generates good statistical data generates media interest in conservation easy to monitor from year to year Disadvantages may disturb the birds if not done carefully the whole colony may not be approachable nests in dense colonies are not easy to count

Caution Most species of birds are sensitive to disturbance during the breeding season, so extreme caution should be taken while counting the nests or pairs. It is preferable to count the nest from a distance. Nest count can also be made after the breeding is over (e. g. flamingo, pelicans). A bare minimum time should be spent near a nesting colony. Most counts should be done in the morning or evening, and never during hot mid-day. In protected areas, first get permission from the forest officials.

If you cannot follow rules, DO NOT COUNT