Thread
Print

[Others 其他] Trends of bird species and population from 1/ 07 - 5/09 鳥種及數目趨勢

Trends of bird species and population from 1/ 07 - 5/09 鳥種及數目趨勢

The following figures are the trends of bird species and trends of bird population.
The survey was conducted from January 2007 to May 2009 on weekly based.
Please feel free to comment and discuss.

George

Note:
‘Bird number / 10’ means the bird number (population) is being ten-divided.

[ Last edited by Webcreeper at 14/07/2009 12:11 ]

Attachment

tpk.jpg (124.66 KB)

14/07/2009 10:12

Tai Po Kau Nature Reserve

tpk.jpg

sm.jpg (123.34 KB)

14/07/2009 10:12

Shing Mun Country Park

sm.jpg

tms.jpg (111.46 KB)

14/07/2009 10:12

Tai Mo Shan

tms.jpg

kt.jpg (103.36 KB)

14/07/2009 10:12

Kam Tin

kt.jpg

TOP

Thanks for showing the data. It looks interesting.

TOP

Two and a half years is rather a short time to be looking at population trends.

Given that, from the 3-year 2007-2009 graphs, there appears to be a general downward trend in bird population (the pink line) in all four locations.
Species numbers (the blue line) appear to have increased from 2007 in at least Tai Po Kau and Shing Mun but this may be just because of your greater familiarity with the territory.

I would be interested in your own comments on these two observations and anything else you think should be pointed out.
Is the decline in population, if it exists, limited to a few species in each place, or a general decline?

Finally, was there an invasion of one or more species causing the large increase in bird numbers in Tai Po Kau at the beginning of this year?

The consistent recording of information is clearly of great value.

[ Last edited by wgeoff at 14/07/2009 17:34 ]

TOP

I Think it is clear that we have a general decline of bird population in all survey sites, from 2007 to 2009.  The decline in Kam Tin is most alarming, and this is clearly related to the deterioration of the site.  The decline is not only in Starlings, but also in grassland species and general species.

I understand that the peak in Jan-Feb 2009 in Tai Po Kau is mainly due to Japanese White-eye.  George should correct if I am wrong.

HF Cheung

TOP

The surveyor should have his objective in conducting the survey and some idea to interpret the data. Solely presenting the graph mean almost nothing to the outsiders.
If HF suggest that there was a major decrease on bird population in Kam Tin, does it imply that there was also a big change of habitat. If this is the point you want to make, you may also present the habitat data as supportive information.
Just look at the graph, it could relate me to influx of migrant, climate change, seasonality, birds and habitat, bird waves....etc.
I think it must be some interesting stories underneath. Please try to be focused on the point you want to make and we will have more new stories to learn.
Thank you.

TOP

This post is to encourage discussion, not as a final report.
I want to tell you that these data is available to us, and people interested can have a look at it.
I myself find these data very interesting and useful, but I do not know everything yet.

HF Cheung

TOP

Appreciated the effort. Here are my comments. Pls do not mind.

It will be better to post the methodology. Readers should know the methodoogy first prior to commenting on the results. Otherwise, we do not know how the surveyor(s) conducted the result and thus not sure whether the survey was conducted scientifically and systematically.

The observer effect (whether the observer(s) hold the same rule during the survey period) is sometimes very crucial when interpreting the data.

Recording birds in forest like Tai Po Kau is even more complicated. Kai set the boundary of 25 to 30 m (not sure the exact value, I may be wrong here) from each side of the survey route as it is near the visual limit of the detectability of birds in forest by a typical human being. Also, whether birds heard are considered. If so, how to quantify them.

The abundance of birds could be varied if these parameters have not been defined. The report should state clearly about this.

If the abundance of one single spp can influence the result, then a table/figure of the spp composition should be presented.

The report should also make reference to previous studies, e.g. Kai's studies about forest birds at Tai Po Kau. All previous studies in urban (By Fiona), shrubland (by Mike Leven), forest (by Kai) used the no of birds/ha. It will be great if the raw data at Tai Po Kau can be transformed to the density of birds as well for making meaningful comparison.

Just showing up and down means the seasonally variation only. Relating it to other abiotic and biotic factors could tell more.

Captain

[ Last edited by wcaptain at 15/07/2009 22:29 ]

TOP

One more enquiry. What is the value of y axis? What does it mean?

Was the survey conducted once a month? If more than once, then we could have a mean value. It looks likely that the data is so fluctated that there is no statistically difference between yrs.

If the surveyor wants to study the annual trend, then it could be either (i) log10 the data or (ii) make an index (like our Hang Seng Index) - set the result of 2007 as the baseline yr and then use it as the reference pt.

"statistics for ornithologists" by Jim Fowler and Louis Cohen. BTO guide No 22. provides a lot of information on handling data.

Although the project title is "the Bird index of HK", the data present here appear not to show any type of index to the readers.

Captain

[ Last edited by wcaptain at 15/07/2009 22:28 ]

TOP

Thank for the interest in the survey.
As I said before, this is not a final report.  Therefore we do not intend to present all informations.
People interested in this can find out more by talking to us.

While I do not want to bore the reader with all the technical details, I will only say that the survey is carried out once a week,o
follow a fixed path at a more or less fixed time of the day, and count all birds seen/heard at all distance.
I am sure there are different opinions on the methodology, but since the method is sufficiently fixed/standardized,
I am satisfied that the method can give good comparison on relative bird abandunce of the sites from year to year.
There are a lot more information than those presented here.  For example, I guess some of you may find the species composition
data interesting.  That will certainly be discussed in the final report.

The horixontal axis of some graph shows a number up to 360 and slightly beyond.  This is representing the day in a year by 1 to 365.  
I hope this is not too difficult to understand.

Go back to the data, do any of you have the feeling that bird population in these sites has decreased in the last few years?
Or do you feel that it is the opposite?  If you have your own data, that would be even better.  Do you mind sharing that with us?

HF Cheung

TOP

Thanks for all friends' comments.

The methodology is given as following.
Part II Methodology (from the survey manual for “Studying Landbirds in Important Bird Habitats of Hong Kong, especially in the IBAs”)
2.1 Line transect technique is used to survey the species (diversity) and number of birds (total population) and to include the best sites in terms of bird density and species diversity.
2.2 Is limited by accessibility: road and footpath are the only choices in this project Whole transect is sub-divided into about 10 sections, so that data from each section are recorded independently
2.3 All species were recorded visually or aurally on the two sides of transect.
2.4 Bird survey will be conducted between 0800h to 1200h or 1400h to 1900h once per week that is around 3 - 4 hours would be spent in each bird survey throughout the study period
2.5 Survey time is based on the peak activity of birds during the pilot survey
2.6 Arrival time at the beginning and ending of each survey sub-section is recorded, so as the weather of the whole survey period
2.7 In summer, bird survey frequency might be reduced to biweekly in case of bad weather, such as heavy raining or typhoon.
2.8 The survey frequency is typically once a week, the survey data is not sufficient to monitor day to day variation.
2.9 Week to week variation could not be clearly seen either, although some particular case studies might be done. Realistically, month to month variation what we can safely expected. That would be sufficient to study the changes in abnormal summer or winter months.

[ Last edited by hgeorge at 17/07/2009 16:04 ]

TOP

Appreciate George's effort of carrying out the survey. Good job.

TOP

Thread