Virtual Museum: open for refreshments!

As you drive your new car out of the showroom, its value drops dramatically, and then keeps on dropping. It’s not quite as bad with the Virtual Museum. But, from the time it is uploaded, the “value” of every record does slowly decrease. A record in the Virtual Museum has three components: species, place and date. The record is evidence that a particular species was recorded at the place on the date. But, as that date recedes into the past, the record becomes less and less valuable as evidence that the species STILL occurs at the site. The record needs to be “refreshed”.

This blog aims to answer two questions. (1) How do I find when a species was last recorded at a place? (2) What is the rate at which a record loses value through time, and how often does it need to be “refreshed”?

We first need to define what we mean by a “place”. For Virtual Museum purposes this is a  “quarter degree grid cell” (QDGC).  These are almost exactly square on the equator, about 27 km north to south and 27 km east to west. At the northern and southern limits of Africa, about 35°N and 35°S, a QDGC is still 27 km north-south, but has shrunk to 23 km east-west. Not quite square, which is why we talk about “cells”. There is nothing magic about the choice of the QDGC as “place”. But this unit of area has long been used in biodiversity mapping in Africa, especially southern Africa. There are about 2,000 QDGCs in South Africa, and about 50,000 in Africa as a whole. If you define the “place” as some smaller unit, then you have to worry about refreshing records in an even bigger number of places. The QDGC represents a convenient trade-off between a fine grid and a manageable number of grid cells.

So the first question boils down to: “How do I find when a species was last recorded in a quarter degree grid cell?” The blog will help you find the six-character code for grid cells south of the equator in Africa. The trick described there is to google the name of the place and ask for the coordinates (eg search for “Kuruman coordinates”, to try to find coordinates in the format of decimal degrees (Kuruman is 27.450°S 23.433°E) and then follow the instructions in the blog to find the code for the QDGC (2723AD)).

To get a list of the scorpions of Kuruman, the incantation is

and you will find that (at the time of writing this blog) one species is listed, Pseudolychas ochraceus, photographed on 8 November 2018 (see

But the focus of this blog will be on QDGC 2528CA, which covers central Pretoria, and the area northwards just west of the N1 (“Pretoria North”). It is an area characterized by rapid and relatively recent urban sprawl:

This map is available at

The map is the same no matter what you put for “vm=” in the incantation. This section of this blog is going to focus on the reptiles for this grid cell, so I chose vm=reptilemap.

Below the map is a list of the 78 species of reptile recorded here. The top 14 lines look like this:

The columns are self-explanatory. For the purpose of this blog, only one is important: “Last recorded”. This gives the date of the newest record of the species in the grid cell. The dates, quite frankly, are alarming! In row 8, the last record of Boomslang was in 1990. That is three decades ago. Does this snake still occur here?  It almost certainly does, in spite of the pressures of development. But it still needs to be formally “refreshed” and confirmed by the submission of a new photographic record.

In this list of 14 species, the oldest date in the “last recorded” column is for the Dusky Worm Lizard. There are two records of this reptile, the most “recent” from 1911. This highlights the fact that the ReptileMAP database includes all the museum specimen data, going back to the year dot, assembled for the SARCA project. These old museum records are really valuable in pointing out what species we should be on the lookout for now.

The “youngest” record is for the Southern Tree Agama, last “refreshed” on 10 February 2016 (curated at There are 18 records of this species in this QDGC; you can see them by clicking on “Records” at the end of the line when you have this “live” on your screen (it won’t work on this screenshot).

Here are the last eight species on the list of 78 for QDGC  2528CA, Pretoria North:

It tells us that Puff Adder was last refreshed on 10 September 2018, less than a year ago. This is the most recent of 59 records, and can be found at

The bottom row is key. It tells us that there are a total of 1,271 reptile records for this QDGC. That is a lot of records. Then come two dates. The top date, with the single asterisk, is the median of the 78 “Last recorded” dates. So half the species were last seen before 14 December 1988, and half the species after that date. 1988 is a long time ago. There is a massive need to refresh the reptiles in this QDGC.

How do some of the other sections of the Virtual Museum compare on this criterion? To get the dragonflies and damselflies from OdonataMAP, the incantation used above changes to

The map is the same as the map above,  but the species list looks like this:

There are only 10 species. The median date of the most recent records is 4 February 2017, which is excellent. The Little Wisp has a “Last recorded” date in 1999. This is a specimen record: – it has no photograph. It is a candidate to be “refreshed”. (The lower date, with two asterisks, 22 November 2016, is the median date for all 16 records.)

Here is the LepiMAP incantation:

It shows that there are 2,390 records of butterflies and moths for the QDGC 2528CA, and that 178 species have been recorded here. That is awesome. But the median date for the “Last recorded” column is 8 January 2009. That is a whole decade ago. To keep this information up-to-date, there are lots of opportunities here for species to be refreshed!

For BirdPix, the incantation

shows that there are 118 records of 58 species, and that the median date of “Last recorded” is 4 September 2013. That is six years ago. There is a general need even here for refreshment. Please explore these ideas for the QDGCs and species groups that interest you. The pattern of the incantation to the website always has this format:

You need to change lepimap to the section that interests you, the 2528CA to the quarter degree grid cell which you want to explore.

The recommended fieldwork strategy in the QDGC is to have a target list of “long-in-the-tooth” species that you want to refresh, but to grab every opportunity that comes your way. You can pre-empt the need to refresh species by keeping the entire data base “young”.

It would be fantastic to be able to produce “up-to-date” distribution maps for species using data from, say, the most recent three years. Here is an amazingly encouraging pair of maps:

The top map shows the distribution of the Common Dotted Border, using 2,483 records since 1 January 1980. It includes historical data from museums and private collections assembled during the SABCA project. In QDGCs with only historical data, the shading is grey. If there is photographic Virtual Museum data,  the shading is green. The bottom map is based on 555 Virtual Museum records submitted in three-and-bit years since 1 January 2016. The lower map is inevitably sparser, but it is identical in overall pattern. This is a remarkable achievement by the citizen scientists of Team LepiMAP!

These two maps illustrate the value of keeping the database refreshed!

(The inset illustration of this butterfly is one of the most recent submissions of a Common Dotted Border. The photograph was taken on 29 June 2019 by Neil Thomson in the Waterberg,  Namibia – see It is one of three records of the species from QDGC 2017AD, and is a bit to the north of the two Namibian records shown on the maps.)

Now we need to tackle the second question. How quickly does a record lose value? It would need a workshop of biodiversity experts to provide a good answer, but here is a first stab at this. Suppose a record has value 100% at the time when it is made. The “value” is the strength of the evidence that  the species occurs at this date and place. After three years the value might be 80%.  After five years, the value might drop to 50%, after 10 years to 10% and after 20 years, the record might have no value at all. In other words, the fact that a species was recorded in this grid cell 20 years ago is useless as evidence that I can still expect it to persist there.  These suggestions can be turned into a graph, with the gaps between the values above joined by straight lines:

Not everyone would agree with this precise curve, but the general shape is likely to be right. In an era with unprecedented rates of development and climate change, records of biodiversity need to be “refreshed” at regular intervals to provide ongoing evidence of the persistence of a species at a locality. For the Virtual Museum, we would love records to be refreshed before three years have elapsed. This would enable us to generate up-to-date distribution maps, such as the one above for Common Dotted Border.

The Virtual Museum is open for refreshments.


Quarter-degree grid cells made simple

Lump it or leave it, quarter-degree grid cells are entrenched in the mapping system of South Africa. In an article called “SA Mapsheet Referencing“, the government tells us that “each map of the National Map Series is identified by its unique number (e.g. 2830CB) … “. Quarter degree grid cells are also deeply entrenched in the biodiversity mapping of South Africa. The first bird atlas project in South Africa was the Bird Atlas of Natal, which collected data from  1970 to 1979. We have no idea why the coordinators of that project, Digby Cyrus and Nigel Robson, selected the quarter-degree grid cell system, but most subsequent biodiversity mapping (= atlasing) projects in South Africa followed their lead.

Often we know (or can easily use Google to find out) the coordinates of a place, but we need to know the “code” for the quarter-degree grid cell (QDGC) the place falls into. One of the objectives of this blog is to enable us to do this.

But first we need to know how the system works. The code for a QDGC has four numbers and two letters (and only A, B, C and D are used). 2830CB is an example. First, I will attempt an explanation in words, and then use two diagrams.

The four numbers, 2830, tell us which degree square we are in. 2830 is the code for the degree square with 28°S and 30°E in its northwest (top left) corner. Next divide the degree square into quarters. Call the two on the top row A and B, and the two in the bottom row C and D. Top left is A, top right is B, bottom left is C and bottom right is D. So the C in 2830CB tells us that this QDGC is in the bottom left corner of the degree square; call it section C. Repeat the process one more time. Split each quarter degree square into four, and label them A and B, C and D, in the same way. So the B in 2830CB points us to the top right quarter of section C. It’s a messy explanation, and there must have been far simpler ways to do this. But we are stuck with this “official” system, and we have to come to  grips with  it!

In diagrams, it is far easier to explain. These are the official diagrams.

This is degree square 2830.

This diagram shows how degree square 2830 is divided firstly into four quarters, and how the first level of letters works. Then the bottom left quarter, section C, is subdivided for the second time, and quarter-degree grid cell 2830CB is shaded. Easy.  Now that you have seen the diagrams, read the explanation in words again!

There are 16 quarter-degree grid cells in a degree square. But the lines that demarcate the grid are at 15′ (fifteen minute) intervals. and 15′ = a quarter of a degree. So it is a quarter-degree grid, and we talk about quarter-degree grid cells (where you are free to hyphenate in any way you choose!).

Suppose now you want the Virtual Museum‘s list of the Lepidoptera, the butterflies and moths, which have been  recorded in QDGC 2830CB. The incantation you need to do this is

When you go to this website, the information you get looks like this, but it is bigger:

2830CB turns out to be in rural KwaZulu-Natal. The village called Tugela Ferry is in the southeastern corner. This website provides a map of the quarter-degree-grid cell, and a list of the species of Lepidoptera recorded. There are 16 so far. Unpacking the rich amount of information on this page is going to be the topic of another blog. This incantation works for all grid cells and for all the sections of the Virtual Museum. The two places where you need to make changes are in green:

You need to customise the project, and chose the QDGC (called a “locus” in the incantation).

But this is the rub. Suppose I want a list of the scorpions of Calvinia. How on earth do I find the code for the QDGC that Calvinia falls into? You start by googling “Calvinia coordinates”. You get an embarrassment of riches – the answer comes in three formats: 31°28′30″S 19°46′22″E / 31.47500°S 19.77278°E / -31.47500; 19.77278. The easiest one to work with is the second one, in decimal degrees. Calvinia is 31.475 degrees south and 19.773 degrees east. Split these numbers into whole degrees and the decimals (i.e. 31 and 0.475, and 19 and 0.773, rounding off to three decimal places). Combine the degrees (31 and 19) into a string of four numbers. We are in degree square 3119. Next I take a piece of paper, and do this scribble:

The fractions are in decimals (and NOT minutes). The left hand edge is the north-south part. The north-south fraction is 0.475, which lies between 0.25 and 0.5. So Calvinia is in one of the cells along the second row of the scribble (AC, AD, BC, BD). The east-west fraction is 0.773, which is bigger than 0.75 (along the top edge). So Calvinia is in the fourth column. The intersection of second row and fourth column is BD. So the code for Calvinia’s QDGC is 3119BD. Let’s give this a whirl for scorpions. The incantation is

It delivers this information

Gosh, there is only a single record of a scorpion recorded from this grid cell in ScorpionMAP. Fieldwork is needed!

What QDGC lies one to the west (left) of 3119BD Calvinia? The quickest way to find this is to go back to the scribble. It is 3119BC. But what QDGC lies one to the east. It is beyond the edge of the  scribble; it is in the next degree square, one to the east of 3119. This must be degree square 3120. Looking at relative positions of the QDGCs in the scribble, the cell to the east of letter code BD must have code AC. So the QDGC one to the east of 3119BD  is 3120AC. There MUST have been an easier way to do this business of giving codes to the QDGCs!

The paragraphs that follow deal mainly with the complications of being in the northern hemisphere. Most readers can skim to the end.

A Norwegian geographer, Ragnvald Larsen, visited Africa. He thought our quarter degree grid cell system was the best thing since sliced bread, and devised a way to apply it to the whole world. If you google “QDGC”, the second item is his enthusiastic blog post. The first item is a Wikipedia article, in which the author(s) think of the QDGCs as “tiles” that cover the earth’s surface. This is quite a nice concept.

But there are two problems. The first problem gets really serious when you reach about 60°N or 60°S. Quarter-degree-grid cells are not really “squares”. In the north-south direction, every QDGC is 27.4 km long. On the equator, they are also 27.4 km wide. They shrink in width as you go north or south, because the lines of longitude get closer together, and converge at the poles:

This is what a QDGC looks like in northern Greenland at 83°N! In the far north, and south, QDGCs get narrower and narrower because the lines of longitude all converge to meet at the poles. It would be unthinkable to do a bird atlas in northern Europe using quarter-degree grid cells. In this extreme example, in northern Greenland, the QDGC is still 27.4 km north to south, but a sliver of land 4 km wide from east to west.

But at the northern and southern ends of Africa, for example at Cape Agulhas, they are only a little narrower, at 23.0 km. So, throughout Africa, QDGCs are, for all practical purposes, approximately square.

The second problem is that codes like 3119BD assume that the 31 is south and the 19 is east. The system makes no provision for north or for west, and does not admit to the fact the east and west longitudes can be three digit numbers, as they are in Australia! Rene Navarro devised a universal system.

As an example, have a look at record, a butterfly from Ethiopia. It’s in grid cell which he has called NE_010037BC. This means that it is 10°N and 37°E. Looking at the scribble above, the BC means that the record is between 10.25°N and 10.5°N and 37.5°E and 37.75°E.

It is agony to think about it, but the actual layout of the scribble in the northern hemisphere, and in the western hemisphere has to be different. If you are north and east, like in Ethiopia, AA needs to be in the bottom left corner, because north is increasing from bottom to top. If you are north and west, like in Senegal, AA needs to be in the bottom right corner, because both north and west are increasing in the opposite direction to what it does in the south and east! It’s messy.

For  some light relief at the end of this blog, here is the butterfly from Ethiopia mentioned above. It is curated at and was recorded in grid cell NE_010037BC. It is a Citrus Swallowtail Papilio demodocus, submitted to LepiMAP by Tesfu Tujuva on 9 April 2019. This butterfly has a range throughout Africa south of the Sahara Desert, and then extends northeast through the Horn of Africa as far as Oman in the Arabian Peninsula. Go to to see where this QDGC is, and to find out whether other species of butterflies and moths have been  recorded here.

The QDGC system has proved exceptionally valuable for the purpose of generating maps of the distribution of species in southern Africa. We would love the maps to be on a finer scale, but for most groups of species (except the birds), the quarter-degree grid system is the best we can achieve with the available data. It is likely to prove useful throughout Africa. This is why we have extended the naming system so that it can be used throughout the continent (and even worldwide).

From the official mapping perspective, we have no idea who invented the system for naming the quarter-degree grid cells. But it has been in use for almost a century, since the first quarter-degree grid cell maps were produced by the section of government known at the time as “Trig Survey.” The system has stood the test of time. We have no choice but to get to grips with it.

Exploring data: the median and the mean, and everything in between

Here is the assignment for this blog. “Write a report on the progress being made in the Western Cape by the BirdPix section of the Virtual Museum.” The report needs to communicate to the citizen scientists who participate in the project. It needs to provide them with insights into how well the project is doing.

This map shows one aspect of progress. It provides the number of records submitted to BirdPix per quarter degree grid cell. There are lots and lots of numbers. This is not anecdote; this is real data, from which recommendations need to be made.

I could write something like this.  One long paragraph could start “Grid cell 3018DA Kliprand in the far north has 32 records,” … later on it would say … “grid cell 3318CD Cape Town has 1322 records,” …  and it would end by saying … “grid cell 3420CC at Cape Agulhas has 67 records.” This paragraph would fill up a few pages with utterly boring and useless text. It is just providing essentially the same information as is presented a lot more effectively in the map. What is needed is some sort of a summary of the data. I need to convey the overall picture, and not get bogged down in detail.

In general terms, the first task of any statistician is to summarize lots of numbers down to a tiny handful of numbers. The message in the data cannot be accessed by reading every number, or by simply eye-balling the data. There are just too many numbers to absorb. To extract succinct stories out of data is the role of the statistician, .

Now this blog is supposed to be a tutorial, and not a full scale data analysis, so I will illustrate the ideas with a subset of the Western Cape; this map goes from Langebaan to Cape Town, and inland.

The first thing to determine is the sample size, the number of numbers. This is 23. The statistician would write n=23. Statisticians have a convention that they reserve the letter n for sample size. Woe and betide any statistician who comes along and uses n for any other purpose. Mathematics is full of these little conventions and rules; if you know these secret codes, equations can often be understood far more quickly.

The sample of size 23 is big enough not to be trivial.  But it is small enough to be manageable. Here are the numbers, copied row by row off the map: 20 430 2 43 13 9 12 67 57 93 64 33 16 72 258 484 44 59 1322 1022 132 54 86. These are the numbers of BirdPix records per gridcell.

The first (and obvious) thing to try is the mean,  also called the average. We have known how to calculate this since we were schoolkids. Add the numbers together, and divide by the number of numbers. In this case, it is 4392/23=191.0. So the mean is 191.0 records per grid cell. We do this bit of trivial arithmetic, and move on to the next task. But, hey, let’s stop and look at this more carefully. Does 191.0 make sense? Does it really communicate what is going on in the sample? A little thought shows that the mean is doing a ghastly job of summarizing the data. Only five of the 23 values are larger than the mean: 258, 430, 484, 1022 and 1322. And the remaining 18 are smaller than the mean. The arithmetic is perfectly correct, but somehow it doesn’t make sense. The mean does not really communicate where the “middle” of the data really lies.

Statisticians have a strategy for dealing with this problem. They simply sort the data, and pick the number in the middle. When they are sorted the 23 numbers look like this: 2 9 12 13 16 20 33 43 44 54 57 59 64 67 72 86 93 132 258 430 484 1022 1322. The number in the middle is 59. There are 11 numbers which are smaller than 59 and 11 which are larger. This number in the middle has a technical name. It is called the median.

We are trying to communicate how well BirdPix is doing. In this situation, the median, 59 records per grid, communicates the reality far better than the mean, 119.0, does. The mean is biased, pulled upwards by the two grid cells with more than 1000 records. In contrast, the median is unfazed by “outliers”. If the largest number in the dataset was 13220 instead of 1322, the impact on the mean would be dramatic (it would change to 708.3), but there would be no impact on the median. The number in the middle remains 59. There is a technical term for this property of the median. In their jargon, statisticians say that it is robust against outliers.

There is a formula for finding the “rank” of the median. Once the numbers are sorted, the median has rank (n+1)/2. With n=23 numbers the median has rank (23+1)/2 = 12. It is the 12th largest number.

This works fine when the sample size is an odd number. But if n is even, there’s a problem. Suppose n=24. The the median has rank (24+1)/2 = 12½. The trick is to use the average of the pair of numbers in the middle of the sorted sample as the median. The median would be defined as the average of the 12th and 13th largest numbers in the sample of 24 numbers.

We now go back to our sample of 23 sorted numbers: 2 9 12 13 16 20 33 43 44 54 57 59 64 67 72 86 93 132 258 430 484 1022 1322.

The smallest number is 2, the largest number is 1322, and the median, the number  in the middle, is 59. A more subtle question is to ask to what extent the numbers are concentrated around the median, or are they spread out towards the two extremes. One clever way to get a handle on this is to compute the medians of the top half  and the bottom half of the data.  In broad brush terms, the “lower median” will be a quarter of the way from the smallest number, so it gets called the lower quartile, and the “upper median” will be a quarter of the way from the largest number, so it gets called the upper quartile.

There are various ways of doing this. The right way is to take this set of numbers  as the bottom half of the data: 2 9 12 13 16 20 33 43 44 54 57 59. Note that it includes the median, 59. There are now 12 numbers. 12 is an even number. So we need to find the middle pair of  numbers; they are 20 and 33. Their average is 26.5. This is the lower quartile. The top half of the data also contains 12 numbers, because the median is used again: 59 64 67 72 86 93 132 258 430 484 1022 1322. The middle pair is 93 and 132, and their average is 112.5. This is the upper quartile.

The grid cell for Hopefield (3318AB) has only two records. Here is one of them. Black Stork Ciconia nigra submitted to BirdPix by Linda and Eddie du Plessis. This record is curated at


Our sample size of 23 is not exactly divisible by 4, so the rest of this paragraph is only approximately true. But as the sample size n gets bigger, it gets closer and closer to the truth.  A quarter of the sample lies between the upper quartile and the largest number, and a quarter lies between the lower quarter and the smallest number. So that means the remaining half of the sample lies between the lower quartile and the upper quartile. Half the sample is greater than the median and half the sample is less than the median. So these five numbers, smallest number, lower quartile, median, upper quartile and largest number provide a neatly interpretable summary of the numbers in our sample. We call them the five-number summary. However large n is, this strategy crunches the sample down to just five numbers. These provide real insight. But it is rare for them to be presented in a paper and actually called the five-number summary. Usually, they are plotted in a particular style, and that graphic is universally called the box-and-whisker plot. One of the next blogs in this series is  devoted to the box-and-whisker plot! And that blog will also reveal the person who invented this crazy name.

So for the small sample of n=23, a statistician would write the five number summary using this notation: (2, 26.5, 59, 112.5, 1322). Now that you are initiated into the secrets of unpacking and interpreting this, we know: (1) all the numbers lie between 2 and 1322, (2) half the numbers lie between 26.5 and 112.5, (3) half the numbers are smaller than 59 and half the numbers are greater than 59.

The grid cell for Hopefield (3318AB) has only two records. Here is second of the two. Steppe Buzzard Buteo buteo submitted to BirdPix by Mark Stanton. The record is curated at


Just for the record, for the Western Cape as a whole, confining ourselves to the 200 grid cells with at least one record, the five-number summary is (1, 7, 25, 58.5, 1322). Interpretation: (1) all the numbers lie between 1 and 1322, (2) half the numbers lie between 7 and 58.5, (3) half the numbers are smaller than 25 and half the numbers are greater than 25. (And there are 62 grid cells without data!) This summary becomes interesting when it is put alongside the summaries for the other nine provinces, and that is what the box-and-whisker plot, coming to you in a blog soon, will achieve. The simple recommendation out of this analysis is that vastly more data are needed before we can claim that BirdPix has a comprehensive dataset for the Western Cape.

The mean and the median seem to be very different animals. Try this exercise. The trimmed mean is calculated by finding the mean after the smallest and largest values in the sample (2 and 1322 in our small dataset) have been eliminated. The trimmed mean is 146.1. we can repeat the process. Chop off the two largest and two smallest  numbers, and find the mean of the remaining 19 numbers: the 2-trimmed mean is 107.2. Keep going. If the sample size is odd, you ultimately are left with a single number, which is the median. If the sample size is even, ultimately you reach a point were you need to take the average of two  numbers, which is also the median. So the mean and the median are the two ends of a spectrum. The trimmed mean is a real strategy for describing the “middle” of the data,  in situations where there are only occasional outliers.

When should you use the mean  and  when should you use the median? There are no strict  rules. The median is always good, and it has a simple interpretation. The mean works fine if the sample has no outliers. If the mean and the median are close together, then the mean will be fine.  In fact the mean is then preferred. This is because a vast amount of sophisticated statistical theory has been built up around the mean, and so it has become the dominant way to measure the “middle” of a sample. But,  beware, as in the example used here, the mean is often misleading.

The blog draws heavily on the textbook IntroSTAT. If you are impatient to move faster than these blogs on “data and statistics” do, then you can download the whole book. It is an amazingly small file (1.6MB).

Karis Daniel  produced the maps.

Exploring data: The difference between data and an anecdote


If you only have one data point, then you have an “anecdote”. Like the photo below:

The baboon was on the roof of a holiday house, snacking on the red lentils it had found inside (


The information we have is that of a single Chacma Baboon Papio ursinis on a date (30 December 2018) at a locality (Bettys Bay, Western Cape, South Africa), engaging in a nefarious activity (housebreaking and theft). This is an anecdote. From this single observation, we cannot draw the conclusion that lone baboons regularly raid holiday homes at Bettys Bay in summer. This is a sample of size one. To make decisions about how and when baboons in Bettys Bay need to be managed, a much larger sample of data is needed.

Sometimes a sample of  size one is massively important. It can alert us to a new and emerging issue. There is an awesome paper in Biodiversity Observations by citizen scientists John Fincham and Nollie Lambrechts. It is called “How many tortoises do a pair of Pied Crows Corvus albus need to kill to feed their chicks?” The abstract reads: “This paper presents proof of heavy predation on tortoises by a pair of Pied Crows at a single nest site in order to rear successive broods of chicks. ” The operative word is “single”. This is a sample of size one. From this single observation, it would be irresponsible to decide to cull Pied Crows to save the tortoise.

The question in the title of the paper in the ejournal Biodiversity Observations is: “How many tortoises do a pair of Pied Crows Corvus albus need to kill to feed their chicks?”. This is the photograph which contains the answer. There are 315 Angulate Tortoises Chersina angulata in it. The paper is at The record of the tortoises is curated in ReptileMAP at Photo: Nollie Lambrechts


The paper by John Fincham and Nollie Lambrechts comes to precisely the correct conclusion by saying: “A comprehensive survey to establish the extent to which this degree of damage is replicated needs to be undertaken urgently.” This is an important “biodiversity observation”. The authors might be onto a real conservation issue for tortoises. But they might equally well have discovered an unusual pair of Pied Crows! You cannot take management action on an anecdote, a sample  of  size one

If the sample size is two, then you really just have two anecdotes, two data points. You cannot draw conclusions from a sample of size two. How about three? How large a sample do you need to be able to draw reliable conclusions? How many data points do you need before you can decide whether an intervention is needed? A statistician would talk about “sample size” and denote this unknown number with the letter n.

There is (unfortunately) no straightforward answer to questions about sample size. There is no rule of thumb. Ultimately the answer lies in discovering how variable the thing you are trying to measure is.

If you are a budding astronomer, say in Ancient Egypt, and you wanted to  find out the number of days from one full  moon to the next. The answer is very dull: 29½ days. After you have got the same answer repeatedly, it is clear that you got it right first time. All you really needed was a sample of size one. If there is no variability, a sample of size one is adequate. But you cannot know that at first!


Nola Parsons measuring an oystercatcher egg at the Koeberg Nuclear Power Station. She is using “dial callipers”; the “ruler” shows that the egg is a little bit more than 60 mm, and the “dial” reads about 1 mm (but you need to look at it from directly above to get an accurate reading, to 0.1  mm). She used the size measurements of the egg and its mass to estimate how long the egg had been incubated for. You can read this in her PhD is entitled Quantifying abundance, breeding and behaviour of the African Black Oystercatcher. Here is a photo of one her study birds:


The eggs of the African Black Oystercatcher Haematopus moquini are variable in length, so you definitely need to measure more than one egg to get a good handle on average egg length. But this a not a particularly variable characteristic, so once you have measured a small  sample,  you have a pretty accurate estimate of egg length.

This herd of African Bush Elephants Loxodonta africana is curated at


In contrast to the lengths of oystercatcher eggs, the sizes of African Elephant herds are very variable. So to get a good estimate of “average”  herd size, a large sample size is essential.

In this blog, we  have learnt that a sample of size one can (usually) be dismissed  as an “anecdote”. We have learnt that, as the thing we want to measure gets more variable, we need larger and larger sample sizes to be able to draw conclusions from the data. Most of the time, large variability is a pain, requiring that we get a large samples to estimate the “average” of the thing we want to measure.

In future blogs, we will think about sensible ways to measure the average in a sample of data, and about how we measure variability.




Namaqua BioBash 2019: Citizen Science in the Karoo

Five days in the Northern Cape with a team of avid birders?! It sounds like a dream come true, and for me, that dream recently became a reality. Wednesday, June 12, a team of six citizen scientists (Jerome Ainsley, Chris Cheetham, Tino Herselman, Salome Willemse, Les Underhill, and myself) gathered in Handvol Gruis guesthouse near Calvinia in the Northern Cape for a 5-day BioBash. The previous BioBash here had been in midsummer. From Wednesday through Sunday, we aimed to explore as many pentads and grid cells in the surrounding area as deeply as we could. Though primarily searching for birds, we also took time to peek under rocks for scorpions and snap photos of passing springbok for the Virtual Museum. While, Jerome, Chris, Tino, and Salome focussed on atlasing for SABAP2, Les and I used the trip to collect VM data.

Springbok, curated at

Each day, we set out before sunrise to photograph birds. For me, this week was incredibly exciting; I encountered several “lifers”, birds to add to the list of species I have seen in my life. These included Karoo Eremomela, Karoo Korhaan, Fairy Flycatcher, Black Harrier, Black-Chested Snake Eagle, and Martial Eagle.

Karoo Korhaan, curated at

A personal highlight, though, was Layard’s Tit-babbler—not a particularly uncommon species, but notoriously difficult to see and photograph. With a bit of patience, this one eventually emerged from its bush and perched just long enough for a BirdPix photo.

Layard’s Tit-Babbler, curated at

Among a bevy of larks, chats, canaries, and other common karoo residents, we did have a few surprising finds as well. One of these was a population of Cape Sugarbirds spotted by Chris, quite far north of their expected occurrence. We also encountered a Goliath Heron far out of its typical range!

Familiar chat, curated at
Goliath Heron, curated at

We also recorded identifiable road kill found throughout the week. Though no longer living, photos of these animals still provide valuable information on their distributions (examples at

Though each day held its triumphs, these were often overshadowed in my mind by a sense of loss. On the drive up to Calvinia, we passed a dry riverbed labelled “Droge leegte”, “Dry Void”. Driving through the karoo or even looking at a satellite image of the region, those two words seem apt descriptors for much of the landscape. The karoo farmland has been devastated by drought and overgrazing in the past decades, and local farmers are struggling to raise crops and livestock. These time-lapse images from Google Earth Engine show the changes in the landscape between 1984 and 2018.

Time-lapse images of Calvinia and surrounding region. To view the years in between or play the full time-lapse, follow the link on the photo.

As we connected with landowners and farmers along our routes, conversations constantly revolved around rainfall. We spoke to the owner of a sheep farm near Nieuwoudtville, who shared that their land received only 4 millimetres of rain in 2018. The difficult reality is that in many ways, this ecosystem is dying. Overgrazing has left its nutrients depleted beyond hope of restoration, and the warming climate brings less and less vital rain each year. As a result, those who do live on the land are faced with the difficult choice of either struggling to continue farming, or leaving their homes, often of generations, to start over again elsewhere.

Sheep on a farm near Doringbos.

On the first day of data collection, we witnessed both of these outcomes. On a remote but active farm, Salome and I encountered sheep that were starving, unable to subsist on what little vegetation remained. We also passed homes that were abandoned; drapes still hanging in the windows, goldfish still swimming in the pond. Throughout the week, these eerily quiet scenes served as the sobering backdrop to our work, reminders of the severity of the drought and the effects of unsustainable land use.

Chris Cheetham searches for birds at an abandoned farmhouse near Loeriesfontein.

In spite of the hardship, I was struck by the resilience and kindness of the people we met. Residents readily connected to us through a mutual love of the land and the life it sustains, and were quick to offer information on the species they had encountered, the best locations to spot raptors, which roads to avoid, and more.

My takeaways from this (my first!) BioBash are twofold: I am deeply aware of the challenges facing this region of the country, but am also encouraged by the impact one small group can make in such a short period of time. Over the course of the week, the atlasing team (Salome, Chris, Jerome, and Tino) contributed 55 cards to SABAP2, and between Tino, Salome, Les, and myself, 840 records were added to BirdPix! The BirdPix additions are clearly visible in this map of the Namakwaland district municipality, which shows the number of species recorded in each quarter degree grid cell prior to and following the BioBash*. The diagram in the lower left corner shows the location of Calvinia (quarter degree grid cell 3119BD).

*The new map is not entirely up-to-date, as records are still being identified by the BirdPix team. It will be replaced with an updated map once all species ID’s have been confirmed.

To me, this success is just another beautiful reminder of conservation without borders: unique people united by a shared passion for the world around us. Even within our team, the diversity was incredible—we were a group comprised of statisticians, scientists, accountants, nature guides, craftsmen, and more! We need individuals from all backgrounds, disciplines, and cultures to generate strong and lasting conservation action, and projects like this are brilliant examples of what effective collaboration can achieve.

Colour ringing birds at Fynbos Estate

birdpix 81942

Bird ringing was conducted at Fynbos Estate from 17-21 June 2019, with two students from France, as part of their field project. During the ringing sessions Emilie  and Manon learned to identify and measure local bird species. Two species, Common Fiscals and Cape Robin-chats, were also colour ringed to enable observations on individually identifiable birds. During ringing sessions, territories of at least two pairs of Common Fiscals were noted. The Common Fiscals at Fynbos Estate often hunt from surprisingly low perches in the vineyards.

birdpix 81943
Cape Robin-chat


birdpix 82579
Common Fiscal


Table: Number of birds ringed and recaptured at Fynbos Estate, 17-21 June 2019

Species English Ringed Recaptured Total
316 Cape Turtle Dove 1 0 1
391 White-backed Mousebird 1 0 1
432 Acacia Pied Barbet 0 3 3
440 Greater Honeyguide 1 0 1
543 Cape Bulbul 5 0 5
581 Cape Robin-chat 6 1 7
621 Long-billed Crombec 1 0 1
622 Bar-throated Apalis 3 0 3
665 Fiscal Flycatcher 1 0 1
672 Cape Batis 2 0 2
678 Fairy Flycatcher 1 0 1
707 Common Fiscal 3 1 4
751 Malachite Sunbird 1 2 3
786 Cape Sparrow 1 0 1
799 Cape Weaver 76 11 87
803 Southern Masked Weaver 13 6 19
808 Southern Red Bishop 1 0 1
810 Yellow Bishop 6 0 6
1105 Olive Thrush 2 1 3
1172 Cape White-eye 32 6 38
4139 Karoo Prinia 3 2 5
4142 Southern Grey-headed Sparrow 2 0 2
TOTALS 162 33 195


Close to 200 birds were caught and processed. Recaptures included birds from the first ringing trip to Fynbos in May 2018. 22 species were handled, including three species not recorded here before: Greater Honeyguide (an immature male), Long-billed Crombec  and Fairy Flycatcher. The Greater Honeyguide is a brood parasite of hole nesting birds like woodpeckers and barbets and potential hosts at Fynbos are Cardinal Woodpecker and Acacia Pied Barbet.

birdpix 81942
Fairy Flycatcher


birdpix 81941
Greater Honeyguide


Would you like to colour-ring a bird? Book a trip with African Ringing Expeditions!

Longevities of Cossypha robin-chats

There are 14 species of Cossypha robin-chats in Africa. They are medium sized thrushes and sexes are alike. They have much red in the plumage and many species have a white eyebrow. The juveniles are spotted above and mottled below, but the tail pattern is the same as that of the adults. These birds have long tarsi and are usually found on or near the ground in forest or thickets, although some species occur in gardens that are near well wooded areas. They are most active at dawn and dusk, and are accomplished songsters and often mimic the calls of other species.

White-throated Robin-Chat

The Chorister Robin-Chat has the greatest longevity at 24 years, based on a recapture of a colour ringed bird (read story here). The Cape Robin-chat has been well ringed and has a high longevity of 17 years (see Table below). The other South African robin-chats have lower ringing rates, but still high longevities at 9 years or more. The other African robin-chats have very low ringing totals with SAFRING rings, but there are probably many more ringed for the Kenyan species.

Cape Robin-chat

The Birds of Africa handbook (Vol 4, 1992) provides two additional longevity records from various published studies. In Malawi the Olive-flanked Robin was recorded with a longevity of over 10 years. In Kenya a Blue-shouldered Robin-Chat had a longevity of 8 years (and there is a record of 24 years of a bird in captivity).

White-browed Robin-Chat

Many more longevity records are to be expected from this well loved group of birds, as more of them are ringed!

Red-capped Robin-Chat

Table. Number of robin-chats ringed and longevity records from SAFRING database, data extracted 7/06/2019

Species Latin Ringed / retraps / recovered Longevity Ringno
Olive-flanked Robin C. anomala 31 / 2 / 0 1y 0m 10d CV48526
Chorister Robin-Chat C. dichroa 1401 / 206 / 4 24y 295904
Red-capped Robin-Chat C. natalensis 3466 / 582 / 12 11y 11m 27d BD20060
White-browed Robin-Chat C. heuglini 1271 / 255 / 10 11y 9m 0d A109741
Cape Robin-chat C. caffra 19370 / 4538 / 197 17y 5m 12d F68667
White-throated Robin-Chat C. humeralis 1179 / 189 / 4 9y 8m 18d AA37869
Grey-winged Robin-Chat C. polioptera 40 / 3 / 0 3y 6m 15d FA12016
Mountain Robin-Chat C. isabellae 114 / 21 / 1 6y 4m 29d BH91815
White-crowned Robin-Chat C. albicapilla 6 / 1 / 0 0y 0m 8d 4A52476
Archer’s Robin-Chat C. archeri 3 / 0 / 0
Blue-shouldered Robin-Chat C. cyanocampter 3 / 0 / 0
Snowy-crowned Robin-Chat C. niveicapilla 116 / 28 / 0 7y 6m 12d 4A48948
Rueppell’s Robin-Chat C. semirufa 0 / 0 / 0
White-headed Robin-Chat C. heinrichi 0 / 0 / 0

Chorister Robin-Chat

Would you like to ring a robin-chat? Book a trip with African Ringing Expeditions!

Updates at Longevities

Virtual Museuming at Mwamba Conservation Centre, A Rocha Kenya

Mwamba is the Conservation Centre of A Rocha Kenya. It is on the coast, and offshore is the Watamu Marine National Park. Just inland is the Arabuko-Sokoke Forest, the largest remnant of a dry coastal forest which originally stretched from Somalia to Mozambique. Just to the south of Mwamba lies Mida Creek, which has mangrove forests, intertidal sandflats and lots of waterbirds, especially migrant waders. The Mwamba Conservation Centre is a superb resource for conservationists, researchers, students, community groups and even holiday makers.  Over the years, it has developed a smorgasbord of practical conservation programmes; there are initiatives in scientific research, environmental education and community conservation.

The Mwamba Conservation Centre is just inside the eastern boundary of quarter degree grid cell 0339BD, more or less under the U in the label for the Watamu Marine National Park. The boundaries of the Arabuko-Sokoke Forest are clearly shown in the Google image. Mida Creek lies between the forest and the sea. If you arrive by air, you fly into Malindi,  near the northeastern corner of the map. About 100 km to the south is Mombasa.

Dr Colin Jackson is the Scientific Director of A Rocha Kenya. I had supervised his PhD, which was entitled The moult and migration strategies of Lesser Sand Plover, Greater Sand Plover and Terek Sandpiper. Colin graduated in April 2018 and a substantial component of his data had been collected at Mida Creek. During a visit to Nairobi to participate in a Tropical Biology Association course, I took the opportunity to travel to the Kenyan coast and visit Colin “at home” at Mwamba, and also to make a belated pilgrimage to the study site at Mida Creek which we had talked about a lot.

In a place abounding with biodiversity, it was a great chance to make a contribution to the Virtual Museum. This blog is a small sample of the records.

This unassuming butterfly is the African Migrant Catopsilia florella. As its name implies this is a migrant butterfly, but (like the Brown-veined White Belenois aurota) the migrations are one-way! It occurs through most of Africa. There are currently 3,982 records of this species in LepiMAP,  the Atlas of African Lepidoptera. But we need a lot more records to be able to produce a respectable distribution map. This record is at

While I was at Mwamba, the commonest moth by far was Cyligramma latona. This moth is so well-known and widespread that it has an English common name: Cream-striped Owl. This large moth has a wingspan of about 80 mm. This “specimen” is curated in the Virtual Museum at It was the first record of the species in quarter degree grid cell 0339BD. Currently 71 species of Lepidoptera (moths and butterflies) have been LepiMAPped for this grid cell. The full list is here Colin Jackson says that 125 species of butterflies (let alone moths) have been recorded on the Mwamba property alone and that the Arabuko-Sokoke Forest is an even more magnificent destination for the lepidopterists.

On to reptiles! The Mwamba Conservation Centre has a designated “compost heap”.  I  am doubtful about the amount of compost produced … because it actually serves as the dining room table for …

… Water Monitors Varanus niloticus. This specimen is preserved in the ReptileMAP section of the Virtual Museum at There is an earlier record in exactly the same compost heap made in October 2012: It is fantastic to confirm the continued presence of Water Monitors here, and to be able to refresh that seven-year-old record. To see the full list of 12 reptiles (it might soon become longer as other records are submitted by visitors) go to

The grounds of the Mwamba Conservation Centre are similar to the others along this piece of coastline: about 4 ha in area, long, thin rectangles with about 100 m along the shore, and stretching back about 400 m to a main road parallel to the shore. Most of the area, across all the properties, is natural coastal vegetation, and it is used by birds as a coastal highway. Colin has set up a few mistnetting sites at right angles to the coast, and runs a regular bird-ringing programme that monitors coastal passage. One of these ringing sessions took place while I was there.

Bird ringing demonstrations have a long history of making people of all ages passionate about birds. Here, Jeff Davis, who teaches biology at the Rift Valley Academy, discusses the bird that he has just removed from the mistnet with the young people who were staying at the Conservation Centre at the time.

This bird left Colin amazed. It was the first Basra Reed Warbler Acrocephalus griseldis ever recorded at Mwamba. Its threat status is Endangered. This migrant has a large non-breeding range stretching from the Horn of Africa to Mozambique.  But it has a minute and shrinking breeding range in the Mesopotamian marshes of southern Iraq (amazingly, a few were found breeding in a reedbed in Israel in 2006).  This bird (but not this photograph) is curated in the BirdPix section of the Virtual Museum at

This is an (Asian) Lesser Cuckoo Cuculus poliocephalus. It has an unusual migratory pattern. This cuckoo breeds from Afghanistan, along the southern foothills of the Himalayas, in southern China and eastwards to Japan! They migrate southwestwards to India and East Africa, and are in Kenya from about November to April. This one, on 14 April 2019, would have been close to departure time. This photo, and two more, are curated at

Provided the winds and the clouds collaborate, flights in and out of Malindi provide a great view of Mida Creek. It is an inlet to the sea, and tidal, but there is no river running into it.

Mida Creek has an area of about 3,200 ha, a combination of tidal sandflats and mangrove forests.

Mida Creek is both a stopover site for migratory waders on passage northward (with an April peak) or southward (with a September-October peak) and a migratory destination in its own right (with birds staying from September to April). It is one of the most important sites for waders along the western rim of the Indian Ocean. To the north, the next really important stopover site is Barr al Hickman in Oman, about 4,000 km away.

Juma Badi stands between Colin and me. He is a Community Guide. While Colin was in Cape Town finishing his PhD, Juma diligently collected one of the best data sets of its type in Africa. Every five days, for two years, he counted the waders along a 1 km transect across Mida Creek. This provided fantastic information on the timing of arrival and departure of waders.

I saw my first Crab Plover at Mida Creek. Not just one, but hundreds. Mida Creek is a key non-breeding site. Curated at


Here is a Lesser Sand Plover. It has a leg-flag and is one of Colin’s study birds. It is curated at Colin’s PhD considered the migration and moult of three species along the Indian Ocean “rim”: Greater and Lesser Sandplovers and the Terek Sandpiper at localities in Kenya, India and Australia. In comparison with wader species which migrate along the coastlines of the Atlantic Ocean and Pacific Ocean, these species are poorly researched,  and Colin’s thesis helps fill an important gap in knowledge.

Colin also placed leg flags on Greater Sand Plovers. On 16 September 2013 he added a ring flag to a bird he had first ringed at Mida Creek nine years earlier. It was resighted on the shore in Kutch, India, on 10 April 2015. It  was resighted the following year on the same beach in India, on 29 March 2016. Great Sand Plovers breed in Kazakhstan and western Mongolia. The full story is here!

Colin was surprised to see a House Crow deep into Mida Creek, and quite far from  any human habitation. It was at the interface between the mangroves and the mudflats. This invasive alien started its African career at Mombasa. It is now so pervasive that there is no chance of eradicating it. This record is curated at

The photographic bird list for the grid cell is currently 31 species, so there is lots of opportunity to push it higher:

We will explore ways in which the Biodiversity and Development Institute can collaborate with A Rocha Kenya, and especially at the Mwamba Conservation Centre.

Filling in the Gaps: Citizen Science on the Road

An early start at 5:45 a.m. enabled us to reach Matjiesfontein railway station just after first light. We were traveling from Cape Town to Hanover in the Northern Cape, a journey which would typically require seven hours transit time. Instead, we used nearly twelve!

Though our final destination was Karoo Gariep Nature Reserve, the journey itself provided a chance to contribute to the Virtual Museum, and in particular, BirdPix.

Road trips are golden opportunities to fill “holes” in Virtual Museum datasets; many grid cells along the national roads have either very few records or none at all. Currently, BirdPix data are being used to generate species distribution maps which will be compared to SABAP2 maps. If BirdPix-generated maps are able to closely replicate the SABAP2 distributions, we may be able to apply the same methods to other VM datasets, modelling the distributions of butterflies, dragonflies, reptiles, and more!

With this in mind, we decided to use our trip to improve the existing coverage of the BirdPix dataset. By planning ahead, we identified a few sparsely populated cells to target along the journey, the first of which was Matjiesfontein.

Matjiesfontein falls in quarter degree grid cell 3320BA, which previously contained only two records: Speckled Pigeon and Fiscal Flycatcher, both reported by Zenobia van Dyk in 2018. Following a brief twenty-minute exploration of the area, however, we added 28 records to the list! These include species such as Blacksmith Lapwing, Cape Weaver, Common Starling, Laughing Dove, and this White-backed Mousebird, one of several encountered.

White-backed mousebird in Matjiesfontein, record curated at

Our next target was Leeu-Gamka, another small railway station approximately 350km northeast of Cape Town. This grid cell contained no records. We stopped near fields and buildings, searching for ibises among irrigation systems and doves along rooftops, and were well-rewarded for our efforts. In under an hour, Leeu-Gamka’s grid cell was brought from 0-18 species! One of my favourite finds was a Pririt Batis, hunting for insects just outside a supermarket.

Pririt batis, record curated at

In addition to their important place in data collection, long-distance road trips also allow for interesting observations on species ranges. For instance, in Matjiesfontein, we encountered Cape Bulbuls, but in Leeu-Gamka, just 150km northeast, were greeted by these African Red-eyed Bulbuls.

African red-eyed bulbul, record curated at

The distributions of both species are closely linked with the landscape: Cape Bulbul occurs in fynbos and coastal scrub vegetation, whilst the African Red-eyed Bulbul is commonly found in arid savanna and riparian bush.

We stopped at a few other locations along the road to Hanover, including an abandoned cluster of buildings near Nelspoort in the Nuweveld Mountain Range. The disused structures are now colonised by a variety of avian species, including Red-headed Finch, House Sparrow, Cape Sparrow, Southern Grey-Headed Sparrow, and Familiar Chat.

Familiar chat, record curated at

Red-headed finch, record curated at

The “prize” finding of the day, though, came much closer to our destination, and was easily my favourite: a pair of secretary birds just outside of Hanover.

Secretary bird, record curated at

These beautiful birds typically avoid busy areas, so spotting these two just off the national road was a surprising treat.

At the end of the day, 59 records were added across 5 grid cells! Contributions like these show the sizeable impact that a few extra stops can have in unpopulated or remote regions. Paired with some cautious driving, roadsides become fantastic spots for birding and generating citizen science data.

The scope is by no means limited to birds–you can contribute on your next journey via any of the Virtual Museum’s collections, and share the species you encounter along the way!

Keep exploring!

… participating in a Tropical Biology Association course in Nairobi

It was a great privilege to be part of the team teaching at a course for the Tropical Biology Association in Nairobi during April. The course was called “Analysing and interpreting citizen science data.”The Tropical Biology Association (TBA) is making a fantastic contribution to capacity building for biodiversity conservation in Africa and southeastern Asia. The impact of the TBA is awesome, and browsing its website is an inspiration. It punches way above its size. The TBA has helped launch the careers of more than 2,000 conservation champions. My Nigerian PhD students have talked to me enthusiastically about the TBA courses they participated in.  For example, Dr Zingfa Wala, who did a four-week TBA course in Madagascar, says: “The TBA course is the one that had the most impact in my career. It exposed me to practical and hands-on experimental design.  It was my first experience working with a team and it opened up great networking opportunities.”

But back to Nairobi, April 2019. The course I participated  in was advertised as Citizen Science in Africa – making it work for conservation. The TBA motivated this course on their website: “Across Africa, citizen science is producing a growing body of data and inspiring significant numbers of people to connect with local and national conservation issues. This exciting development has led to an urgent need for training to enable African scientists to analyse and interpret these data in order to generate useful outputs to inform practice and policy.” The course was a partnership between the Tropical Biology Association, the National Museums of Kenya, the British Trust for Ornithology and the Biodiversity and Development Institute.  It was funded by Cambridge Conservation Initiative.

An indication that the TBA had identified a genuine need is the fact that more than 200 people applied, 10 times more than could be accommodated. The 22 participants who were selected came from 10 African countries: Liberia, Benin, Nigeria, Cameroon, Congo, Uganda, Kenya, Rwanda, Tanzania and Botswana. We gathered at the National Museums of Kenya on Monday 8 April for a week-long course that started at 8 am each morning and finished at 5 pm.

Two sets of numbers from 1 to 11 were placed on the tables, and each participant had to find the other person with the same number, and introduce that person. Clara Cassell from Liberia got paired with Judith Mirembe from Uganda.  Judith told us that Clara does yoga, and Clara told us that Judith plays hockey. Judith is also leader of Uganda Woman Birders and was recently featured by the Audubon Society:

The main thrust of the course was the analysis of the data collected through citizen science projects. Dr  Rosie Trevelyan, who heads up the TBA, facilitated many discussions. This is the output from a session in which we discussed the challenges of  doing citizen science in Africa.

The participants were taken step-by-step through the basics of R, developing the skills needed to produce maps of their own countries in R. This part of the course was led by Dr Simon Gillings, of the British Trust for Ornithology. Here he is describing some of the outputs from the citizen science projects he is responsible for, dealing with birds in the UK.

My own contribution to the week consisted of six presentations. This slide illustrates how ideas about the distribution of the Amethyst Sunbird chopped and changed. The truth emerged in 1997 as a result of the distribution map generated by citizen scientists in the first bird atlas. This is the map in the bottom right hand corner of the slide.

There were visiting lecturers. One was David Clarence, who is professionally an econometrician, and an enthusiastic contributor to the bird atlas project in Kenya. He has applied his statistical skills to the Kenyan bird atlas data, and he described to us the huge strides he has made in the analysis.

During the course, there was lots of time allocated to working on the datasets the participants had  brought with them. This enabled them to get help when stuck.

We spent one morning in the Nairobi City Park. In part it was a great excuse to get outside and in the sun, but more importantly it was an opportunity to test protocols for collecting data. City Park is an excellent resource for this kind of activity. This Blotched Leopard Lachnoptera ayresii was the first record for the species in LepiMAP, the Atlas of African Lepidoptera, for the grid cell (0136BD) into which Nairobi falls (see Here is the map of the Nairobi grid cell (City Park is in the north-centre of the grid cell) and list of species for the Nairobi grid cell Currently that list has 68 species.

Quite a large part of the museum precinct is natural vegetation. After the classes one day a group of participants found this horned chameleon. It got submitted to ReptileMAP ( and was identified as Jackson’s Chameleon Trioceros jacksoni. One of the participants on the course was Laban Njoroge, who is the entomologist at the museum. The museum’s collection of insects was on the same floor as our course venue, so Laban had the shortest distance between his office and the course! In the course WhatsApp group, Laban told us about the Jackson whose name is attached to this chameleon: “This man Jackson was a seasoned and passionate naturalist. In one of the AGMs of the then East African Natural History Society (now Nature Kenya, Nature Uganda and Nature Tanzania) he came up with an idea of making natural history collections. To kick start his idea he donated a drawer of butterflies. This is how the huge insect collection at National Museums of Kenya was started. This drawer still remains in our collection to date.”

Laban took me on a tour of the insect collection he curates. Since Jackson’s time, the National Museums of Kenya’s collection has grown to more than three million specimens. Here is Laban showing me the Black Emperors Anax tristis in the collection. This is Africa’s largest dragonfly, with a wingspan of about 13 cm. This is also one of Africa’s largest and most valuable insect collections.

Back to Jackson’s Chameleon! It is astonishing that an animal like a chameleon can become a problematic invasive alien when introduced elsewhere. Chameleons form part of the pet trade, and some Jackson’s Chameleons were released in Hawaii.  Here is part of the abstract of the paper in the journal Biodiversity and Conservation featured above: “Native Hawaiian invertebrates, including four individuals of an endangered species, were discovered in the dissected stomachs of wild caught Jackson’s Chameleons collected from June to November 2009 on the island of Oahu, Hawaii. Jackson’s Chameleons were introduced to the Hawaiian Islands in the early 1970s. Of particular concern is the fact that introduced chameleons have previously only rarely been found in native Hawaiian habitat. One concern is that Jacksons’s Chameleons may be undergoing a range expansion into upper elevation pristine forests. If they reach and establish populations in these areas, devastating impacts to the native ecosystem are possible.”


At the end of the course, each participant received a certificate. Here, Sidney Shema, who leads the bird atlas project in Kenya, is presented with his certificate by Professor Mary Gikungu, who is the Director of National Repositories and Research at our host institution, the National Museums of Kenya. Professor Gikungu is a entomologist specialising in bees.

Read the TBA’s own report on the course. It is entitled “Launching a new wave of citizen scientists.”

Thank you to Rosie Trevelyan  and Anthony Kuria (TBA) for the invitation to participate. It was a rich and rewarding experience. We hope that the BDI and the TBA can find ways to collaborate in the future.