2 W. Szafer Institute of Botany, Polish Academy of Sciences, Lubicz 46, 31-512 Kraków, Poland; ibnalepka
ib-pan.krakow.plAdam Walanus1, Dorota Nalepka2;
ib-pan.krakow.pl
Pollen data are stored in tables, with taxa in columns and samples (spectra) in rows. Such tables, typically are stored in spreadsheets (MS Excel), however it may have simple form of text, with values separated by tabulator (txt files) or by coma or semicolon (, ; csv files). For storing and printing of smaller tables (up to 30-50 taxa) also text editors (like MS Word) can be used.
In case of any of mentioned forms, and probably of some other, it is enough to copy the table to the Clipboard (Ctrl-C) and then, to click the menu option "Load table from Clipboard" in the described here application, to have data ready for diagram plotting. However, in POLPAL system, simple database may be constructed, with one, consistent list of taxa for all pollen tables. Advanced results can be obtained using such database.
Of course, not only pollen and spores data may be handled in POLPAL, also other remains counted in reasonable amount, like Cladocera, diatoms, molluscs, charcoal particles, fruits, seeds and other micro and macro objects.
Shown below are pollen tables in Notepad, Excel, and Word.
Application Diagram plots pollen diagrams, in many different forms. Also numerical (statistical) analysis of data (ConSlink, PCA, Rarefaction) may be added to the diagram, by two mouse clicks.
The final result, the diagram, is in the standard MS Windows format of bitmap (*.bmp file). It is very easy to add to the bitmap some additional descriptions, lines etc. Printing bitmap is easy as well, simply because it is standard format. Even the diagram of 300 taxa and 300 samples may be printed with perfect quality (see for example Ralska-Jasiewiczowa et al. 1998).
Bitmaps, however, are large files, even when saved as monochrome. It can be recommended to save bitmaps in GIF format, which is compressed one, with no lost of information. The JPG is also possible, however, not recommended since the compression rate is correlated with the lost of picture quality.
A few examples of diagrams are shown below (some figures are sized down).
Diagram in color:
Monochrome, large diagram (compressed):
Cyclogram (pie), see Layout options
Diagrams AP/NAP and special form of presentation of Pinus and Betula (see Miscellaneous information):
Different forms of pollen "curve" (the last one is obtained with help of MS Paint):
There is additional application PPTable that may be used instead of Excel. Counting pollen grains is supported in that application, with use of special coding of taxa.
| 1 | t | ab | Abies alba |
| 2 | t | ac | Acer |
| 44 | h | acd | Aconitum |
In order to keep many pollen tables in consistence, the list of taxa may optionally be used in POLPAL. Such a list consists of taxa names, taxa codes, taxa numbers, and life form code, which help in diagram construction.
There is no installation of Diagram application, simply copy the file Diagram.exe into some folder (for example, where you already keep your pollen data). Then run the program. (The shortening to the application may be obtained by dragging it with the right mouse key).
Below is fragment of Windows screen with useful programs, and folder with pollen data (xls, csv, txt) and results (bmp).
It is assumed that pollen table is in MS Excel, or in other spreadsheet, or in text editor. Below is pollen table selected in MS Excel.
Taxa names are in the first row. Depths are in the first column. Other cells contain counts (empty cell = 0). Select table, copy it to the Clipboard (Ctrl-C), and then click the menu option Load table from Clipboard in Diagram.
How to select large table:
It operates in text editors also.
In case of difficulties in using Load table from Clipboard option, paste the content of clipboard into the Notepad (simply start Notepad from Windows/Programs/Accessories and press Ctrl-V). Because of long taxa names that form of table is not very convenient, anyway Excel may read it back. If the table is in Notepad; it is worth to save it, since ...
Pollen table may also be loaded to Diagram from the file. Click in the menu Load text table. In the Open dialog two types of file may be chosen: *.CSV or *.TXT. The first is produced by Excel ("save as csv"), the second may be obtained in Excel ("save as text") as well, and in Write, Word, Notepad and many other text editors. In case of CSV, data (numbers) are separated by coma (,) or semicolon (;), in case of TXT, the separator is TAB (tabulator).
Pollen table below may be copied (select it, and Ctrl-C) from here, and used by Diagram.
| depth;Abies alba;Acer;Alnus undiff.;Betula undiff. |
| 108;1;2;15;32 |
| 112;4;3;76;100 |
| 121;4;0;79;125 |
| 131;4;1;61;66 |
With taxon name additional information can be transferred to the program. The symbol '#' is used here. For example: Alnus#tg3 means, that taxon Alnus is a tree, the curve is to be of color "g" (defined in the program), and that taxon is to be the third in the diagram. Those three parameters are optional (if there is one only letter, it is treated as life form - used in pollen sum definition). However, if you have more than one table, it is recommended to use list of taxa.
Taxon name length is limited to 255 letters (Myriophyllum spicatum/verticillatum is much below that limit); taxa names can be edited in the program (as well as taxa color and position), during process of diagram adjusting.
Maximal number of taxa =350, the same is the maximal number of samples. Since MS Excel has only up to 256 columns, text file (txt or csv) is to be used in case of more taxa (see also application PPTable).
Below is the result of loading to the application Diagram pollen table given on figure above, and clicking the button Diagram.
Button Taxa (see above) recalls the taxa of the table, with some parameters.
Five (editable) columns contain:
The second column i.e. sequential number is of high importance. There are many mechanisms for manipulating them from the menu (Taxa arrangement). At start, numbers that correspond to the position of taxon in the table are used (if numbers are not given with taxa names - Alnus#50). However, 10, 20, ... not 1, 2, ... are used to make it possible to insert some taxa. One general rule is applied here: taxa with the same number will be summed up, and plotted under the name of the first taxon to be summed up. For example: 10 Abies, 10 Alnus, 10 Betula,....10 Ulmus will produce curve of trees (change Abies to trees). Taxa with no number (empty cell, or letter in the cell) will not be plotted.
Some buttons are visible only if available. For example Taxa is invisible when spreadsheet with taxa names is actually displayed.
At the left part of the Text tab of Parameters dialog, there are options for placing text instead of sample depths.
At the right part of the Text tab of Parameters dialog (see figure above), there are options for adding zones description at the end of the diagram. The text of zones definition has the following structure. In each row there is: depth - tabulator - letter s or d or nothing - tabulator - zone name. It is impossible to type tabulator into the proper field in the dialog, so the text is to be prepared in Excel, Word, or Notepad.
There are many parameters used in plotting diagram. They are stored in the text field. Most of them are accessible from menu; however, some more advanced parameters are to be customized in the editable taxt field. Click menu Parameters to get it, or press F10. The values (numbers or texts) after the "=" mark can be modified. The text field of All parameters can be saved and loaded back. File Param.txt, if exists, is loaded automatically. Note that the parameters may be rearranged. User can move the frequently used parameters to the front.
Click on the diagram with the right mouse key to show popup menu. There are four options: Start here - X, Start here - Y, Start here - X and Y, and Up to the program. It may be decided to start plotting of the next curve (in Multi diagram mode) at the horizontal (X) or vertical (Y) position defined by the right mouse click. The last option make free the decision about the Y position (see y0 external=n).
The final result, the diagram in form of bitmap can be enhanced, for example using the simplest graphical program - MS Paint, part of Windows Accessories.
For many reasons it is good to have one list of taxa names. Such a list may be easily maintained, errors in names has to be corrected once only, not in each table. In the POLPAL list of taxa, in addition to taxa names the life form is added (one letter code: t-tree, s-shrub, d-dwarf shrub, h-herb, n-non-pollen, l-local, p-spore, i-indicator, x-other, ...).
As identifier of taxon, its number or its code may be used. Code is composed of two, three or four letters. The code, as well as number is to be unique in the list (application PPTable checks it). Codes are used mainly for counting pollen.
Below is Notepad with list of taxa. Tabulator separates columns. It is possible to create list of taxa from scratch in Notepad, however better is Excel. From Excel the list of taxa is to be exported as text (txt) file. List of taxa saved in file named taxa.txt, in the same folder where is application Diagram.exe, will be loaded automatically by Diagram.exe (as well as by PPTable) after running it.
Other form of list of taxa:
Below is the example of pollen table when list of taxa is in use.
Used are numbers of taxa (4, 24, 370, 29), codes (bn, abi), and simply names (Boraginaceae, Artemisia, Grass). In any of that 3 cases there are two possibilities: (1) program will find the number, the code or name in the list of taxa or (2) will not. In the first case, the name of taxon from the list of taxa will be taken to diagram, in the second, the "name" from the table will by taken (probably "Grass" does not exists in the list of taxa). In the first case, also the life form of the recognized taxon will be taken to diagram from the list of taxa. Note, that while using full taxon name in the table, it must be precisely the same taxt as in the list of taxa. For example "Allium vineale t.", "Allium vineale t", and "Allium vineale" are three different names.
Taxon name length is limited to 255 letters (Myriophyllum spicatum/verticillatum is much below that limit).
In the list of taxa, please start taxa names with capital letters (Betula not betula).
Set of taxa is a structure as simple as useful. It is a set of taxa names (or taxa codes or numbers - if list of taxa exists), in separate rows. Below are three sets of taxa in Excel and one in Notepad. The order of taxa in set of taxa will be used in plotting diagram.
Taxa can be summed up. Character "&" in front of taxon name or code will result in summing the taxon up to the previous one. Here, in the first set Tilia platyphyllos will be added to Tilia cordata-type. (In application Diagram taxa will get numbers 10 Pinus, 20 Betula, 30 Abies alba, 40 Quercus, 50 Artemisia, 60 Tilia cordata-type, 60Tilia platyphyllos.) In the second set Betula nana-type cf. will be added to &Betula nana-type; Tilia platyphyllos and Tilia cordata-type will be added to Tilia. In the third set taxon number 478 will be added to taxon bb.
Sets of taxa from Excel can be sent to application Diagram using Clipboard (Crtl-C in Excel, click menu Taxa arrangement/Set of taxa (paste from Clipboard) in Diagram). Sets of taxa stored in separate text files (for example: Set of taxa Trees.txt) can be leaded to Diagram by menu option Set of taxa (load from file). Having adjusted sequence of taxa in the program Diagram, the sequence can be saved as set of taxa (in text file) - see menu option Taxa arrangement - Save chosen taxa as Set.
Using Excel: (1) select row with taxa, (2) copy it (Ctrl-C) to the Clipboard, (3) go to some free spreadsheet, (4) select cell D1 (5) use menu option Edition/Paste special, and check option Transpose, then (5) click OK. (6) Fill columns A, B, and C:
(7) Save spreadsheet as text file taxa.txt.
Using Notepad or Word: (1) remove all data but taxa names, (2) in Notepad manually, or in Word automatically, change tabulators (or ;) to new lines, (3) add in front of each taxon its number, life form and code (separated by tab or ;), and (4) save result as text file taxa.txt.
Application PPTable is a spreadsheet for manipulating pollen table. It may be used instead of Excel. Some operations, typical for palynology is simpler here. However, the option of counting pollen grains may be of special interest.
PPTable is able only to read pollen table stored as text files (txt or csv). Anyway, using PPTable one need no to be familiar with text files. It is enough to write numbers to the spreadsheet or use Counting option, to fill the table, and then to Save it.
PPTable uses list of taxa and checkes any loaded table if it is compatible with the list of taxa. Any unrecognized taxon is to be identified with some from the list of taxa (see below). Note that the list of taxa contained in file taxa.txt, in the same folder where is the program is loaded automatically.
Both mentioned features of PPTable enforce mutual consistence of pollen tables, what is necessary for performing comparisons of profiles, and integration of data.
Note, that columns can be dragged form one position to the other. To do that press left mouse key on the header of a column and move cursor left or right.
Note that; if the amount of indicator added differs from sample to sample, the indicator counted is to be recalculated. Choose an amount of indicator added as standard (for example 13500) and use it for that window. However, indicator counted must be recalculated as follow: icrecalculated=icoriginal*13500/ia, where ia is for indicator added (to given sample), and instead of exemplar 13500 some other value may be used. Round the obtained value off to the integer.
In the upper part of the dialog for counting (see below) there is green text field. Here letters and other keys should be typed in. If program does recognize a code, the taxon name will be displayed (program also may tell you the taxon name, using wav files). One count will be added to the taxon in the sample selected in the main dialog. If taxon is absent in the table, it will be added at the end of the table. During counting, two sums are displayed. To define the sums (also its names) use proper lines in Parameters field. Sums are defined by life form.
The codes dictionary is obtained by pressing capital letter, and, if necessary, next letters of the taxon name (in lower case). After making use of dictionary (by reading and remembering taxon code), press Esc. Double click on taxon in the dictionary field adds one count. Single click only selects taxon, then one of the function keys (F1, F2, ..., F9, F11, F12) may be assigned to the taxon by keystroke.
Menu options shows: field of some parameters, record of the process of counting, and picture with simple help. Clicking menu option switches these items on and off.
Parameters may be saved and loaded. The blue field of Parameters may be edited as text file, however, only the text on the right side of the "=" mark may be changed. Most of lines are of obvious meaning, other are commented below.
Confirmation of counts by voice
To use sound, prepare wave files: V_.wav V_bb.wav V_abc.wav ... V_error.wav, and store it in the same folder where program is. In file names taxa codes are used. File V_.wav is used for all taxa which has no its own wav file (for example "something else"). File V_error.wav is used if code entered during counting remained unrecognized.
Simple help summarizes all options:
Coding taxa: ab ac al bb cr ff fx pc pp po qq tx uu tt my lx cohx fr hi jj rha ss bn cv em ach ca cal cea ch cra cya faa .........
INQUA Working Group on Data-Handling Methods Newsletter 11: January 1994 OPTIMIZING TAXON CODES IN POLLEN COUNTING, Adam Walanus
There seems to be a rule that whenever a computer keyboard gets too close to the microscope a new program for counting pollen arises. However, many programs "emulate" mechanical counters only, whereas even the oldest XT can do much more (Davis, 1993). In my POLPAL (Walanus, 1991) a kind of optimization of the taxa codes is applied. In the process of counting, two things should be minimized: the number of keystrokes and the amount of human memory occupied by the codes. That is possible under the obvious assumption that some taxa are frequent, some are rare and the rest are very rare. Up to the ten most frequent taxa are coded by the function keys F1..F10. The next group (in principle up to 26 x 26 = 676 taxa) is coded by the sequence of two letter keys. The remaining "very rare" taxa are coded by the three-letter codes, which are four keystrokes in fact, because the space bar must be pressed before the letters. How many taxa should be coded with one-, two- or three-keys codes is a matter of which kind of effort is preferred: to search for the key on the keyboard or to remember codes. It must be mentioned that a codes dictionary is available in POLPAL any time by pressing Enter and the first letter of the taxon's name. However, searching for the code on the screen is the most time-consuming task during counting. It is really a personal decision whether it is worth to remember: F5 = Alnus, and use one keystroke per Alnus grain, or to use: al = Alnus, easy to remember but with the 100% additional wear to the finger. Only the most frequent taxa should be coded by Fn keys. The less frequent taxa are coded by the two-letter codes. However, like the Fn's, not all possible pairs of letters must be used. There is no reason to code Corylus avellana by, say, iy because co, ca and comparable options are already occupied. Certainly, the three-letter cor is a pretty combination for remembrance. The rareness of a taxon can be estimated automatically. Having to use the dictionary frequently to look up a taxon's code will soon fix the combination in your long-term memory!
There are two small, additional applications used in manipulating large bitmaps produced by Diagram:
Program Sum_d.exe makes it easy to combine two diagrams. By mouse click you mark the point on the "1" and on the "2" diagram. Pressing Merge button adds diagrams in such a way that marked points coincide. New diagram is as large as necessary to comprise both summed up.
Program Cut_d.exe makes it easy to combine some curves from diagram, which is already in the form of bitmap. By two mouse clicks you mark the vertical fragments of "source" diagram. Pressing Extract and add button adds the fragment to the new "resulting" diagram. Note, that in order to merge curves from different diagrams, both diagrams have to have the same vertical position of the first sample; use option y0 external with some numerical parameter (for example y0 external=500) to ensure that.
The next step in pollen analysis may be numerical analysis of pollen table. Below three types of available numerical analysis are added to piece of pollen diagram.
Click on the menu option Numerical analysis shows dialog:
The color of curves resulted from rarefaction, and PCA analysis is defined by the parameter Color PCA & Raref (see All paramaters)
The next stepe in Numerical data analysis may by attempt to compare two pollen tables. Eeach sample from one table is to be compared with each sample from the second.
Application PP_SSM.exe has six tabs: List of Taxa, Table 1, Table 2, Set of Taxa, SMM, and Options. The sequence of tabs is equivalent to the typical sequence of steps in producing Similarity Matrix.
The List of taxa is visible at the first tab. List of taxa is necessary here, since two pollen tables are to be compared. Taxa from one table are to be identified with the proper taxa from the second table. If some list of taxa is in the file Taxa.txt (in the same folder where is the program), that list is loaded automatically at the start of the program. However, any other list of taxa can be loaded.
Two tables are to be loaded (using tabs Table 1, Table 2), to be compared. Taxa from tables, which program has been unable to identified with any taxon from the list of taxa, are listed in special field. If there are taxa not to be used in the analysis, there is no problem.
For numerical analysis, as a rule, only taxa more frequent than one grain in one sample, are used. The Set of taxa can be loaded to define taxa to be used in SSM analysis.
List of taxa may be pasted from Excel (Ctrl-V) to the upper white field. List of taxa can be edited in that field. It is enough to change one letter to remove taxon. List of recognized taxa is given below the editable field. Taxa form the first or the second table can be used as set of taxa (use buttons Table 1, Table 2).
Tab SSM is very simple. Click button Draw to obtain the resulting figure. Deselect Draft option to produce final, printing quality plot. The button Save shows typical dialog; use it to save bitmap obtained.
Options tab.
Default options ore OK for the most of typical cases. However, depth scale of final plot sometimes should be adjusted. Especially, in order to combine typical pollen diagrams with SSM matrix (see Sum_d.exe) it is necessary to use the same scale. The Vertical scale from Diagram is to be used here. It is possible, of course, to calculate scale for already obtained diagram:
In MS Paint read (in pixels) the vertical position of two samples. Here sample 210 is at 35 pixels, sample 250 is at 316 pixels, so vertical scale=(315-35)/(250-210)=7 [pixels per depth unit].
Let us denote by p1,i,k and p2,j,k the transformed (or not) values for k-th taxon from the i-th sample from the first table, and from the j-th sample from the second table. Attenuated Euclidian difference of these numbers is calculated:
di,j,k=abs(p1,i,k-p2,j,k)/sqrt(p1,i,k+p2,j,k)
where abs is a function abslolute value, and sqrt is a square root. Attenuation (achieved by division by sqrt of sum of %) is for diminishing influence of the most abundant taxa. It has little or no influence for present/absent data, as well as for data transformed to Equal Taxa Weight.
The values di,j,k are summed over all taxa used for analysis, to give the final value di,j of dissimilarity between sample i and j. Since it is more interesting to see which samples are similar, rather then dissimilar, squares plotted on the matrix are rare while di,j,k is small. The value of di,j,k is normalize so, that the largest di,j,k is equal to 1, and the value of 1-di,j,k is used in plotting black squares. However, the additional transformation may be used for 1-di,j,k (see Options tab).
A number of pollen tables with taxa names (or codes or numbers) consistent with one list of taxa make a database. Data acquired from such a database can be used, for example, for creation of iso-pollen maps. Below example is presented, obtained using data base of about 200 pollen tables (sites). Hundreds of such maps are created automatically, for different time horizons and different taxa (or groups of taxa).
Nalepka D. & Walanus A. 2003. Data processing in pollen analysis. Acta Palaeobotanica 43(1): 125-134.
Ralska-Jasiewiczowa M., Goslar T., Madeyska T. & Starkel L. 1998. Lake Gościąż, Central Poland. A monographic study. Part 1. W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków.
Ralska-Jasiewiczowa M., Nalepka D., Goslar T. 2003. Some problems connected with the vegetation development during the oligocratic/Homo sapiens phase of Holocene interglacial in central Europe. Vegetation History and Archaeobotany 12(4).
Walanus A. & Nalepka D. 1997. Palynological diagram drawing in Polish POLPAL for Windows. INQUA Working Group on Data-Handling Methods. Newsletter 16: 8-10.
Walanus A. & Nalepka D. 1999. POLPAL. Program for counting pollen grains, diagrams plotting and numerical analysis. Acta Palaeobotanica Suppl. 2: 659-661.
Walanus A. 1989. Saving computer memory in storing tables of pollen count. Pollen et Spores 31 (1-2): 161-164.
Walanus A. 1994. Optimizing taxon codes in pollen counting INQUA Working Group on Data-Handling Methods. Newsletter 11: 6.
Walanus A. 1995a. Pollen Data in Space and Time - Local Approach. INQUA Working Group on Data-Handling Methods. Newsletter 13: 13-14.
Walanus A. 1995b. Pollen data in space and time - local approach, INQUA-Comission for the Study of the Holocene. Working Group on Data-Handling Methods. Newsletter 13.
