Biological Diatom Index intercalibration : IBD NF T 90-354

on the river Loup (Alpes Maritimes, France)

First results on 04/04/2000

On 31st March 2000, the results of following participants (alphabetical order) were arrived at the Agence de l'Eau Artois-Picardie :

Almeida Salomé, University of Aveiro, Portugal
Druart Jean-Claude, INRA Thonon les Bains, France
Honoré Marie-Ange, Institut Pasteur de Lille, France
Iserentant Robert, Université de Louvain, Belgique
Lalanne-Cassou Christian, DIREN Ile de France, France
Lesniak Christophe, Agence de l'Eau Artois-Picardie, France
Nazart Maxence et Guillard Didier, DIREN Pays de Loire, France
Nouchet Nathalie, Bi Eau Cie, France
Peeters Valérie, DIREN Bourgogne, France
Peres-Weerts Florence, FPW Cie, France
Prygiel Jean, Agence de l'Eau Artois-Picardie, France
Rumeau Alain, Conseil Supérieur de la Pêche, France
Sabater Sergi, Université de Barcelone, Espagne
Straub François, Lycée Blaise Cendrars, La Chaux de Fonds, Suisse
Tudesque Loïc, Centre de Recherche Public Gabriel Lippmann, Luxembourg
Vidal Henri, Conseil Général des Alpes Maritimes, France
Vizinet Jessica, Aquascop Cie, France.

Two exploitation types are foreseen. A first one concerns the index values and their significance. The other concerns essentially the systematic.

 

Table 1 : IBD intercalibration results (September 1999) for 13 operators

 

P= sampling ; L= slide ; I= inventory ; Witness (slide) : slide realized from a unique sampling and transmitted to each operator

 

A

 

B

 

C

 

D

 

E

 

F

 

G

 

H

 

I

 

J

 

M

 

N

 

O

 

Witness

 

19,1

 

19

 

19,2

 

18,6

 

19,5

 

16

 

16,8

 

16,4

 

20

 

17,2

 

17

 

19,3

 

19

 

P1L1I1

 

20

 

18,9

 

19,3

 

18,7

 

19,9

 

16,2

 

15,9

 

17,7

 

20

 

16,7

 

16

 

19,5

 

19,3

 

P1L1I2

 

20

 

19,2

 

19,3

 

19

 

19,1

 

16,3

 

16,1

 

17,4

 

20

 

16,8

 

15,6

 

19,2

 

19,2

 

P1L1I3

 

20

 

18,9

 

19,4

 

19,4

 

19,6

 

16,2

 

16,1

 

17,3

 

20

 

17,1

 

15,7

 

19,2

 

18,6

 

P1L2I1

 

20

 

18,9

 

19,1

 

19,6

 

19,1

 

16,3

 

15,3

 

17,2

 

20

 

17,2

 

16,3

 

19,1

 

19,2

 

P1L2I2

 

20

 

17,2

 

19,5

 

20

 

19,5

 

15,9

 

15,3

 

17,1

 

20

 

17,2

 

16,4

 

19,5

 

19,5

 

P1L2I3

 

20

 

17,8

 

19,4

 

19,8

 

19,3

 

16,1

 

15,8

 

17,3

 

20

 

17,1

 

16,2

 

19,3

 

19,4

 

P1L3I1

 

20

 

18,1

 

19,4

 

20

 

19,3

 

15,9

 

15,6

 

17,2

 

20

 

17,2

 

15,6

 

19,3

 

19,3

 

P1L3I2

 

20

 

17,9

 

19,6

 

20

 

19,5

 

16,1

 

15,4

 

16,9

 

20

 

17,1

 

15,6

 

18,8

 

19,2

 

P1L3I3

 

20

 

17,6

 

19,1

 

20

 

19,3

 

16,1

 

15,3

 

17

 

20

 

17,1

 

16

 

19,6

 

19,1

 

P2L1I1

 

20

 

17,6

 

19,3

 

18,5

 

20

 

16,4

 

15,8

 

17,3

 

20

 

16,9

 

16,8

 

18,7

 

19,5

 

P2L1I2

 

20

 

16,8

 

19,3

 

18,7

 

20

 

15,8

 

15,4

 

17,2

 

19,9

 

16,9

 

16,8

 

18,7

 

19,3

 

P2L1I3

 

20

 

17,2

 

19,4

 

18,3

 

20

 

15,6

 

16,1

 

17,1

 

20

 

17,1

 

16,9

 

18,8

 

19

 

P2L2I1

 

20

 

17,4

 

19,6

 

19,1

 

20

 

16

 

15,6

 

17

 

19,8

 

17

 

16,6

 

18,8

 

19,2

 

P2L2I2

 

20

 

16,8

 

20

 

19

 

20

 

16,2

 

15,7

 

17,2

 

20

 

16,9

 

16,5

 

18,8

 

19,1

 

P2L2I3

 

20

 

17,4

 

19,8

 

18,7

 

20

 

16,3

 

15,5

 

17

 

20

 

16,8

 

16,6

 

18,7

 

19,1

 

P2L3I1

 

20

 

17,5

 

19,7

 

18,9

 

20

 

15,9

 

15,4

 

17

 

20

 

17,1

 

16,3

 

18,6

 

19,3

 

P2L3I2

 

20

 

17,4

 

20

 

18,6

 

20

 

15,9

 

15,3

 

17,3

 

19,9

 

17,2

 

16,6

 

18,7

 

19,3

 

P2L3I3

 

20

 

18,1

 

19,6

 

19

 

20

 

15,9

 

15,8

 

17,2

 

20

 

17

 

16,5

 

18,8

 

19,4

 

P3L1I1

 

19,9

 

19,2

 

18,8

 

18,9

 

20

 

16,1

 

15,5

 

17,1

 

20

 

16,9

 

12,2

 

19,6

 

19,7

 

P3L1I2

 

19,9

 

18,7

 

19

 

19

 

20

 

15,7

 

14,8

 

16,8

 

20

 

17,3

 

12,1

 

20

 

19,7

 

P3L1I3

 

20

 

18,6

 

18,8

 

19,3

 

20

 

15,6

 

15,5

 

17

 

20

 

17,3

 

12,1

 

19,6

 

19,6

 

P3L2I1

 

20

 

18,7

 

18,5

 

19

 

20

 

16,2

 

14,9

 

17,1

 

20

 

16,8

 

11,8

 

19,6

 

19,5

 

P3L2I2

 

20

 

19,3

 

18,6

 

18,9

 

20

 

15,8

 

15,1

 

17

 

20

 

17

 

11,7

 

19,8

 

19,7

 

P3L2I3

 

20

 

18,8

 

18,8

 

18,6

 

20

 

15,9

 

15

 

17

 

20

 

16,7

 

12,1

 

19,8

 

19,6

 

P3L3I1

 

20

 

18,1

 

18,5

 

19,1

 

20

 

15,8

 

15,3

 

17,1

 

20

 

17

 

11,8

 

20

 

19,7

 

P3L3I2

 

19,9

 

18,8

 

18,8

 

19

 

20

 

16

 

15,4

 

17,2

 

20

 

16,9

 

11,9

 

19,6

 

19,8

 

P3L3I3

 

20

 

18,6

 

19,1

 

19,2

 

20

 

16

 

16

 

16,9

 

20

 

17,1

 

11,9

 

19,3

 

19,8

 
  1. First results

The expressed results in the shape of index values have been reported in Table 1 in a confidential way, a letter has been given to each operator in a random manner. The data of 4 operators could not been entered because a lack of time.

Though the statistics have not yet been done, some first observations can be made from the results of the 13 operators whose results have been treated. 

The sampling seems the most delicate phase. If the sampling formalities described in the IBD standard project are respected, the index results are very homogeneous for the set of the 3 samplings. One single exception is to be mentioned (operator M, sampling P3) : the index results are homogeneous for the samplings P1 and P2 (IBD values between 16 and 17), and for the sampling P3 (IBD value between 11,7 and 12,2). The rapid examination of the floristic lists shows a very perceptible difference in the composition of the floristic suite indicating in particular that the sampling P3 has not been realized in the same conditions than samplings P1 and P2 and in a general way than the whole samplings of the other operators. So all the inventories are dominated by small Achnanthes/Achnanthidium while the inventories corresponding to sampling P3 of operator M are largely dominated by Amphora pediculus, which is a species almost absent in the other inventories. This seems to translate a sampling realized in conditions, which are not conform to the standard. 

From these rapid observations, it appears thus that when the sampling formalities are respected, a unique inventory realized from one single slide, itself issued from one single sampling is sufficient to characterize the water quality. The repeating of the sampling formalities has been verified and the formalities can be considered as reliable. 

  1. Effect of the variations of the counting formalities on the IBD values 

    Some variations of the counting formalities have been observed. The IBD standard prescribes a counting of 400 individuals. Only 5 operators (B, C, H, J, N) have respected this number, the other operators having frequently overrun this number, sometimes reaching 450 counted individuals. This seems not to have a big effect on the obtained results (see results of operators C and O for example). This value of 400 had been imposed to minimalize the difference due to the fact that all taxa intervening in the IBD calculation must present in all figure cases a relative abundance superior to 7,5 ‰ (3 individuals on 400) and for a great number have a relative abundance superior to an abundance threshold established for each of the taxa of the IBD method. 

  2. Determination effect on the IBD values

If the differences between the IBD values are inferior to 1 unit for each operator taken separately, sometimes very perceptible differences have been noted between operators. These differences are only due to the determination of the taxa from the IBD method. In particular, the IBD distinguishes Achnanthes minutissima Kützing var. minutissima = AMIN (under which figures also Achnanthes minutissima Kützing var. affinis (Grunow) Lange-Bertalot = AMAF), Achnanthes minutissima Kützing var. jackii (Rabenhorst) Lange-Bertalot = AMJA, and Achnanthes biasolettiana Grunow var. biasolettiana Grunow = ABIA.

Certain operators (A, B, C, D, E, I, N and O) have distinguished AMIN and ABIA and give high IBD values (from 18,5 to 20). Others (H, J and M) have only considered AMIN and present IBD values comprised between 16 and 17. Others finally (F and G) have distinguished AMIN and AMJA and present IBD values between 14,8 and 16,8. The variations between the IBD values should thus be due to the determination effort. This is confirmed by the examination of the IBD results obtained from the witness slides prepared from a unique sampling by a same operator. Apart from the natural variability from around 1 unit, the variations of sometimes 4 points between the different operators can only result from the manner to determine the IBD taxa. For example, the abundance of taxa AMIN and ABIA which have been distinguished by the operator O have been added and affected to AMIN for the inventories P1L1l1 and P1L1l2. The IBD values pass respectively from 19,5 and 19,2 to 16,9 and 16,7 and thus approach from the IBD values furnished by the operators not having distinguished AMIN and ABIA. 

2 First conclusions

Finally and as a first approach, we can say that : 

  1. The IBD sampling formalities are strong and when they are respected, a sampling, a slide and an inventory are sufficient to characterize the general water quality by the mean of the IBD ; 
  2. The natural variability of the IBD is of maximum 1 unit. It should be taken into account during the exploitation of results from intra- or interannual follow-ups ; 
  3. The fact to determine the diatoms systematically by the species or following the determination level required by the IBD does not involve significant variations at the IBD note level ; 
  4. The systematic remains the most delicate point in the implementation of diatomic indexes. The IBD has been conceived for an application on monitoring network. Certain taxa, which are difficult to differentiate by non-specialists, have intentionally been regrouped in paired taxa even if their ecology does vary a little bit. Confusions remain possible even with adapted works. Apart the fact that improvements could be brought to the IBD determination keys to solve ambiguities, it is thus indispensable in the same time to organize training sessions to the systematic where identification criteria are explained and recognized by every trainer and to implement a reference slide collection presenting the IBD taxa available for consultation by everybody.

 

References

Agences de l'Eau/Cemagref, 2000 - Guide méthodologique pour la mise en œuvre de l'Indice Biologique Diatomées NF T 90-354, avril 2000, sous presse.

AFNOR, 2000 &endash; Détermination de l'Indice Biologique Diatomées (norme NF T 90-354), sous presse.

Conseil Général des Alpes-Maritimes, 1995 &endash; Etude de la qualité des eaux du bassin du Loup. Juillet & septembre 1994. Conseil Général des Alpes-Maritimes (DATDE) - Agence de l'Eau Rhône-Méditerranée-Corse, 15 p. + annexes.

PRYGIEL J., COSTE M. & ECTOR L., 1999 &endash; Projets d'intercalibration européens et mise en place d'une charte de qualité pour l'IBD (et l'IPS). In : Ector L., Loncin A. & Hoffmann L. (eds), Compte rendu du 17e colloque de l'Association des diatomistes de langue française. Luxembourg, 8-11 septembre 1998. Cryptogamie, Algologie 20: 139-142.

PRYGIEL J. & ECTOR L., 2000 &endash; Groupement d'Intérêt Scientifique " Diatomées des Eaux Continentales " Présentation et premières orientations. In : Ector L., Compère P. & Vidal H. (eds), Compte rendu du 18e colloque de l'Association des diatomistes de langue française. Nice, 14-17 septembre 1999. Cryptogamie, Algologie 21.