European Larch breeding programme in Poland Jan Kowalczyk, Jan Matras, Marek Rzońca Forest Research Institute Sękocin Stary, Poland Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
European Larch breeding programme in Poland
Jan Kowalczyk, Jan Matras, Marek Rzońca
Forest Research Institute
Sękocin Stary, Poland
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Outline
• European larch in Poland – Why Larch? – Distribution – Local circumstances
• Brief history of larch research in Poland • Present state of breeding activities
– Strategic aims – Methods – Results
• Plans for future - expected results
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Why Larch?
– Fast growing tree
– Easy to plant
– Visible in the beginning of the grow
– Not seriously affected by pest and disease in Poland
– Durable wood
– Larch is beautiful
0
5
10
15
20
25
0
50
100
150
200
250
300
350
400
450A
rea
of
larc
h [
Ths.
hec
tare
s]
Larch area
% total conifers
[%]
Pâques, L. E. and all. (2013). Forest Tree Breeding in Europe: Current State-of-the-art and Perspectives (Vol. 25). Springer, pp 16.
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Local circumstances
– Distribution is scattered
– Natural distribution– how it influence the practical utilization now?
– Larch was/is often planted
– Seeds was moved and also Japanese larch was introduced
– Hybrids !!! (vigorous grow – quality and performance of future generations)
Brief history of larch research in Poland
IUFR
O 1
94
4
IUFRO 1957
Lub
awka 1
93
2
Ro
gów
19
49
SSO 1
19
66
IBL 1
96
7
IBL 2
00
0
INR
A 2
00
3
Den
dro
Gen
1940 1930 2000 1970
18 provenances 10 from Sudeten 8 form Alps
23 provenances from Poland—6 trials located in all Poland
Kolanów1946 48 provenances from 4 sub area
4 provenances 3 from Polish Lowlands and 1 Sudeten (standard)
20 half sib familes from Świętokrzyskie Mt.
157 half sib families from Świętokrzyskie Mt. 2 trials in Poland and 3 in France (IBL—INRA collection)
67 half sib families from Jesiennik region 1 trials in Poland and 2 in France, 2 in Czech, 2 in Belgium (INRA series)
Several SSO in different age located in all Poland more then 600 half sib families
Breeding value families tested in Zwierzyniec
-2,00
-1,50
-1,00
-0,50
0,00
0,50
1,00
1,50
1730
1387
1857
1417
1702
1849
1402
1413
1419
1397
1408
1391
1420
1400
1394
1405
1418
1421
1388
1851
1396
1410
DBH
Stem straightness
BV
Families
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Conclusions from research
• The large differences in the grow of the population was observed even in a relatively small area
• The regions of occurrence of larch Polish and the Sudeten may be regarded as homogeneous areas.
• For this reason, the breeding of larch have to be conducted primarily at the level of particular seed sources (forest stands, families)
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
Breeding programme
Classical low intensive
Seed stands
Plus tree selection
Classical
Progeny testing
Backward selection
Improvements
• As regards the indigenous European larch, there occurs locally an ‘infestation’ of the gene pool with genes from Japanese larch, in view of the easy crossing of these two species, and this leads to an erosion of the genome of the indigenous species. For this reason it is particularly important to take note of the species purity of the seed bases proposed for confirmation. The existing stands of Japanese larch and of hybrid ones in the vicinity of the populations selected and qualified as basic seed material should be successively removed.
Program … for years 2011-2035
Demand for seed of European larch according to felling requests
0
200
400
600
800
1000
1200
1400
1600
2001 2002 2003 2004 2005 2006 2007 2008 2009
[kg]
Year
Larch seed sources in Poland
Seed sources Number Area [ha] % to all conifers
Selected stands (reserved seed stands)
82 445 4
Plus trees 966 15
Seedling seed orchards 24 172 30
Seed orchards 38 247 30
Intensive breeding
• It is planned to establish breeding populations in the Sudety and in the Świętokrzyskie Mountains. When choosing the initial material use should be made of the existing plus trees and the results of progeny trials. It is recommended to conduct the breeding program in international cooperation with such countries as the Czech Republic and France.
Program … for years 2011-2035
Cel pracy
Seminarium Zakładowe, Sękocin, 4 III 2014
Aims
• Estimation of breeding value half sib families grooving on SSO
• Selecting the best family (tested)
• Creation of SO 1.5 generation and clonal archive (backward selection method)
• Establishing new breeding population
Seminarium Zakładowe, Sękocin, 4 III 2014
Seed transfer
rules
Local circumsta-
nces
Technical aspects,
Time
Cost
Seed demand
Analyses
Condition of trials and
minimum n. of trees per family
Existing low
SO 1.5 gen Existing infrastructure (trial, SSO)
Inwentaryzacja PUN i doświadczeń rodowych
Seminarium Zakładowe, Sękocin, 4 III 2014
SSO 1
SSO 68
SSO 46
SSO 89
SSO 90 SSO 67
Seminarium Zakładowe, Sękocin, 4 III 2014
1 46 67 68 89 90
Plantation year
1966 1992 1996 1996 2001 2001
Number of trees
4000 1911 1669 3784 1837 1849
Number of families
20 36 42 52 67 97
Area [ha] 7,2 12,1 5,3 14,8 5,2 5,1
Studied SSO description
1. Data cleaning
2. Removing extremes – ( 3SD<x<3SD)
3. Calculation for each trials – variance i BV for features
4. Stand arising data
5. Summary index
6. Selection the best families – SO 1.5 generation
7. Calculation BV for single tree
8. Choosing the best trees for crossing
Seminarium Zakładowe, Sękocin, 4 III 2014
Recepture
Przykładowe zróżnicowanie dla PUN1
> summary(DBH.aov<- aov(DBH~Family+Family*Block, data=plik))
Df Sum Sq Mean Sq F value Pr(>F)
Family 19 341628 17980 5.614 3.61e-13 ***
Block 9 134645 14961 4.671 4.77e-06 ***
Family:Block 171 520657 3045 0.951 0.654
Residuals 764 2446870 3203
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Seminarium Zakładowe, Sękocin, 4 III 2014
TUKEY TEST TO COMPARE MEANS Confidence level: 0.95 Dependent variable: DBH Variation Coefficient: 13.60573 % Independent variable: Family Factors Means 5p 453.657407407407 a 7p 443.820754716981 ab 1p 442.84 ab 11p 439.244680851064 abc 4p 424.722222222222 abcd 19 422.96511627907 abcd 18p 422.798076923077 abcd 15 422.518867924528 abcd 6 415.840425531915 abcd 13 413.795454545455 abcd 17p 413.539215686275 abcd 14p 413.142857142857 abcd 20p 409.48 bcd 16 404.076923076923 bcd 3 399.25 cd 2 395.878048780488 d 10 394.7 d 9p 394.031914893617 d 12p 391.12037037037 d 8 384.776315789474 d
Independent variable: Block Factors Means 10 434.021739130435 a 3 426.572222222222 a 8 424.833333333333 a 5 421.016304347826 ab 9 420.103092783505 abc 4 414.651685393258 abc 7 414.556701030928 abc 2 414.546391752577 abc 1 395.775510204082 bc 6 395.024752475248 c
Tytuł slajdu
Seminarium Zakładowe, Sękocin, 4 III 2014
> summary(m.lm) Linear mixed model fit by REML ['lmerMod'] Formula: DBH ~ Block + (1 | Family) Data: plik REML criterion at convergence: 10477.41 Random effects: Groups Name Variance Std.Dev. Family (Intercept) 303 17.41 Residual 3174 56.34 Number of obs: 964, groups: Family, 20 Fixed effects: Estimate Std. Error t value (Intercept) 396.096 6.899 57.41 Block2 17.838 8.074 2.21 Block3 28.941 8.243 3.51 Block4 18.039 8.258 2.18 Block5 25.017 8.190 3.05 Block6 -1.838 8.007 -0.23 Block7 16.474 8.083 2.04 Block8 28.522 7.838 3.64 Block9 22.991 8.087 2.84 Block10 37.732 8.197 4.60 > 4*303/(303+3174) [1] 0.3485764
> ranef(m.lm, level=1) $Family (Intercept) 10 -15.1567705 11p 20.1488115 12p -21.5872893 13 -0.2192024 14p -3.2430080 15 5.9509466 16 -7.9591342 17p -3.3589419 18p 6.6019899 19 6.0476027 1p 22.4833640 2 -15.6581160 20p -5.0588889 3 -13.3290967 4p 8.7460731 5p 31.7171197 6 0.1086741 7p 23.6306150 8 -23.6215560 9p -16.2431928
Wyniki – najlepsze rody
Seminarium Zakładowe, Sękocin, 4 III 2014
site Reg fam ind_me
an
h_perc d_perc str_perc cro_per
c
89 604 1845 1.63 3.67 7.00 1.76 -4.16
90 604 1845 1.63 3.66 3.51 -1.06 -1.34
70 852 684 1.57 16.82 7.24 12.47 7.37
89 604 7006 1.54 0.72 0.33 1.91 0.40
90 604 7006 1.54 3.52 1.99 0.63 -1.13
89 604 1833 1.47 3.69 4.48 1.45 -0.48
90 604 1833 1.47 -0.13 1.88 0.64 1.22
89 604 7696 1.44 2.65 4.27 0.63 -0.58
90 604 7696 1.44 1.38 2.43 0.71 1.78
1 604 5p 1.42 6.09 14.90 -5.32 -20.87
70 852 647 1.35 8.71 4.42 14.50 3.72
89 604 1837 1.32 2.49 0.85 0.44 2.54
90 604 1837 1.32 2.84 4.02 -0.31 -3.48
69 803 601 1.29 12.58 26.49 13.53 -12.16
89 604 1826 1.27 -3.28 -8.36 0.88 7.26
90 604 1826 1.27 2.66 2.79 1.03 11.67
70 852 673 1.25 24.50 23.11 6.77 -8.44
89 604 7017 1.22 3.69 5.14 0.56 0.12
90 604 7017 1.22 0.80 0.17 0.43 0.66
41 803 624 1.05 11.42 17.85 -1.28 -5.86
69 803 624 1.05 11.88 7.68 8.25 7.70
89 604 1849 1.02 2.38 2.12 0.74 0.34
90 604 1849 1.02 -0.08 1.78 0.75 -0.22
Index=1.25*H_BV + DBH_BV
+str_BV + crown_BV
Zysk w % - efekt selekcji
Seminarium Zakładowe, Sękocin, 4 III 2014
SSO h_effect d_effect str_effect cro_effect
1 3.78 11.94 2.50 -11.10
41 6.90 8.68 5.97 -2.38
69 12.79 15.68 4.37 -0.58
Average 7,8 12,1 4,3 -4,7
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
• It is reasonable to choose the best families to establish SO 1.5 generation – we know what we get
• Selection priorities - growth and quality • Not take into account the characteristics of wood in the
selection - task for the future? • We do not include survival • Not take into account the reaction to action of biotic and
abiotic • SSO has a specific history and it affects the interpretation of
the results - necessity more digging in data • The average gain for DBH - 12% for straightness 4%, crown
width -4%
Summary - conclusions
Forest Tree Breeding Conference, Prague, Czech Republic, August 25-29, 2014
1. Research is conduced by - dendrogen.pl
• IBL, IDPAN, UP Poznań, UR Kraków, SGGW
• UAM Poznań, ID PAN, UKW
2. The grant is funded by State Forest – BLP 397 (2013- 2017)
3. Thank you for my coauthors
• – Jan Matras , Marek Rzońca
• Technical staf– Jerzy Przyborowski, Marek Rzońca,
Marcin Konieczyński, Danuta Pieniążkiewicz, Tomasz Wojda, Paweł Przybylski,
Władysław Kantorowicz