Effects of previous land use on genotype-by-environment interactions in two loblolly pine progeny tests

Abstract

Loblolly pine (Pinus taeda L.) is widely used for forestry in southern Brazil, with most seeds coming from clonal seed orchards. This study investigated the effects of previous land use on the adaptability, stability, and productivity of loblolly pine progenies in southern Brazil, aiming to produce a second-generation clonal seed orchard (CSO). Two progeny tests were installed in southern Brazil. In FARA, 52 open-pollinated loblolly pine progenies were tested, and in TUNAS, 65 open-pollinated loblolly pine progenies were tested, 35 of which were common to both tests. FARA is a rotation area previously used for annual agronomic crops (oats, corn, soy, beans), whereas, TUNAS is a recycled loblolly pine plantation forest. The diameter at breast height (dbh) and stem height were measured at six years of age, and the wood volume was calculated. Genetic parameters were estimated using REML/BLUP method in SELEGEN software. The adaptability, stability, and productivity of the progenies were estimated by the relative performance harmonic mean of the breeding values method (MHPRVG). The breeding values for volume were used to estimate genetic gains through selection. The heritability values were slightly higher in FARA compared to those in TUNAS. Most of the individual narrow-sense heritability values were between 0.10 and 0.20 for growth traits. And most of the family-mean heritability values were above 0.50. In FARA, the selection strategy resulted in up to 26.37% genetic gain, while for TUNAS, it reached about 20%. GE effects are significant for volume. Because of the significant GE effects it is necessary to consider previous land use in the selection of genotypes for the formation of loblolly pine CSOs in the tested region. Considering the MHPRVG ranking, the maximum genetic gain estimated is 13.20%. The choice of more stable genotypes can circumvent the challenges imposed by crop rotation in the tested areas.