Discussion and conclusion
This study finds a late bud set of Minnesota (MN) aspen populations in Athabasca, as the southern seed source sets buds later than the local Alberta aspens. The climates variables affect bud set include autumn average temperature, august average temperature, as well as the end date of frost free period. Other environmental variables account for the bud set differences are daylight length, which is closely related with latitude. Thus, the geographical cline of aspen could be observed based on the provenance trials with more genetic information about the populations. However, to predict the northern limit of MN seed source is not promising unless more experiments about the adaptation range of aspen populations are finished to test their responses to more complex local climate conditions. At least, MN aspen is possible to be damaged by early winter frost in Alberta.
As well, a late brown-down trend is also investigated. This may give higher risks of early frost to MN provenance too. The latitudinal gradient of aspen brown-down in western Canada and Minnesota is mixed with temporal differences, which provides evidences of other climate factors related to the brown-down event. The brown-down event more responds to the local daylight length during the critical days in autumn, which is closely related to the latitudinal factor. Multivariate analysis shows the relationship between the brown-down date versus environmental variables. The daylight length, latitude, FFP, MAT etc., all affect the brown-down dates. The annual difference of brown-down date relates to the annual variation of the local climate. The leaf phenological event can last for one month with a radical changing point of foliage reflectance. Because of the high variation of leaf phenology timing, the brown-down data should be validated with the ground truthing data such as the provenance data. In this study they match with each other but the linear relationship is not valid enough.
Thus environmental variables have subtle and complex correlation with bud set and brown-down events. This bud set trait presents the genetic variation at a larger geographical landscape, while the environmental variables such as latitude and longitude indicate the aspen genotype differences. So is the brown-down event. It is possible to predict the transferring distance of aspen in Alberta based on different adaptation traits and climatic risks in case of climate change by combining more provenance trial data with the remote sensing data.
For future studies, there are four possible seed transfer directions (e.g., BS to BP, BP to NBP, BP to FH, and NBP to TP), which all request assessment of adaptation to the local environment. For BS to Alberta, the relationship between bud set phenology and aspen hardiness are not clear. As well more evidences about the climate impact are necessary for validating the statistical results. With the respect of transferring, selecting and breeding superior provenance from the southern region such as Minnesota, it is very necessary to understand the degree to which MN aspen are affected by local frost risks in Alberta as well as the adaptation function of these aspen populations. A longer time range observation of brown-down is necessary for better linking the climatic factors and fall phenology variation. Researches on both fall phenology and other adaptive traits are need as well.
As well, a late brown-down trend is also investigated. This may give higher risks of early frost to MN provenance too. The latitudinal gradient of aspen brown-down in western Canada and Minnesota is mixed with temporal differences, which provides evidences of other climate factors related to the brown-down event. The brown-down event more responds to the local daylight length during the critical days in autumn, which is closely related to the latitudinal factor. Multivariate analysis shows the relationship between the brown-down date versus environmental variables. The daylight length, latitude, FFP, MAT etc., all affect the brown-down dates. The annual difference of brown-down date relates to the annual variation of the local climate. The leaf phenological event can last for one month with a radical changing point of foliage reflectance. Because of the high variation of leaf phenology timing, the brown-down data should be validated with the ground truthing data such as the provenance data. In this study they match with each other but the linear relationship is not valid enough.
Thus environmental variables have subtle and complex correlation with bud set and brown-down events. This bud set trait presents the genetic variation at a larger geographical landscape, while the environmental variables such as latitude and longitude indicate the aspen genotype differences. So is the brown-down event. It is possible to predict the transferring distance of aspen in Alberta based on different adaptation traits and climatic risks in case of climate change by combining more provenance trial data with the remote sensing data.
For future studies, there are four possible seed transfer directions (e.g., BS to BP, BP to NBP, BP to FH, and NBP to TP), which all request assessment of adaptation to the local environment. For BS to Alberta, the relationship between bud set phenology and aspen hardiness are not clear. As well more evidences about the climate impact are necessary for validating the statistical results. With the respect of transferring, selecting and breeding superior provenance from the southern region such as Minnesota, it is very necessary to understand the degree to which MN aspen are affected by local frost risks in Alberta as well as the adaptation function of these aspen populations. A longer time range observation of brown-down is necessary for better linking the climatic factors and fall phenology variation. Researches on both fall phenology and other adaptive traits are need as well.
