Environmental Response Of Radial Growth Of Larix Principis-rupprechtii Trees At Multiple Time Scales In Liupan Mountains

The impacts on forest trees by both the long-term climatic change and the short-term meteorological extreme events have been increasing in recent decades.Therefore,it is urgent and important to improve the forest stability in the dryland regions and to coordinate the conflict between forest and water,based on the deep understanding and quantitative description of the impact by environmental factors and stand structure on the tree growth at multiple temporal scales?e.g.,daily and yearly scale?.However,the most current studies at yearly scale focused on the relations between annual radial growth and the annual means of meteorological factors,without paying enough attention to the effect of stand structure.At short time scales,the response mechanism of stem radial growth to environmental factors was still not clearly understood,the lack of daily radial growth model,which couples the effect of multiple factors,limits the deep understanding and accurate prediction of tree growth in a changing environment.Therefore,this study selected the sample trees with different dominances in the larch?Larix principis-rupprechtii?plantations of Liupan Mountain where is the semi-humid climate zone,to analyze the response of stem radial growth to environmental change at the yearly and daily scales by using ring width measurement and dendrometer monitor,to determine the main factors influencing the tree radial growth at different time scales,and to establish the radial growth models coupling the multiple influencing factors at daily and yearly scales,with the aim of clarifying the response mechanisms of the stem radial growth of Larix principis rupprechtii trees to the environmental changes including the short-term meteorological extreme events.The main results are as follows:?1?Within the age range studied?0-45 years?,the larch trees growing at better sites showing a small difference in the basal area increment?BAI?among the trees with different dominances at the age range of 0-20 years,but gradually increasing difference at the age range higher than 25-30 years;however,the BAI difference among the differently dominant trees appears earlier?at the age of 15 years?if growing under worse conditions?e.g.,too high tree density,poorer light condition,thin soil?and keeps basically stable after the age of 25 years.The response of basal area increment?BAI?among the trees with different dominances to stand density followed a exponential decrease equation.Under the same conditions of site quality,for the same tree age,the greater the difference in tree dominance,the greater the difference in BAI.Therefore,a BAI model was established to consider the effects of stand density,tree age and tree dominance,besides the effects of annual precipitation and annual mean air temperature,in the form of multiplicative equation,and with high accuracy?R²=0.81 and 0.83?.These models can well reflect the response of annual stem radial growth of trees to changing environmental and stand structure conditions.?2?The diurnal variation of stem radius is closely related with meteorological conditions.In sunny days,the stem radius showed a regular change of daytime shrinkage and nighttime expansion,while in cloudy and rainy days,the stem radial shrink does not appear,and the overall tendency is a"platform"or continuous increase.The radial growth of all the larch trees with any dominance is faster at early stage and slower at late stage of the growing season.The growing season can be divided into three growth periods:the rapid growth period?May to June?,the slow growth period?July to August?,and the dormant growth period?September to October?.?3?As an important factor affecting the diurnal stem radial variation,tree transpiration needs to be understood and quantified accurately.The diurnal variation of both sap flow density?Js?and daily transpiration of single trees?T?with different dominances showed a single peak curve.The greater the tree dominance,the earlier the start time and the later the stop time of sap flow density,and the faster rate to reach its peak.The peak values of sap flow density or daily transpiration of single trees exhibited the following order in every month:dominant tree>sub-dominant tree>intermediate tree>pressed tree.The response of daily transpiration of single trees?T?followed a binomial equation to potential evapotranspiration?PET?,and followed a saturated exponential growth function to relative extractable water?REW?of the 0-60 cm soil layer.The models of T considering the effects of PET and REW were fitted using multiplicative equation for each tree dominance.For dominant trees it is T_d=?-0.3889PET²+6.7521PET-2.1419?×?1.7817?1-exp?-2923.7950REW???;for sub-dominant trees is T_sd=?-24.0219PET²+261.6174PET-104.8776?×?0.0269?1-exp?-2588.0041REW???;for intermediate trees is T_m=?-37.6743PET²+515.6138PET-1203.0833?×?0.0073?1-exp?-2662.6181REW???;for pressed trees is T_p=?-10.7981PET²+280.80434PET-68.1772?×?0.0051?1-exp?-3765.2377REW???,respectively.These modes have an accuracy?R²?between 0.64-0.73,and they can well predict and evaluate the variation of daily transpiration of single trees.?4?The monthly means of daily maximum shrinkage?MDS?showed a bimodal curve within the growing season,with the first peak at June and the second peak at August.The responses of MDS to potential evapotranspiration?PET?,soil temperature?Ts?and daily transpiration of single trees?T?follow a function of exponential growth that tends to be saturated,and the response to the relative soil water content?REW?follows a quadratic polynomial function,regardless of the rapid,slow,and dormant growth period.Based on these response equations,the MDS models considering the effects of PET,Ts and REW were fitted in the form of multiplicative equation,for each tree dominance in each growth periods.These modes?R²?have an accuracy between 0.58-0.77,and they can well predict and evaluate the variation of MDS in each growth period.?5?The monthly means of the daily stem radial expansion amplitude?EA?showed a bimodal curve within the growing season.The daily EA increases with rising daily rainfall amount?P?,with a rapidly increase and large fluctuate when the P is below 10 mm,and the increasing rate of EA becomes slow-down when the P is above 10 mm.The responses of EA to the daily means of air temperature?Ta?and vapor pressure deficit?VPD?follow a binomial function,and the response to the relative soil water content?REW?follows an exponential growth function that tends to saturate,at any growth periods.Based on these response equations,the EA models considering the effects of P,Ta,VPD and REW were fitted in the form of multiplicative equation,for each tree dominance in each growth periods,and these models can well predict and evaluate the EA variation in any growth periods.?6?The daily stem radius increment model can be formed when coupling the daily radial expansion model and the daily radial shrinkage model.This daily stem radius increment model can predict the daily radial growth,and also the yearly radial growth when accumulating all the daily radial growth within a year?growing season?.Such predicted annual radial growth can reflect the effects on tree radial growth by the short-term environmental changes,which can be used to make up the gap that the normal tree growth model at yearly scale cannot reflect the effects of short-term environmental changes on stem radial growth.