Font Size: a A A

Carbon Density And Its Spatial Heterogeneity Of Pinus Massoniana Forest Ecosystem In Jiangxi Province

Posted on:2019-01-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:P PanFull Text:PDF
GTID:1363330548987900Subject:Forest management
Abstract/Summary:PDF Full Text Request
The global climate change characterized by the increase of atmospheric CO2concentration and temperature has become one of the hot topics of governments and scientists.Forest ecosystem is the main body of terrestrial ecosystem and the largest terrestrial carbon pool.It plays an important role in the global carbon cycle and mitigation of climate warming.Therefore,it is widely concerned that how to accurately estimate forest carbon storage.The carbon density is an important index to measure the carbon sequestration capacity of the forest ecosystem.The carbon density and its spatial characteristics of different forest ecosystems in the same area and the same forest ecosystem in the different areas are different due to the influence of natural and human factors.Pinus massoniana is the main native coniferous tree species in subtropical area.And it is also one of the main forest types in Jiangxi Province,which area accounts for28.9%of the province's arbor forest area.This study investigated Pinus massoniana forest in Jiangxi Province to analyze the distribution characteristics of its carbon density and to explore the spatial variability of its carbon density by using geographically weighted regression?GWR?and other spatial regression models.The study results can provide a scientific basis for the management and planning of Pinus massoniana forest ecosystem carbon sequestration in the study area,and also provide a theoretical and method reference for further revealing the influence factors and spatial heterogeneity of forest ecosystem carbon density.This study has important value in theory and application.The main results and conclusions were as follows:?1?The carbon content of Pinus massoniana forest ecosystem and its two types of natural forest and plantation in different age classes were as follows:The tree layer was leaf>trunk>branch>root.The shrub layer was branch>leaf>root.The herb layer was aboveground>underground.The litter layer was undecomposed layer>semi-decomposition layer.The different components carbon conernt of tree layer,shrub layer and herb layer in Pinus massoniana forest ecosystem and its two types of natural forest and plantation were not obvious law.The soil layer carbon conernt of Pinus massoniana forest ecosystem and its two types of natural forest and plantation increased with the increase of stand age.The carbon content increase range of plantation in different age classes is higher than that of natural forest.?2?The different layers carbon density of Pinus massoniana forest ecosystem and its two types of natural forest and plantation in different age classes were as follows:The tree layer carbon density were 45.68 t·hm-2,48.64 t·hm-2 and 42.62 t·hm-2,and increased with the increase of stand age,which were trunk>branch>root>leaf.At the same age class,the tree layer carbon density of natural forest was higher than plantation.And the carbon density of natural forest was significantly higher than plantation at age class??P<0.05?.There were no significant differences at other age classes?P>0.05?.The understory vegetation layer carbon density were 4.52 t·hm-2,5.06 t·hm-2 and 3.95 t·hm-2,and had no obvious law with the increase of stand age.The understory vegetation layer carbon density of natural forest was higher than plantation at the same age class.The carbon density of natural forest was significantly higher than plantation at age class??P<0.05?.The litter layer carbon density were 1.81 t·hm-2,1.88 t·hm-22 and 1.73 t·hm-2,and had no obvious law with the increase of stand age.The litter layer carbon density of natural forest was higher than plantation at the same age class.The soil layer carbon density were 70.32 t·hm-2,77.54 t·hm-22 and 62.87 t·hm-2,and increased with the increase of stand age.And at the same age class the soil layer carbon densityof natural forest was higher than plantation at the same age class.?3?The total carbon density of Pinus massoniana forest ecosystem and its two types of natural forest and plantation in different age classes were respectively 122.33 t·hm-2,133.12 t·hm-2 and 111.17 t·hm-2,and increased with the increase of stand age.At the same age class,the total carbon density of natural forest was higher than plantation.And the carbon density of natural forest was significantly higher than plantation at age class??P<0.05?.The ratio of carbon density to total carbon density at all levels was soil layer>tree layer>understory vegetation layer>litter layer.The proportion of soil layer decreased with the increase of stand age.The proportion of tree layer was on the contrary.The proportions of understory vegetation layer and litter layer increased firstly and then decreased with the increase of stand age.?4?The global Moran's I value of carbon density in Pinus massoniana forest ecosystem and its tree layer,understory vegetation layer,litter layer and soil layer were greater than zero and decreased with the increase of distance.The carbon density of the ecosystem and its tree layer,litter layer and soil layer almost had no spatial autocorrelation when the distance exceeded 400 km.However,the carbon density of understory vegetation layer had almost no spatial autocorrelation when the distance was more than 450 km.The results showed that the carbon density of Pinus massoniana forest ecosystem and its different layers had significant positive spatial autocorrelation.The spatial distribution of carbon density was aggregated tended to be random with the increase of distance.The Z value characteristics of local Moran's I showed that the spatial autocorrelation of carbon density in Pinus massoniana forest ecosystem and its different layers increased with the increase of distance,which indicated that the local spatial distribution was not uniform.And the difference of local spatial distribution gradually increased with the increase of distance.?5?The carbon density of Pinus massoniana forest ecosystem and its different layers were fitted by the semi-variogram model.The results showed that the exponential model was the best fitting model for the carbon density of tree layer and understory vegetation layer,which the ratio of nugget and sill were 0.38 and 0.50.The spherical model was the best fitting model for the carbon density of litter layer,soil layer and ecosystem,which the ratio of nugget and sill were 0.49,0.35 and 0.36.The fitting results showed that the carbon density of Pinus massoniana forest ecosystem and its different layers had spatial moderate correlation.The carbon density of Pinus massoniana forest ecosystem and its tree layer,understory vegetation layer,litter layer and soil layer were spatially interpolated by Kriging according to the parameters of semi-variogram model,and mainly concentrated on85.14-153.52 t·hm-2,39.71-81.30 t·hm-2,3.44-6.82 t·hm-2,1.49-2.43 t·hm-2 and33.86-124.45 t·hm-2.The carbon density spatial distribution law of Pinus massoniana forest ecosystem and its different layer were similar.The trend was low in the middle region and high in the peripheral region,which was consistent with the landform characteristics in the study area.?6?The multiple linear stepwise regression method was used to screen out the influence factors which were related to the carbon density of Pinus massoniana forest ecosystem and its different layers,and did not exist multiple collinearity.The results showed that the main influence factors of carbon density in each layer were different.And the influence factors of tree layer carbon density were altitude,slope,diameter at breast height?DBH?,stand density,annual temperature and annual precipitation.The influence factors of understory vegetation layer carbon density were altitude,slope position,understory vegetation coverage,stand density and annual precipitation.The influence factors of litter layer carbon density were altitude,understory vegetation coverage,stand density,annual temperature and annual precipitation.The influence factors of soil layer carbon density were altitude,slope,slope position,soil depth,stand age,annual temperature and annual precipitation.The influence factors of ecosystem carbon density were altitude,slope,soil depth,DBH,annual temperature and annual precipitation.?7?The carbon density and its influence factors of Pinus massoniana forest ecosystem and its different layers were respectively fitted by the ordinary least squares model,spatial error model,spatial lag model and geographically weighted regression model?GWR?.The results showed that GWR was the best fitting model for carbon density of Pinus massoniana forest ecosystem and its tree layer,understory vegetation layer,litter layer and soil layer,which R2 were 0.8043,0.8638,0.6958,0.6697 and 0.7767 respectively,and AIC and MSE were smaller than other three models.The tests showed that the GWR could effectively reduce the spatial autocorrelation of the model residual.?8?The fitting coefficients of GWR were used to analyzing the spatial influence of factors on carbon density of Pinus massoniana natural forest and plantation ecosystem and its different layers.The results showed that the carbon density of Pinus massoniana natural forest and plantation ecosystem and its different layers were consistent with the growth or decrease trend of site,vegetation and meteorological factors.With the increase of site,vegetation factors and annual temperature,the carbon density of Pinus massoniana natural forest ecosystem and its different layers increased faster than plantation,and decreased more slowly than plantation.With the increase of annual precipitation,the carbon density of Pinus massoniana natural forest ecosystem and its tree layer,understory vegetation layer,soil layer increased more slowly than plantation,and the carbon density of litter layer decreased faster than plantation.
Keywords/Search Tags:Pinus massoniana, forest ecosystem, carbon density, distribution characteristic, spatial heterogeneity, geographically weighted regression
PDF Full Text Request
Related items