| Forest ecosystems were recognized as an important role in global carbon(C) cycling and in C sequestration. To quantify the potential C sink and understand its implication to the regional carbon budget and future forest management is important. This thesis is based on the data of field inventory and biomass survey for the different ages Chinese fir stands in Jiangle County, Sanming City, Fujian Province. This study examines the C concentrations of various above-ground biomass components (bark, stem, foliage, living branch and dead branch), the understory (shrub, herb), litter and soil. The C storage of trees, understory, litter, and soil was also estimated. Meanwhile, the author established biomass and carbon inversion model based on high resolution satellite images. The main conclusions are summarized as follows:(1) Chinese fir belongs to shallow root system of plant, and the trees biomass are mainly concentrated in the aboveground. The biomasses of aboveground represents 76.09% to 78.49% of the whole, while that of belowground takes up 21.51% to 23.91%. The aboveground biomass of tree organs was ranked in descending order as stem> roots> branch> foliage except for young stand. The belowground biomass was ranked in descending order as large roots> thick roots> medium roots>small roots, and ranked as 20-40cm> 10-20cm>40-60cm>60-80cm>80cm by different soil depth.(2) The aboveground C concentration Chinese fir plantation is greater than that of the belowground part and the mean/average C concentration approximately is 46.97%. The C concentration differed significantly at the same ages among the different organs, but not significant in the same organs at different ages. In this study, the C concentration of aboveground tree organs was ranked in descending order as foliage (49.58%)> dead branch (47.81%)> stem (47.62%)> living branch (47.45%)> bark (47.31%), and in the belowground was large roots (46.66%)> medium roots (46.17%)> small roots (46.13%)> thick roots (45.80%).(3) The shrub C concentration is greater than that of the herb. The average shrub C concentration approximately is 43.56%, and the mean C concentration of branch, foliage and root are 44.49%,43.50% and 42.71% respectively. The mean herb C concentration approximately is 36.10%, and the mean C concentrations of above- and below-ground are 41.72% and 30.47%. For the litter layer, the mean C concentration was 40.41%, and ranked as:no decomposition layer> little decomposition layer> completed decomposition layer. The C concentration in the mineral soil decreased with increasing soil depth. The C concentration differed significantly among different soil depth and different ages.(4) A compatibility of biomass model was constructed using diameter at breast height (DBH) and elements of total tree biomass. The equation was subsequently utilized to estimate tree carbon storage. Based on data of tree biomass, we controlling jointly from level to level were used to establish the mono-element compatible tree biomass model of Chinese fir, and the precision of models reached beyond 93.4%.The carbon storage of Chinese fir plantations and their tree layers increased with the increasing of age at first, and reached maximum at maturity. With the highest proportion of total biomass, tree layer biomass approximate between 89.17 and 223.01 t·hm-2;The shrub layer biomass was approximate from 0.44 to 13.27 t·hm-2, and the herb layer biomass was approximate from 1.61 to 4.90 t·hm-2.The litter layer biomass was approximate from 2.97 to 5.75 t·hm-2(5) The C storage in Chinese fir plant biomass in stands aged young, middle, near-mature, mature and over-mature stand was 182.61,182.29,243.69,245.16 and 210.05 t·hm-2 respectively. The vegetation layer accounted for between 44.49% and 61.95%, the litter layer accounted for between 0.73% and 1.25%, and the soil layer accounted for between 37.20% and 54.76% of total C storage. The tree layer C storage was largest in vegetation layer, accounting for between 96.47% and 97.76%, while the undergrowth C storage accounted for between 2.24% and 4.50%.(6) Based the plots data, with the quadratic function as the basic model, the random effects of plots to crown area were considered, then the crown area-diameter linear mixed effects (LME) model was established The LME model with the constant plus function as the variance functions showed the highest precision for the model of crown area and diameter model. Results showed that crown area-diameter model had a better fit effect and was selected as the base model. The model accuracy of adj-R2 increased from 0.3267 to 0.4953, and the verify accuracy of adj-R2 increased from 0.0889 to 0.3127.(7) Crown surfaces of single tree level of China-fir plantations in different age groups were extracted by using high-resolution satellite images and watershed segmentation algorithm with control tag and morphological image analysis together. On average, results show that young, mature and over-mature stage get higher precision, while both middle-aged and near-mature stage get poor accuracy, of which the young, middle-aged, near-mature, mature and over-mature forests achieves average tree number accuracy of 96.02%,85.11%,79.77%,86.92%,93.15%,respectively, and user accuracy of 82.93%, 60.46%,54.00%,69.65%,79.96%, respectively.(8) Based on the crown extraction results, combining the linear mixed effects crown area-DBH model and tree biomass prediction modeling, the estimation of Chinese fir plantations DBH and the estimation of the biomass during different development stages were implemented, with the results indicating that the predict accuracy during different growth stages is 67.95%,71.94%,73.35%,96.10%, 9315%, respectively. |
PDF Full Text Request