Signs and symptoms of root rot in Eucalyptus pellita plantations in Indonesia

This thesis investigates the identity of fungi causing root rot in Eucalyptus pellita plantations in Indonesia. It explores the question of how to detect root-rot infected E. pellita trees from their crown symptoms and whether root-rot disease can be detected at an early stage in the cycle of disease development. This study described and identified the fungi associated with root-rot disease in E. pellita, putatively caused by a species of Phellinus. Macro- and microscopic observations and DNA analysis were used to describe sporocarp morphology and the fungal cultures growing from the symptomatic root samples. Results showed that at the 12 sites investigated, and contrary to expectations, species of Phellinus are less commonly associated with root-rot disease than are Ganoderma philippii and G. mastosporum. There were several potential fungal agents of root rot present at any one site; the sporocarp types observed and the external appearance of the roots were not consistently good indicators of the active pathogen as isolated from roots. A visual assessment method to assess the crown condition of trees in plantations of E. pellita was developed. Eight aboveground variables were used as indicators to classify the E. pellita crowns into five different classes. Repeatability, reproducibility and reliability of this method were examined by conducting repeated surveys. Analysis of the data showed that the crown variables adequately discriminated between crown-condition classes when they were assessed by experienced assessors. However, in repeated surveys which were conducted by less experienced assessors, the crown variables did not sufficiently discriminate between crown-condition classes. Applicability of the method to indicate root-rot incidence and severity at individual-tree level was tested during the first survey. An aboveground assessment of crown-condition using the methodology developed does indicate, at the plot level, the incidence and severity of root rot. At the tree level, the aboveground variables were not significantly correlated with root-rot incidence and severity, as indicated by Spearman correlation (a = 0.05). There was an approximate probability of one out of two that poor crown health was associated with visible signs and symptoms on the roots exposed around the tree. Probability of these indicators for estimating root-rot incidence and severity in an individual tree is 61.4% and 41.6%, respectively. A pathosystem model of Eucalyptus nitens trees artificially inoculated with Armillaria luteobubalina was set up to investigate early physiological responses associated with root-rot infection. Trees were inoculated with two different isolates of A. luteobubalina. Root systems were either wounded or left intact before inoculation. Three photosynthetic parameters, i.e. photosystem II yield (\(F_v\)/\(F_m\)), chlorophyll content and photosynthetic rate (`A_max`) were assessed during six-months of observation. Photosystem II yield was the most sensitive to root-rot infection. A significant difference in `F_v`/`F_m` between the unwounded control and other treatments was observed. Chlorophyll content and photosynthetic rate (`A_max`) decreased for all trees, including controls, during the period of the experiment. The decrease was more marked in treated than control trees. The root systems of inoculated trees were examined and reisolations of A. luteobubalina from symptomatic roots were carried out to confirm infection with Armilllaria of E. nitens trees. This preliminary trial of a model pathosystem was successful and did indicate that there were detectable physiological changes associated with early infection by Armillaria. However the experiment required a longer duration for more widespread physiological changes to be detected. The findings of this project reinforce the importance of comprehensive efforts to reduce the severity and incidence of root-rot disease, especially in terms of early detection. These efforts include the correct identification of fungal causal agent/s, the regular monitoring of crown condition, the application of physiological indicators such as photosystem II yield (`F_v`/`F_m`) to detect stress in plants, including that caused by root rot. Further studies are required to scale up these findings to an operational and cost effective level of plantation management. Pest management strategies, especially for root rot, such as site-hazard rating, species-site matching and mixed-planting systems are discussed in reference to plantation expansion in Indonesia