Thumbnail Image

Predicting site productivity drivers for Podocarpus totara and modelling its growth with 3-PG

Forests have been identified as key mitigation strategies to reduce net greenhouse gas emissions as they sequester and store carbon through growth. The Emissions Trading Scheme (ETS) and One Billion Tree (OBT) programme aim to increase tree plantings to offset these emissions by identifying exotic and indigenous tree species suitable for carbon sequestration across New Zealand. However, the rate of planting indigenous tree species has been slow as the species ability to sequester carbon is not well documented and legislation has prevented harvesting. The endemic conifer tōtara (Podocarpus totara D. Don) is an indigenous tree species which could be suitable for timber and carbon forestry. Previous studies have identified that tōtara can grow across a wide range of climatic and environmental conditions, however, few studies have investigated the key site productivity drivers and examined how they potentially influence productivity across sites. The aims of this study were to identify the main productivity drivers for tōtara across sites in the North Island of New Zealand and model growth using the 3-PG (Physiological Principles in Predicting Growth) forest growth model developed by Landsberg and Waring (1997). This study had two main hypotheses; 1) site fertility and temperature would be significant drivers of tōtara productivity and 2) the 3-PG growth model would be able to predict the growth of tōtara (stem diameter, basal area, volume, height and stand density) moderately well, providing a good fit to both the calibration and validation datasets with minimal error (RMSE). To test these hypotheses, 21 previously described planted and naturally regenerated tōtara stands across the North Island were selected. Fifteen sites were re-measured to investigate site productivity drivers for tōtara and to calibrate the 3-PG forest growth model. A multiple linear regression analysis using the backward elimination method was conducted to examine 22 growth related variables. Reineke’s (1933) stand density index (SDI) approach was used to identify trends in mortality as a function of stand density and size across sites. In addition, four volume equations by Ellis (1979), Coomes et al. (2002), Beets et al. (2012), and Todoroki and Steward (2019), that had previously been used to estimate volume of tōtara were tested to identify the most appropriate fit to two tōtara specific datasets. After model calibration, six additional sites, drawn from previous mensuration data were used to validate the model. The results from this study identified that climatic and soil physiochemical properties were significant drivers of tōtara productivity in planted and naturally regenerated stands. Temperature (max, mean and min), rainfall, elevation, soil total phosphorus, HCP ECa (horizontal coplanar receiver, apparent electromagnetic conductivity), and other soil macro and micronutrients (e.g. potassium and manganese) were selected as significant drivers of tōtara productivity from the multiple linear regression analysis; thus Hypothesis 1 of this study was supported. Currently, it is unclear how much these variables contribute to productivity and further research is recommended to investigate the level of contribution these significant drivers have on tōtara. Further investigation of more sites with collection of detailed growth and soil sampling data is recommended to develop a site fertility index for tōtara to quantify the role of fertility across sites. The volume equation developed by Todoroki and Steward (2019) provided the most appropriate fit to the tōtara dataset. A species-specific equation can therefore improve current estimates of volume in stands. In addition, the SDI index successfully identified relative stand densities for maximum and optimal stocking and can provide a basis for informing density management across tōtara sites. The 3-PG forest growth model predicted the growth of tōtara across sites with variable success. The 3-PG model provided reasonable estimates of stem diameter, height and stand density across the calibration sites (R2 ≥ 0.60 with low RMSE). However, the model consistently over-estimated basal area and volume. The model performed poorly (R2 ≤ 0.50) when tested with the validation dataset, and significantly over-estimated all growth variables apart from stand density. Therefore, the results did not support Hypothesis 2. This result could be due to the limited quantity and quality of data available and further investigation into the model’s parameters and algorithms is recommended. Overall, even though the results were less accurate than expected, they indicate that process-based models, like 3-PG, as opposed to empirical models, can be used to model species with limited datasets to identify base line information on growth over time and identify where additional research efforts should be directed to improve predictions. This thesis emphasises the need to investigate a wider range of tōtara sites across varied soil fertility and productivity to improve the quality of the data available for growth modelling and forest management. This should include regular re-measurement of growth and long-term monitoring of climatic and edaphic factors across sites. This study has significantly improved the knowledge base available for tōtara to assist landowners, forest managers, iwi and the government to grow and manage tōtara for both commercial and non-commercial purposes.
Type of thesis
Wade, A. V. (2020). Predicting site productivity drivers for Podocarpus totara and modelling its growth with 3-PG (Thesis, Master of Science (Research) (MSc(Research))). The University of Waikato, Hamilton, New Zealand. Retrieved from https://hdl.handle.net/10289/13732
The University of Waikato
All items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.