Comparison of Non-Destructive Tools for Measuring MOE of Southern Pine Trees

Comparison of Non-Destructive Tools for Measuring MOE of Southern Pine Trees | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Comparison of Non-Destructive Tools for Measuring MOE of Southern Pine Trees Chandan Kumar, Henri Bailleres, Vilius Gendvilas, Ian Last, Dominic Kain, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7050599/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Various non-destructive technologies have been employed to assess log quality, facilitating informed decision regarding sorting, segregation and processing decisions. However, there is a lack of comprehensive comparative evaluations of non-destructive tools for MOE assessment in standing trees, specifically regarding their accuracy, efficiency, practical advantages, and their correlation with the quality of sawn boards. Therefore, this study first investigated the correlations between log MOE, log density, and average MOE and MOR of boards obtained from each log. Then, the study evaluated the predictive performance of estimating measured log MOE (Measured_MOE) and average board MOE using a destructive (HM200_MOE), a lab-based non-destructive (USMOE), and two field-deployable non-destructive (ST300_MOE and Resi_MOE) log MOE measurement tools. Finally, a comparison between the destructive and non-destructive tools for log stiffness measurement, focusing on deployability, efficiency, and prediction power was presented. The results showed that as tools become less destructive, their predictive power in estimating log MOE decreases. The destructive method, HM200_MOE, was the most precise (with an 0.78), in estimating log MOE due to its direct measurement on felled logs, followed by the non-destructive m \(\:{R}^{2}=\:\) ethods: USMOE ( \(\:{R}^{2}=0.67\) ), ST300_MOE ( \(\:{R}^{2}=0.61\) ), and Resi_MOE ( \(\:{R}^{2}=0.57\) ). USMOE explained the highest variability in the average board MOE ( \(\:{R}^{2}=0.71\) ), followed by HM200_MOE ( \(\:{R}^{2}=0.69\) ), ST300_MOE ( \(\:{R}^{2}=0.64\) ), and Resi_MOE ( \(\:{R}^{2}=0.43\) ). Green log density showed very weak correlation with log MOE and average board MOE. In contrast, resin extracted log density had moderate to strong correlations with log MOE ( \(\:{R}^{2}\) from 0.39 to 0.63) and a high positive correlation with average board MOE ( \(\:{R}^{2}=\:0.84\) ). Destructive methods like HM200 are very precise but costly and unsuitable for field applications. Lab-based non-destructive tools such as USMOE can achieve a balance between precision and accuracy and potential field applicability but are slower, more expensive, and not field-deployable like Resi_MOE and ST300_MOE. Field-deployable tools, such as ST300 and Resi, offer practical solutions for operational forestry due to their efficiency and portability. However, their reduced predictive performance underscores the need for improved precision and accuracy through improved modelling or calibration. These findings highlight the trade-offs between predictive performance and operational efficiency, with the choice of tools depending on the required predictive performance, field deployability, sampling needs, and the cost of equipment and measurements. Figures Figure 1 Figure 2 Figure 3 Introduction Assessing the quality of logs in forest plantations is essential for optimising resource allocation, utilisation, and processing efficiency. Several methods are available for evaluating log quality, which can be broadly categorised into two main approaches based on whether tree felling is required: destructive and non-destructive methods. Tools like Hitman HM200 (Fibre-gen, Christchurch, New Zealand) and Beam Identification by Non-destructive Grading (BING) (CIRAD, France), that use resonance acoustic methods on log to measure dynamic MOE from felled trees, provide high predictive performance. Non-destructive evaluation (NDE) methods, in contrast, generally do not require tree felling and vary in their degree of invasiveness. Over the past few decades, significant advancements have been made in the NDE methods and tools for measuring the Modulus of Elasticity (MOE) in logs. These non-destructive methods often involve no or minimal sampling from standing trees without causing significant damage to the tree (Ondrejka, Gergeľ, Bucha, & Pástor, 2021 ; L. Schimleck et al., 2019 ). These non-destructive methods can be further categorised into two methods: Lab-based non-destructive methods: These methods require small samples taken from standing trees, with measurements conducted in laboratory settings. Examples include measurement and prediction from core samples using ultrasound (Kumar et al., 2021 ), SilviScan (Evans, Ilic, & Matheson, 2000 ), and DiscBot (Lee, 2024 ). Field-deployable non-destructive methods: These methods enable direct measurements on standing trees in situ and are portable for use in forest. These include resistance drilling (Downes & Lausberg, 2016 ; Downes et al., 2018 ; Rinn, Schweingruber, & Schär, 1996 ), Pilodyn, ST300 (Fibre-gen, Christchurch, New Zealand), and Fakopp. In lab-based non-destructive methods, the increment borer is a fundamental tool for assessing the wood quality of forest resources. This tool is instrumental in extracting cores from living trees, allowing for analysing growth trends by examining tree ring patterns. Once these cores have been obtained, they must be appropriately stored and transported to a laboratory for further analysis, where wood density and stiffness can be determined (Gao et al., 2017 ). Increment cores extracted from standing trees can also facilitate measurement of many other wood properties, including basic density, dendrochronological analysis, fibre length, and microfibril angle using techniques such as ultrasound, X-ray imaging, and near-infrared (NIR) spectroscopy (Downes et al., 1997 ; Gao et al., 2017 ; Gendvilas et al., 2022 ; Kumar et al., 2021 ). Valid tree density and MOE estimates are typically derived from cores spanning bark to pith, as they capture the full radial variation. This radial variation can be used to predict MOE of logs and the boards that can be obtained from the logs (Kumar et al., 2021 ; Psaltis et al., 2021 ). However, this process can present challenges, such as reaching the pith in large trees and core extraction difficulties can arise in denser trees (Gao et al., 2017 ; Grissino-Mayer, 2003 ). Near-infrared (NIR) spectroscopy can be used to analyse samples obtained from increment cores. The analysis involves a calibration process through a chemometric approach using training samples (calibration set) to develop the mathematical relationship between the NIR spectra and the property of interest (Wessels, Malan, & Rypstra, 2011 ). The SilviScan®, developed by CSIRO team in 1992, is a specialised laboratory device for measuring wood characteristics parameters such as density and microfibril angle. Its main components include an optical cell scanner for various wood features, an X-ray densitometer for density profiles, and an X-ray diffractometer for microfibril angles and cellulose crystallization (Evans et al., 2000 ). The device provides MOE estimates based on wood properties, calibrated using the acoustic resonance technique. SilviScan® has demonstrated high accuracy in multiple studies (Buksnowitz, Müller, Evans, Teischinger, & Grabner, 2007 ; L. R. Schimleck, Evans, & Matheson, 2002 ). The DiscBot, developed by the New Zealand Forest Research Institute trading as Scion, uses scanning technologies to assess various wood properties crucial for quality and final product outcomes. It utilises automatic motion of wooden discs beneath sensors, including a high-quality colour image camcorder, wood fibre angle measurement, an infrared spectrograph, X-ray measurement for wood density, and an ultrasonic device to evaluate wood strength (Ondrejka et al., 2021 ). In contrast, field deployable non-destructive methods on standing trees allow for faster data collection and processing with minimal or no lab work. Among them, acoustic techniques are widely employed and commercially viable methods due to their affordability, speed, robustness, and practicality for field applications. These tools have played a crucial role in assessing standing trees before harvesting, facilitating effective resource management, planning, harvesting, and wood processing to maximise the resource's value extraction potential (Ondrejka et al., 2021 ; L. Schimleck et al., 2019 ). These methods measure stress wave speed propagating through the stem, typically generated by tapping one end with a light hammer. The Time of Flight (TOF) of the acoustic wave is precisely determined between the transmitter probe and receiver probe (Wang & Ross, 2002 ; Wessels et al., 2011 ). Subsequently, wave speed and a dynamic Modulus of Elasticity (MOE) are calculated from the TOF measurements and the wood density data. Some examples of these devices are Fibregen Director ST300 Tool, ArborSonic 3D Acoustic Tomograph, IML Impulse Hammer, and FAKOPP company tools (Microsecond Timer, Resonance Log Grader, ArborElectro Impedance Tomograph). Another field deployable tool, the Resi technology refers to a resistance drilling method, developed in the early 1990s, primarily used to detect rot in trees and poles (Rinn et al., 1996 ). It operates by driving a specialized 3 mm diameter drill (needle) through a tree at specified feed and rotation speeds while measuring the resistance to rotation (torque). This process generates a profile of resistance at 0.1 mm intervals, which is correlated with wood density and stiffness (Downes & Lausberg, 2016 ; Downes et al., 2018 ; Rinn et al., 1996 ). The resistance drilling is gaining popularity due to its affordability for field use, digital data collection capability, and relatively high-resolution data output (Gao et al., 2017 ). Pilodyn, using a spring-loaded sticker pin, offers a minimally invasive method where penetration depth inversely correlates with wood density, yet its accuracy and reliability for tree selection, especially in breeding programs, is questioned due to its limited evaluation of outermost wood rings, making it unrepresentative of stem mean density (Cown, 1978 ; Gao et al., 2017 ). Given this, use of the pilodyn is limited and not discussed further here. The field deployable methods involved specialised operators, and their deployment in the field can be disturbed by weather events such as rain, wind, snow, etc. The various destructive and non-destructive evaluation (NDE) methods mentioned above have been employed to assess log quality, yielding varying degrees of accuracy in their results. While the destructive methods can be regarded as having high predictive performance, they are slow, expensive, and not suitable for faster in-field assessment. Laboratory-based non-destructive methods often achieve higher accuracy but can be time-consuming and costly. Emerging technologies such as integrated sensors and industrial artificial intelligence systems will enhance both the efficiency and usability of these systems. In contrast, portable tools like resistance drilling and ST300 provide immediate, in-situ measurements, allowing for large-scale, economical assessments (Baillères et al., 2019 ). The forest industry requires tools that enable efficient plot evaluation with a relatively large number of trees, balancing sampling rates, cost-effectiveness, and the speed required for field implementation. However, a noticeable gap remains in understanding the comparative performance of these tools, particularly regarding their relative predictive power, sampling rate, comparing their advantages and disadvantages, and the relationship among the MOE measured using those tools. Additionally, limited research exists on how tree and log quality assessments using these tools correlate with the mechanical performance of sawn boards produced from these logs. The objectives of this study are threefold: First, to investigate the correlation between log MOE measured using various destructive and non-destructive tools, log density, and the MOE and MOR of sawn boards. Second, to assess the accuracy of the log MOE measurement tools in predicting log MOE and the average MOE of sawn boards derived from these logs. It is worth noting that this average does not align closely with the board mechanical grade distribution, which determines the value of the production, determined using classification methods based on the distribution of the board’s mechanical performance (Baillères et al., 2019 ). Finally, destructive and non-destructive tools should be compared with a focus on measurement accuracy and precision, efficiency, predictive capability, and operational practicality. Methodology Data were collected from a total of 68 sawlogs, each 3.9 m in length, extracted from a height of 2.4 m to 6.3 m above the stump, from trees harvested from plantation forests in Tuan, Beerburrum and Toolara in southeast Queensland (SEQ), Australia. All trees were from the locally developed hybrid between slash pine ( Pinus elliottii var. elliottii ) and Caribbean pine ( P. caribaea var. hondurensis ), known as the PEE × PCH hybrid, from a range of germplasm types and silvicultural regimes. The sample comprised 30 F 2 hybrid trees of open-pollinated seedling origin from Tuan, aged 28 years, with a final crop stocking density of 388 stems per hectare; 30 F 1 hybrid trees from Beerburrum, aged 19 years, including an F 1 hybrid full-sib family and two clonal varieties, C625 and C887, with stocking densities ranging from 200 to 1000 stems per hectare; and 8 F 1 hybrid trees from Toolara, aged 24 years, all of clonal variety C887, with stocking densities ranging from 1006 to 2660 stems per hectare. Detailed sampling methods and locations are described in Kumar et al. ( 2021 ) and Baillères et al. ( 2019 ). 3.1 Destructive log MOE measurement The reference log MOE, referred to hereafter as ‘ Measured_MOE’, was obtained for each log using Beam Identification by Non-destructive Grading (BING), a resonance acoustic method for estimating MOE (Paradis, Brancheriau, & Baillères, 2017 ). This value was then used as a benchmark for comparison with other log MOE measurements. Details of the method are described in other papers (Baillères, Hopewell, & Boughton, 2009 ; Brancheriau, 2014 ; Faydi, Brancheriau, Pot, & Collet, 2017 ; Kumar et al., 2021 ). Another destructive log MOE measurement, referred to hereafter as ‘HM200_MOE’ was measured in felled logs using resonance acoustics with the HM200 (Fibre-gen, Christchurch, New Zealand). The density of the green logs referred to as ‘Green_Log_Density’ was measured from dimension and weight of the logs and was used to estimate Measured_MOE and HM200_MOE. 3.2 Non-destructive log MOE measurement 3.2.1 Lab based non-destructive measurement Bark to bark cores of dimensions approximately 16 mm × 16 mm were extracted from logs at 2.4 m height and cores were cut into 20mm segments. The segments were then air dried and their extracted density was estimated using air density and resin content measured using wet chemistry and near infrared (NIR). More detail can be found in (Baillères et al., 2019 ). Then ultrasonic transmission velocity (UTV) was measured for each segment. From UTV and extracted density, MOE of each segment was calculated. A five-parameter logistic (5PL) function was used to describe the radial variation of MOE obtained from segment data (Kumar et al., 2021 ). Then the area weighted average of the segment’s ultrasonic MOE and extracted log density were used to calculate the MOE of logs (‘USMOE’) and extracted log density (Extr_Log_Den_Est) following the method described in Kumar et al. ( 2021 ). 3.2.2 Field based non-destructive methods ‘ST300_MOE’ was calculated using time of flight (ToF) acoustics method using the Director ST300 (Fibre-gen, Christchurch, New Zealand) on standing trees and using a constant green density of 1000 kg.m -3 . ‘Resi_MOE’ and ‘Resi_Density’ were measured using micro-drill torque resistance using the IML (Resi) PD-400 (IML System GmbH, Wiesloch, Germany). A bark to pith Resi trace was collected on same felled logs avoiding branches and defects at a forward speed of 200 cm/min and a rotation speed of 2500 RPM. Resi traces were processed using web-based software ( https://forestquality.shinyapps.io/FWPA-5/ ) (accessed on 10 November 2024) to obtain ‘Resi_MOE’ and ‘Resi_Density’. 3.3 Average board MOE, MOR and density After completing the log measurements, the logs were sawn using industry-prescribed sawing patterns to nominal structural framing dimensions (96 × 40 mm and 72 × 40 mm). The resulting boards were oven dried. The Modulus of Elasticity (MOE) and the Modulus of Rupture (MOR) of the boards were determined by four point static bending testing in accordance with (AS/NZS 4063.1, 2010 ). The average board MOE, MOR, and density obtained from each log, referred to as ‘Avg_Board_MOE’, ‘Avg_Board_MOR’, and ‘Avg_Board_Den’ respectively, were calculated by averaging the MOE, MOR, and density values of individual boards from each log. 3.4 Statistical analysis All analyses were prepared in R (R Core Team, 2022a ) using RMarkdown (Allaire et al., 2023 ) within the RStudio environment (R Core Team, 2022b ). The predictive performance of each tool was evaluated using linear regression by comparing predicted values to measured values. Precision was defined as the degree of statistical variability, quantified using the coefficient of determination (R²), which indicates how consistently a tool predicts values relative to the measured data. Accuracy, on the other hand, was assessed as systematic bias, represented by the slope and intercept of the linear regression. This approach follows established metrological definitions, where precision refers to the closeness of repeated measurements to each other, and accuracy refers to the closeness of the predictions to the true (measured) values. Results and discussion First the coefficients of determination ( \(\:{R}^{2}\) ) matrix between log MOE, log density and average MOE, MOR and density of boards are presented and discussed. Then the results from regression analysis to investigate the precision and bias of predicting log MOE and average board MOE from the destructive and non-destructive tools are presented. \(\:{R}^{2}\) serves as an estimate of precision, indicating the proportion of variance explained by independent variable, while the slope and intercept of linear regression line reflect bias, indicating whether the model systematically underpredicts or overpredicts (accuracy or bias). Finally, the overall comparison between the tools regarding their predictive performance, operational capability and efficiency is presented. 4.1 Correlations between all MOE and density measurement Figure 1 shows the coefficients of determination ( \(\:{R}^{2}\) ) between log MOE measured by various destructive and non-destructive tools, density measurement (green log density and estimated extracted log density), average board MOE and MOR. Measured_MOE exhibited moderate to strong positive correlations ( \(\:{R}^{2}\) ) with other MOE measurement tools, ranging from \(\:{R}^{2}=0.57\) to \(\:{R}^{2}=0.78\) . The highest correlation with Measured_MOE was observed with HM200_MOE ( \(\:{R}^{2}=\) 0.78), likely due to both being resonance-based methods applied to felled logs. This was followed by USMOE ( \(\:{R}^{2}=0.67\) ), ST300_MOE ( \(\:{R}^{2}=0.61\) ) and Resi_MOE ( \(\:{R}^{2}=0.57\) ). However, these coefficients of determination ( \(\:{R}^{2}\) ) alone do not determine which tools are superior, as multiple additional factors need to be considered for measuring or predicting log MOE. These factors include measurement time, accuracy or bias as defined above, sampling speed, and whether the tools are field-deployable for use in commercial settings (discussed further in section 4.4). The average MOE of the boards (Avg_Board_MOE), representing an average performance of the final product (sawn boards), exhibited a strong correlation with Measured_MOE ( \(\:{R}^{2}=\:0.84\) ), USMOE ( \(\:{R}^{2}=\:0.71\) ), and HM200_MOE ( \(\:{R}^{2}=\:0.69\) ), indicating these methods can provide reliable predictions of average board quality. ST300_MOE also showed a moderate correlation ( \(\:{R}^{2}=\:0.64\) ), while Resi_MOE demonstrated a weaker relationship ( \(\:{R}^{2}=\:0.43\) ) with the Avg_Board_MOE. However, this is the average MOE of boards extracted from each log, which does not closely represent the grades recovery. The relatively lower predictability of Resi_MOE may be attributed to its dependence on calibration models and its measurement principle (torque), primarily linked to the wood density, which doesn’t provide a direct account of the microfibril angle (MFA) contribution to MOE and is influenced by other physical variables such as resin or water content. Further analysis could explore the underlying causes of variability in prediction accuracy among these tools, particularly for field-deployable methods like ST300 and Resi. These results suggest that as the degree of destructiveness of the tools increases, the proportion of variance explained in predicting log MOE and average board MOE also increases. The green log density displayed low correlation with all the MOE and density measurements. This result is likely influenced by variations in moisture content, which can fluctuate due to partial drying during transportation, handling, and storage, as well as differences in resin content within the logs. Conversely, extracted log density (Extr_Log_Den_Est) showed strong to moderate coefficients of determination with all log MOE values ( \(\:{R}^{2}\) from 0.39 to 0.63). Extr_Log_Den_Est also exhibited a moderate positive correlation with average board MOE ( \(\:{R}^{2}=\:0.59\) ) and average board density ( \(\:{R}^{2}=\:0.60\) ). This suggests that extracted log density serves as a more consistent predictor of overall board quality than the green log density for southern pine. This indicates that extracted log density is a more reliable metric for predicting both board and log MOE in southern pine species specially where resin and extractive is present, making it a preferable choice over green log density.Resi_Den is as good as Measured_MOE at predicting Extr_Log_Den_Est. As expected, there was a strong correlation between the non-destructive measurements Resi_Density and Resi_MOE (R² = 0.81) as they are not statistically independent. The average board MOR shows a strong relationship with the average board MOE ( \(\:{R}^{2}=\:0.7\) ). However, average board MOR exhibited only weak correlations with log MOE measurements ( \(\:{R}^{2}\) values ranging from 0.19 to 0.47), with the strongest correlation observed with Measured_MOE. This suggests that while log MOE can serve as a reliable predictor for average board MOE, it is not an effective predictor for average board MOR, which can be impacted by knots, grain angle and other features such as compression failure. 4.2 Comparison of Log MOE via Linear Regression Figure 2 shows the regression between Measured_MOE and other log MOE measured using various destructive and non-destructive tools differentiated by germplasm. As described in previous sections, the destructive method (e.g., HM200_MOE) exhibited the highest precision with Measured_MOE ( \(\:{R}^{2}=\:0.78\) ), followed by the lab-based non-destructive method (USMOE, \(\:{R}^{2}=\:0.67\) ), and then the field-deployable non-destructive methods (Resi_MOE, \(\:{R}^{2}=\:0.57\) , and ST300_MOE, \(\:{R}^{2}=\:0.61\) ). When comparing the regression slopes (Fig. 2 ), the USMOE demonstrated the lowest bias (slope = 0.95, close to 1), indicating a strong predictive power followed by HM200_MOE (a slope of 0.77). ST300_MOE exhibited the lowest accuracy or higher bias (slope = 0.47), significantly overpredicting the actual MOE values by approximately 50%, most likely due to sampling of the high-MOE outerwood only. Conversely, Resi-MOE with a slope of 1.22 underpredicted the log MOE by 22%, potentially due to calibration dependencies. The Resi-MOE and the HM200_MOE are biased by about the same degree but in the opposite direction. These results indicate that the destructive and lab-based non-destructive technique outperformed field-deployable tools in predicting log MOE, having both higher precision and lower bias. However, these results are for limited set of data only for 68 trees. One of the four sampled genotypes, clonal variety C625 (orange circles in Fig. 2 ), exhibited consistently low log MOE values, below 8,000 MPa. This resulted in reduced prediction accuracy and higher deviations from the regression lines, particularly for USMOE and Resi_MOE. The poor prediction performance for this germplasm may be attributed to its unique characteristics, such as growth characteristics (e.g. large juvenile wood proportion). Downes GM, Hogg, and Lee (2017) previously reported similar issues, attributing this to the abnormal behaviour of this specific clone. However, there were only 10 data points from C625. A larger dataset with more trees would provide better insight into these effects and overall comparison of the log MOE predictions. 4.3 Comparison of average board MOE via Linear Regression Figure 3 shows linear regression models predicting average board MOE (Avg_Board_MOE) from destructive (HM200_MOE), lab-based non-destructive (USMOE), and field-deployable non-destructive methods (ST300_MOE and Resi_MOE). Among these, USMOE exhibited the highest precision ( \(\:{R}^{2}=\:0.71\) ), followed closely by HM200_MOE ( \(\:{R}^{2}=\:0.69\) ). A regression analysis was conducted between BING log MOE and average board MOE for comparison, yielding an R² value of 0.84 and a slope of 0.94. For the field deployable non-destructive testing method measured on standing trees, ST300_MOE, showed a stronger relationship (R² = 0.64) compared to Resi_MOE (R² = 0.44). The high coefficients of determination for USMOE and HM200 indicate their strong ability to predict Avg_Board_MOE. USMOE demonstrated the least bias, with a slope of 0.97, which is close to 1, indicating that its predictions closely align with actual measured values (better accuracy). This makes USMOE a reliable non-destructive alternative for predicting average board stiffness. Resi_MOE also showed minimal bias (slope = 1.04). In contrast, HM200_MOE, a destructive method, exhibited moderate bias (slope = 0.71), suggesting a tendency to overestimate MOE, particularly for higher values. The highest bias was observed in ST300_MOE (slope = 0.47), indicating that it significantly overestimates actual average board MOE. Similar to the log MOE predictions, one of the four sampled genotypes, clonal variety C625 (indicated by orange circles), exhibited the greatest deviation from the regression line, particularly for non-destructive measurements based on USMOE and Resi. Destructive methods like HM200_MOE remain highly accurate but challenging to implement in field applications. Lab-based USMOE bridges the gap between accuracy and non-destructiveness, though its dependency on laboratory facilities limits its field portability. The results must be interpreted in consideration that this study sampled multiple plantings of varying genetic material and planting stocking. This may have impacted upon the relative usefulness of the different methods. For example, outerwood as sampled by the ST300 may be influenced to a greater extent by competition effects. These tools would typically (but not always) be used in more homogeneous stands. Field-deployable tools like ST300_MOE and Resi_MOE offer practical solutions for operational forestry, but the lower predictive accuracy of Resi highlights the need for improved metrology, calibration techniques and predictive models for MOE. These findings underscore the trade-offs between accuracy and operational efficiency when selecting tools for predicting Avg_Board_MOE. 4.4 Overall comparison Table 1 provides a comprehensive comparison of destructive and non-destructive tools based on their measurement efficiency, operational capabilities, and predictive power (precision and bias or accuracy). As highlighted earlier, the destructive methods (BING_MOE and HM200_MOE) require tree felling and are not field-deployable, limiting their use in commercial forestry operations despite their high accuracy. Table 1 Comparison of Destructive and Non-Destructive Tools for Log Stiffness Measurement for clearfall age Southern Pine trees Point of Comparison BING (Destructive) HM200 (Destructive) USMOE (Non-destructive) Resi (Non-destructive) ST300 (Non-destructive) Tree Felling Required Yes, felling required Yes, felling required No, measures standing trees No, measures standing trees No, measures standing trees Field Deployable Not field-deployable Not field-deployable Field-deployable but lab work needed Field-deployable, no lab needed Field-deployable, no lab needed Type of MOE Measurement Direct measurement, not calibration-based Direct measurement, not calibration-based Direct measurement, not calibration-based Calibration-based (requires training samples) Calibration-based (requires approximate density) Sampling Rate (contiguous trees per hour) Slow: approximately 5 trees/hour Slow: approximately 5 trees/hour Slow: approximately 4 trees/hour Fast: approximately 50 trees/hour Moderate: approximately 20 trees/hour Data Analysis Speed Fast Fast Moderate Fast Fast Radial Variation Measurement No No Yes Yes No Full Radial Analysis (e.g., Age, Density, Extractives) No No Yes, full analysis possible (e.g., cores used for extractives, age, MFA, density) Yes, partial analysis possible (e.g. density variation) No Precision of Log MOE Prediction High High High Moderate Moderate Bias of Log MOE Prediction Moderate Moderate Low Moderate High Precision of Individual Board MOE Not possible Not possible Possible Possible Not possible Precision of Average Board MOE Prediction High Moderate to High Moderate to High Low to Moderate* Moderate Bias of Average Board MOE Prediction Low Moderate Low Low High * Resi accuracy may be improved through calibration and improved predictive models for MOE In terms of bias for log MOE prediction, USMOE (0.95) demonstrated the lowest bias, followed by HM200_MOE (0.77). Resi_MOE (1.22) exhibited a similar level of bias to HM200 but in the opposite direction, while ST300_MOE (0.47) had the highest bias, significantly overestimating log MOE. For predicting board MOE, USMOE (0.97) and Resi_MOE (1.04) exhibited the lowest bias, whereas ST300_MOE (0.47) had the highest bias, significantly overestimating board MOE. HM200_MOE (0.71) showed moderate bias, slightly underpredicting board MOE. As previously noted, ST300 measures the outer wood, and the bias can be corrected provided the physical meaning of stress wave speed is adequately understood and accurately modelled. Additionally, the ST300 is becoming increasingly difficult to service, and its availability from suppliers is gradually declining. However, Resi and ST300 excel in their operational efficiency, with faster sampling rates and no requirement for laboratory work, making them highly suitable for field applications where rapid assessment is prioritised. The Resi tools rely on statistical models which rely on training samples. This approach can introduce uncertainty in the predictive power. Therefore, the type of calibration techniques and the number and quality of training samples are crucial. USMOE, another non-destructive method, requires laboratory measurements of density and ultrasound velocity, making it less practical for large-scale field operations. However, USMOE can provide more accurate estimation (low bias and high precision) of log MOE and average board MOE. However, with ongoing technological advancements, USMOE could become more efficient if density and ultrasonic transmission velocity are measured faster or if alternative methods—such as segmental MOE measurement using NIR—are employed. Improvements in coring technology, including increased speed, reduced effort, greater portability, and innovative drilling techniques, would also make this approach better suited for field measurements. Among the tools compared, only USMOE allow for the measurement of radial MOE variation. As a direct measurement method, USMOE offers additional capabilities such as the prediction of individual board MOE and enabling detailed analyses such as within-tree variation and growth trends. This makes it particularly valuable for research applications requiring a deeper understanding of wood properties. Meanwhile, Resi calibration could be improved in the future to provide radial MOE profiles for suitable species, potentially expanding its analytical applications. Additionally, the cores obtained from standing trees can also be used for other technical traits or research purposes, such as dendrochronology-related investigations; other useful measurements, such as resin content, chemical composition analysis, and predictions of quality of products derived from those trees, can be made using the cores if necessary. For example, Psaltis et al. ( 2021 ) used the segment’s MOE measured by ultrasound to predict the individual board MOE that can be obtained from a forest. In this study, the USMOE, using a 2D integration approach, can reconstruct an estimate of MOE for individual boards that can be sawn from trees with moderate accuracy (R 2 = 0.53) and low bias (2%). Overall, the choice of tool depends on the specific application context: Destructive tools (BING and HM200) are ideal for research and high-precision applications but are impractical for operational forestry due to their lack of field-deployability. USMOE bridges the gap between accuracy and potential field applicability, with future advancements likely to enhance its practicality. Resi and ST300 are better suited for large-scale commercial operations, offering efficiency and portability despite their relatively lower accuracy. Further research should focus on improving the calibration models for accuracy of field-deployable tools like Resi and ST300 and on advancing USMOE toward a fully field-compatible solution. These developments could significantly enhance the efficiency and accuracy of non-destructive log quality assessments. Conclusion This study compared and evaluated the performance of various destructive and non-destructive tools for log MOE and board MOE estimation of plantations grown Southern Pine. The study included log MOE measurement using two destructive tools, BING (Measured_MOE) and HM200, along with three non-destructive tools: one lab-based (USMOE) and two field-deployable (ST300 and Resi). The accuracy of the tools increases with the degree of destructiveness. The destructive tool (HM200_MOE) was the most precise in predicting log MOE (( \(\:{R}^{2}=0.78\) ) followed by USMOE ( \(\:{R}^{2}=0.67\) ), ST300_MOE ( \(\:{R}^{2}=0.61\) ) and Resi_MOE ( \(\:{R}^{2}=0.57\) ). The prediction of board MOE followed the same trend with higher precision and less bias using USMOE than HM200_MOE. Additionally, Germplasm C625 consistently exhibited low log MOE values (< 8,000 MPa), leading to reduced prediction accuracy, particularly for USMOE and Resi_MOE. These deviations align with prior reports attributing this behaviour to unique characteristics of this specific clone. While extracted log density exhibited strong correlations with both log MOE and average board MOE for this species of tree, green log density showed a very weak correlation. The choice of tool depends on the specific application requirements. For high accuracy in research or precision-demanding operations, destructive methods remain the most reliable. For non-destructive tools like USMOE, Resi and ST300 offer a practical balance between predictive power and operational efficiency. Future research should focus on improving the calibration models for non-destructive tools to enhance their accuracy and consistency across diverse field conditions. Declarations Author Contribution C.K. drafted the manuscript, conducted data analysis, and prepared figures and tables.V.G. contributed to the manuscript review and editing and the conceptual development of the introduction.I.L. and D.K. reviewed and edited the manuscript and contributed to site selection, data collection, and tree harvesting.G.D. reviewed the manuscript and validated the resistograph data.D.L. contributed to experimental planning, secured funding, data collection, and provided critical review and manuscript editing.H.B. contributed to the experimental method, conceptual development, reviewed and edited the manuscript Acknowledgement This study combines information from two projects: “Improving returns from Southern Pine plantations through innovative resource characterisation” Project PNC361-1415 and “Assessing and managing mid-rotation wood quality in Australian softwood plantations to produce fit-for-purpose logs”. Project PNB548-2021”. The authors gratefully acknowledge the financial support for the project funded by Forest and Wood Products Australia, DAF Forest Industries, the University of the Sunshine Coast and industry partners: HQPlantations, Forest Corporation New South Wales, Hancock Victoria Plantations and Hyne Timber Pty Ltd. We acknowledge the contribution by all staff at DAF Gympie and Salisbury Research Facility including Henri Bailleres, Gary Hopewell, Rhianna Robinson, Rica Minett, Chris Fitzgerald, Marco Lausberg, Tony Burridge, and John Oostenbrink. References Allaire, J., Xie, Y., Dervieux, C., McPherson, J., Luraschi, J., Ushey, K., & Atkins, A. (2023). Rmarkdown: Dynamic Documents for r. https://CRAN.R-project.org/package=rmarkdown. AS/NZS 4063.1. (2010). Characterization of structural timber, Part 1: Test methods. Standards Australia / Standards New Zealand . Baillères, H., Hopewell, G. P., & Boughton, G. (2009). MOE and MOR assessment technologies for improving graded recovery of exotic pines in Australia. Forest & Wood Products Australia,, Project no: PNB040-0708 Baillères, H., Lee, D., Kumar, C., Psaltis, S., Hopewell, G. P., & Brancheriau, L. (2019). Improving returns from southern pine plantations through innovative resource characterisation. Forest & Wood Products Australia,, Project no: PNC361-1415 (ISBN 978-1-925213-89-8). Brancheriau, L. (2014). An alternative solution for the determination of elastic parameters in free–free flexural vibration of a Timoshenko beam. Wood Science and Technology, 48 (6), 1269-1279. doi:10.1007/s00226-014-0672-x Buksnowitz, C., Müller, U., Evans, R., Teischinger, A., & Grabner, M. (2007). The potential of SilviScan’s X-ray diffractometry method for the rapid assessment of spiral grain in softwood, evaluated by goniometric measurements. Wood Science and Technology, 42 , 95-102. doi:10.1007/s00226-007-0153-6 Cown, D. (1978). Comparison of the Pilodyn and torsiometer methods for the rapid assessment of wood density in living trees. New Zealand Journal of Forestry Science, 8 (3), 384-391. Downes, G., Hudson, I., Raymond, C., Dean, G., Michell, A., Schimleck, L., . . . Muneri, A. (1997). Sampling plantation Eucalypts for wood and fibre properties. . Retrieved from Melbourne: Downes, G., & Lausberg, M. (2016). Evaluation of the RESI software tool for the prediction of HM200 within pine logs sourced from multiple sites across New Zealand and Australia. NZ Solid Wood Innov, 15 . Downes, G., Lausberg, M., Potts, B., Pilbeam, D., Bird, M., & Bradshaw, B. (2018). Application of the IML Resistograph to the infield assessment of basic density in plantation eucalypts. Australian Forestry, 81 (3), 177-185. Downes GM, Hogg, B., & Lee, D. (2017). Evaluating the application of the IML Resistograph to the prediction of key wood properties of the Southern Yellow Pine. University of the Sunshine Coast., In Milestone report 5a-d FWPA PNC361-1415 . Evans, R., Ilic, J., & Matheson, C. (2000). Rapid estimation of solid wood stiffness using SilviScan. Paper presented at the Proceedings of 26th Forest Products Research Conference: Research developments and industrial applications and Wood Waste Forum, Clayton, Victoria, Australia, 19-21 June 2000. Faydi, Y., Brancheriau, L., Pot, G., & Collet, R. (2017). Prediction of Oak Wood Mechanical Properties Based on the Statistical Exploitation of Vibrational Response. BioResources, 12 (3), 5913-5927. Gao, S., Wang, X., Wiemann, M. C., Brashaw, B. K., Ross, R. J., & Wang, L. (2017). A critical analysis of methods for rapid and nondestructive determination of wood density in standing trees. Annals of Forest Science, 74 (2), 27. doi:10.1007/s13595-017-0623-4 Gendvilas, V., Neyland, M., Rocha-Sepúlveda, M. F., Downes, G. M., Hunt, M., Jacobs, A., . . . O’Reilly-Wapstra, J. (2022). Effects of thinning on the longitudinal and radial variation in wood properties of Eucalyptus nitens. Forestry: An International Journal of Forest Research, 95 (4), 504-517. doi:10.1093/forestry/cpac007 Grissino-Mayer, H. D. (2003). A Manual and Tutorial for the Proper Use of an Increment Borer. Tree-ring research, 59 , 63-79. Kumar, C., Psaltis, S., Bailleres, H., Turner, I., Brancheriau, L., Hopewell, G., . . . Lee, D. J. (2021). Accurate estimation of log MOE from non-destructive standing tree measurements. Annals of Forest Science, 78 (1). doi:10.1007/s13595-021-01031-w Lee, J. (2024). Meet DiscBot, our new wood quality detective. Scion . Retrieved from https://www.scionresearch.com/about-us/about-scion/corporate-publications/scion-connections/past-issues-list/issue-17,-september-2015/meet-discbot,-our-new-quality-detective Ondrejka, V., Gergeľ, T., Bucha, T., & Pástor, M. (2021). Innovative methods of non-destructive evaluation of log quality. Central European Forestry Journal, 67 (1), 3-13. Paradis, S., Brancheriau, L., & Baillères, H. (2017). Bing: Beam Identification by Non destructive Grading . Psaltis, S., Kumar, C., Turner, I., Carr, E. J., Farrell, T., Brancheriau, L., . . . Lee, D. J. (2021). A new approach for predicting board MOE from increment cores. Annals of Forest Science, 78 (3). doi:10.1007/s13595-021-01093-w R Core Team. (2022a). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. R Core Team. (2022b). RStudio: Integrated development for r. Computer Program, RStudio, PBC, Boston, MA. http://www.rstudio.com/. Rinn, F., Schweingruber, F., & Schär, E. (1996). RESISTOGRAPH and X-Ray Density Charts of Wood. Comparative Evaluation of Drill Resistance Profiles and X-ray Density Charts of Different Wood Species. Holzforschung, 50 , 303-311. doi:10.1515/hfsg.1996.50.4.303 Schimleck, L., Dahlen, J., Apiolaza, L. A., Downes, G., Emms, G., Evans, R., . . . Wang, X. (2019). Non-destructive evaluation techniques and what they tell us about wood property variation. Forests, 10 (9), 728. Schimleck, L. R., Evans, R., & Matheson, A. C. (2002). Estimation ofPinus radiata D. Don clear wood properties by near-infrared spectroscopy. Journal of Wood Science, 48 (2), 132-137. doi:10.1007/BF00767290 Wang, X., & Ross, R. J. (2002). Nondestructive evaluation of green materials–recent research and development activities. Nondestructive evaluation of wood. Forest Products Society, Madison . Wessels, C. B., Malan, F. S., & Rypstra, T. (2011). A review of measurement methods used on standing trees for the prediction of some mechanical properties of timber. European Journal of Forest Research, 130 (6), 881-893. doi:10.1007/s10342-011-0484-6 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Nov, 2025 Reviews received at journal 17 Nov, 2025 Reviewers agreed at journal 17 Nov, 2025 Reviews received at journal 01 Sep, 2025 Reviews received at journal 20 Aug, 2025 Reviewers agreed at journal 06 Aug, 2025 Reviewers agreed at journal 05 Aug, 2025 Reviewers invited by journal 05 Aug, 2025 Editor assigned by journal 28 Jul, 2025 Submission checks completed at journal 09 Jul, 2025 First submitted to journal 05 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7050599","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":496984257,"identity":"5c9c2cc9-d95b-4bf3-8ccc-2653d6388ac1","order_by":0,"name":"Chandan Kumar","email":"data:image/png;base64,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","orcid":"","institution":"Department of Primary Industries","correspondingAuthor":true,"prefix":"","firstName":"Chandan","middleName":"","lastName":"Kumar","suffix":""},{"id":496984260,"identity":"f49301c9-b971-4e8f-b0fa-7296443499d1","order_by":1,"name":"Henri Bailleres","email":"","orcid":"","institution":"Scion","correspondingAuthor":false,"prefix":"","firstName":"Henri","middleName":"","lastName":"Bailleres","suffix":""},{"id":496984261,"identity":"9516de25-e23f-42cc-95b8-c223ee7144ef","order_by":2,"name":"Vilius Gendvilas","email":"","orcid":"","institution":"University of the Sunshine Coast","correspondingAuthor":false,"prefix":"","firstName":"Vilius","middleName":"","lastName":"Gendvilas","suffix":""},{"id":496984263,"identity":"6e2bfbdb-9233-41b3-be18-5c12c12f8ef2","order_by":3,"name":"Ian Last","email":"","orcid":"","institution":"HQPlantations Pty Ltd","correspondingAuthor":false,"prefix":"","firstName":"Ian","middleName":"","lastName":"Last","suffix":""},{"id":496984265,"identity":"f2876403-15b6-4a07-be3b-76f1def4c6b6","order_by":4,"name":"Dominic Kain","email":"","orcid":"","institution":"HQPlantations Pty Ltd","correspondingAuthor":false,"prefix":"","firstName":"Dominic","middleName":"","lastName":"Kain","suffix":""},{"id":496984267,"identity":"a13c1e5e-95d2-49bb-ab5d-d768c4692347","order_by":5,"name":"Geoff Downes","email":"","orcid":"","institution":"Forest Quality","correspondingAuthor":false,"prefix":"","firstName":"Geoff","middleName":"","lastName":"Downes","suffix":""},{"id":496984269,"identity":"34063c4e-53f6-4d59-8318-4c46e790a58b","order_by":6,"name":"David J. Lee","email":"","orcid":"","institution":"University of the Sunshine Coast","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"J.","lastName":"Lee","suffix":""}],"badges":[],"createdAt":"2025-07-05 05:23:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7050599/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7050599/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88614792,"identity":"496654d6-e7d8-460a-a82b-3ba39544db1b","added_by":"auto","created_at":"2025-08-08 10:29:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":16991,"visible":true,"origin":"","legend":"\u003cp\u003eMatrix of coefficients of determination (R\u003csup\u003e2\u003c/sup\u003e) from squared Pearson correlations between log MOE, density (measured using different tools), average board MOE, MOR, and density. Statistically insignificant correlations (p\u0026gt;0.05) are left blank. Positive correlations are represented in red and negative correlations in blue, with the intensity and circle size proportional to the strength of the correlation.\u003c/p\u003e","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7050599/v1/b68f83b0e1aa6489d5515256.png"},{"id":88613963,"identity":"37b78644-be54-42ec-b8f3-728436bd5148","added_by":"auto","created_at":"2025-08-08 10:21:52","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":19477,"visible":true,"origin":"","legend":"\u003cp\u003eThe linear regression between Measured_MOE and MOE measured using various tools (e.g., USMOE, HM200, ST300, and Resi), differentiated by Germplasm. Regression analysis includes red lines and text indicating regression through the origin (zero intercept) and blue lines representing ordinary regression with an intercept. Marker shapes and colours represent different Germplasm groups, with a single shared legend for all plots. The red dashed line represents the 1:1 reference line, indicating perfect agreement between measured and predicted values.\u003c/p\u003e","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7050599/v1/d8da42a437896ce9ae365b5a.png"},{"id":88613967,"identity":"1f59c2d3-3f4f-42e2-9825-fa6ee31087a0","added_by":"auto","created_at":"2025-08-08 10:21:52","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":21432,"visible":true,"origin":"","legend":"\u003cp\u003eThe linear regression between averaged board MOE and measured log MOE using various tools (e.g. USMOE, HM200, ST300, and Resi). The red dashed line represents the 1:1 reference line, indicating perfect agreement between measured and predicted values.\u003c/p\u003e","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7050599/v1/5f5f7c42cfc6cdd733a22993.png"},{"id":88615072,"identity":"5fcb723e-c407-439e-b22b-a03fa55ac99e","added_by":"auto","created_at":"2025-08-08 10:37:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":968774,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7050599/v1/799ca064-f0a0-4883-8666-1facdd8d908f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Comparison of Non-Destructive Tools for Measuring MOE of Southern Pine Trees","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAssessing the quality of logs in forest plantations is essential for optimising resource allocation, utilisation, and processing efficiency. Several methods are available for evaluating log quality, which can be broadly categorised into two main approaches based on whether tree felling is required: destructive and non-destructive methods. Tools like Hitman HM200 (Fibre-gen, Christchurch, New Zealand) and Beam Identification by Non-destructive Grading (BING) (CIRAD, France), that use resonance acoustic methods on log to measure dynamic MOE from felled trees, provide high predictive performance.\u003c/p\u003e\u003cp\u003eNon-destructive evaluation (NDE) methods, in contrast, generally do not require tree felling and vary in their degree of invasiveness. Over the past few decades, significant advancements have been made in the NDE methods and tools for measuring the Modulus of Elasticity (MOE) in logs. These non-destructive methods often involve no or minimal sampling from standing trees without causing significant damage to the tree (Ondrejka, Gergeľ, Bucha, \u0026amp; P\u0026aacute;stor, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; L. Schimleck et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThese non-destructive methods can be further categorised into two methods:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eLab-based non-destructive methods: These methods require small samples taken from standing trees, with measurements conducted in laboratory settings. Examples include measurement and prediction from core samples using ultrasound (Kumar et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), SilviScan (Evans, Ilic, \u0026amp; Matheson, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e), and DiscBot (Lee, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eField-deployable non-destructive methods: These methods enable direct measurements on standing trees in situ and are portable for use in forest. These include resistance drilling (Downes \u0026amp; Lausberg, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Downes et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rinn, Schweingruber, \u0026amp; Sch\u0026auml;r, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1996\u003c/span\u003e), Pilodyn, ST300 (Fibre-gen, Christchurch, New Zealand), and Fakopp.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eIn lab-based non-destructive methods, the increment borer is a fundamental tool for assessing the wood quality of forest resources. This tool is instrumental in extracting cores from living trees, allowing for analysing growth trends by examining tree ring patterns. Once these cores have been obtained, they must be appropriately stored and transported to a laboratory for further analysis, where wood density and stiffness can be determined (Gao et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Increment cores extracted from standing trees can also facilitate measurement of many other wood properties, including basic density, dendrochronological analysis, fibre length, and microfibril angle using techniques such as ultrasound, X-ray imaging, and near-infrared (NIR) spectroscopy (Downes et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Gao et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Gendvilas et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kumar et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Valid tree density and MOE estimates are typically derived from cores spanning bark to pith, as they capture the full radial variation. This radial variation can be used to predict MOE of logs and the boards that can be obtained from the logs (Kumar et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Psaltis et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). However, this process can present challenges, such as reaching the pith in large trees and core extraction difficulties can arise in denser trees (Gao et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Grissino-Mayer, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eNear-infrared (NIR) spectroscopy can be used to analyse samples obtained from increment cores. The analysis involves a calibration process through a chemometric approach using training samples (calibration set) to develop the mathematical relationship between the NIR spectra and the property of interest (Wessels, Malan, \u0026amp; Rypstra, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The SilviScan\u0026reg;, developed by CSIRO team in 1992, is a specialised laboratory device for measuring wood characteristics parameters such as density and microfibril angle. Its main components include an optical cell scanner for various wood features, an X-ray densitometer for density profiles, and an X-ray diffractometer for microfibril angles and cellulose crystallization (Evans et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). The device provides MOE estimates based on wood properties, calibrated using the acoustic resonance technique. SilviScan\u0026reg; has demonstrated high accuracy in multiple studies (Buksnowitz, M\u0026uuml;ller, Evans, Teischinger, \u0026amp; Grabner, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; L. R. Schimleck, Evans, \u0026amp; Matheson, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe DiscBot, developed by the New Zealand Forest Research Institute trading as Scion, uses scanning technologies to assess various wood properties crucial for quality and final product outcomes. It utilises automatic motion of wooden discs beneath sensors, including a high-quality colour image camcorder, wood fibre angle measurement, an infrared spectrograph, X-ray measurement for wood density, and an ultrasonic device to evaluate wood strength (Ondrejka et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn contrast, field deployable non-destructive methods on standing trees allow for faster data collection and processing with minimal or no lab work. Among them, acoustic techniques are widely employed and commercially viable methods due to their affordability, speed, robustness, and practicality for field applications. These tools have played a crucial role in assessing standing trees before harvesting, facilitating effective resource management, planning, harvesting, and wood processing to maximise the resource's value extraction potential (Ondrejka et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; L. Schimleck et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). These methods measure stress wave speed propagating through the stem, typically generated by tapping one end with a light hammer. The Time of Flight (TOF) of the acoustic wave is precisely determined between the transmitter probe and receiver probe (Wang \u0026amp; Ross, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Wessels et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Subsequently, wave speed and a dynamic Modulus of Elasticity (MOE) are calculated from the TOF measurements and the wood density data. Some examples of these devices are Fibregen Director ST300 Tool, ArborSonic 3D Acoustic Tomograph, IML Impulse Hammer, and FAKOPP company tools (Microsecond Timer, Resonance Log Grader, ArborElectro Impedance Tomograph).\u003c/p\u003e\u003cp\u003eAnother field deployable tool, the Resi technology refers to a resistance drilling method, developed in the early 1990s, primarily used to detect rot in trees and poles (Rinn et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). It operates by driving a specialized 3 mm diameter drill (needle) through a tree at specified feed and rotation speeds while measuring the resistance to rotation (torque). This process generates a profile of resistance at 0.1 mm intervals, which is correlated with wood density and stiffness (Downes \u0026amp; Lausberg, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Downes et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rinn et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). The resistance drilling is gaining popularity due to its affordability for field use, digital data collection capability, and relatively high-resolution data output (Gao et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\u003cp\u003ePilodyn, using a spring-loaded sticker pin, offers a minimally invasive method where penetration depth inversely correlates with wood density, yet its accuracy and reliability for tree selection, especially in breeding programs, is questioned due to its limited evaluation of outermost wood rings, making it unrepresentative of stem mean density (Cown, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1978\u003c/span\u003e; Gao et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Given this, use of the pilodyn is limited and not discussed further here. The field deployable methods involved specialised operators, and their deployment in the field can be disturbed by weather events such as rain, wind, snow, etc.\u003c/p\u003e\u003cp\u003eThe various destructive and non-destructive evaluation (NDE) methods mentioned above have been employed to assess log quality, yielding varying degrees of accuracy in their results. While the destructive methods can be regarded as having high predictive performance, they are slow, expensive, and not suitable for faster in-field assessment. Laboratory-based non-destructive methods often achieve higher accuracy but can be time-consuming and costly. Emerging technologies such as integrated sensors and industrial artificial intelligence systems will enhance both the efficiency and usability of these systems. In contrast, portable tools like resistance drilling and ST300 provide immediate, in-situ measurements, allowing for large-scale, economical assessments (Baill\u0026egrave;res et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe forest industry requires tools that enable efficient plot evaluation with a relatively large number of trees, balancing sampling rates, cost-effectiveness, and the speed required for field implementation. However, a noticeable gap remains in understanding the comparative performance of these tools, particularly regarding their relative predictive power, sampling rate, comparing their advantages and disadvantages, and the relationship among the MOE measured using those tools. Additionally, limited research exists on how tree and log quality assessments using these tools correlate with the mechanical performance of sawn boards produced from these logs. The objectives of this study are threefold: First, to investigate the correlation between log MOE measured using various destructive and non-destructive tools, log density, and the MOE and MOR of sawn boards. Second, to assess the accuracy of the log MOE measurement tools in predicting log MOE and the average MOE of sawn boards derived from these logs. It is worth noting that this average does not align closely with the board mechanical grade distribution, which determines the value of the production, determined using classification methods based on the distribution of the board\u0026rsquo;s mechanical performance (Baill\u0026egrave;res et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Finally, destructive and non-destructive tools should be compared with a focus on measurement accuracy and precision, efficiency, predictive capability, and operational practicality.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003eData were collected from a total of 68 sawlogs, each 3.9 m in length, extracted from a height of 2.4 m to 6.3 m above the stump, from trees harvested from plantation forests in Tuan, Beerburrum and Toolara in southeast Queensland (SEQ), Australia. All trees were from the locally developed hybrid between slash pine (\u003cem\u003ePinus elliottii\u003c/em\u003e var. \u003cem\u003eelliottii\u003c/em\u003e) and Caribbean pine (\u003cem\u003eP. caribaea\u003c/em\u003e var. \u003cem\u003ehondurensis\u003c/em\u003e), known as the PEE \u0026times; PCH hybrid, from a range of germplasm types and silvicultural regimes. The sample comprised 30 F\u003csub\u003e2\u003c/sub\u003e hybrid trees of open-pollinated seedling origin from Tuan, aged 28 years, with a final crop stocking density of 388 stems per hectare; 30 F\u003csub\u003e1\u003c/sub\u003e hybrid trees from Beerburrum, aged 19 years, including an F\u003csub\u003e1\u003c/sub\u003e hybrid full-sib family and two clonal varieties, C625 and C887, with stocking densities ranging from 200 to 1000 stems per hectare; and 8 F\u003csub\u003e1\u003c/sub\u003e hybrid trees from Toolara, aged 24 years, all of clonal variety C887, with stocking densities ranging from 1006 to 2660 stems per hectare. Detailed sampling methods and locations are described in Kumar et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and Baill\u0026egrave;res et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Destructive log MOE measurement\u003c/h2\u003e\u003cp\u003eThe reference log MOE, referred to hereafter as \u003cem\u003e\u0026lsquo;\u003c/em\u003eMeasured_MOE\u0026rsquo;, was obtained for each log using Beam Identification by Non-destructive Grading (BING), a resonance acoustic method for estimating MOE (Paradis, Brancheriau, \u0026amp; Baill\u0026egrave;res, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). This value was then used as a benchmark for comparison with other log MOE measurements. Details of the method are described in other papers (Baill\u0026egrave;res, Hopewell, \u0026amp; Boughton, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Brancheriau, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Faydi, Brancheriau, Pot, \u0026amp; Collet, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Kumar et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother destructive log MOE measurement, referred to hereafter as \u0026lsquo;HM200_MOE\u0026rsquo; was measured in felled logs using resonance acoustics with the HM200 (Fibre-gen, Christchurch, New Zealand). The density of the green logs referred to as \u0026lsquo;Green_Log_Density\u0026rsquo; was measured from dimension and weight of the logs and was used to estimate Measured_MOE and HM200_MOE.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Non-destructive log MOE measurement\u003c/h2\u003e\u003cdiv id=\"Sec5\" class=\"Section3\"\u003e\u003ch2\u003e3.2.1 Lab based non-destructive measurement\u003c/h2\u003e\u003cp\u003eBark to bark cores of dimensions approximately 16 mm \u0026times; 16 mm were extracted from logs at 2.4 m height and cores were cut into 20mm segments. The segments were then air dried and their extracted density was estimated using air density and resin content measured using wet chemistry and near infrared (NIR). More detail can be found in (Baill\u0026egrave;res et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Then ultrasonic transmission velocity (UTV) was measured for each segment. From UTV and extracted density, MOE of each segment was calculated. A five-parameter logistic (5PL) function was used to describe the radial variation of MOE obtained from segment data (Kumar et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Then the area weighted average of the segment\u0026rsquo;s ultrasonic MOE and extracted log density were used to calculate the MOE of logs (\u0026lsquo;USMOE\u0026rsquo;) and extracted log density (Extr_Log_Den_Est) following the method described in Kumar et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e3.2.2 Field based non-destructive methods\u003c/h2\u003e\u003cp\u003e\u0026lsquo;ST300_MOE\u0026rsquo; was calculated using time of flight (ToF) acoustics method using the Director ST300 (Fibre-gen, Christchurch, New Zealand) on standing trees and using a constant green density of 1000 kg.m\u003csup\u003e-3\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u0026lsquo;Resi_MOE\u0026rsquo; and \u0026lsquo;Resi_Density\u0026rsquo; were measured using micro-drill torque resistance using the IML (Resi) PD-400 (IML System GmbH, Wiesloch, Germany). A bark to pith Resi trace was collected on same felled logs avoiding branches and defects at a forward speed of 200 cm/min and a rotation speed of 2500 RPM. Resi traces were processed using web-based software (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://forestquality.shinyapps.io/FWPA-5/\u003c/span\u003e\u003cspan address=\"https://forestquality.shinyapps.io/FWPA-5/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (accessed on 10 November 2024) to obtain \u0026lsquo;Resi_MOE\u0026rsquo; and \u0026lsquo;Resi_Density\u0026rsquo;.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Average board MOE, MOR and density\u003c/h2\u003e\u003cp\u003eAfter completing the log measurements, the logs were sawn using industry-prescribed sawing patterns to nominal structural framing dimensions (96 \u0026times; 40 mm and 72 \u0026times; 40 mm). The resulting boards were oven dried. The Modulus of Elasticity (MOE) and the Modulus of Rupture (MOR) of the boards were determined by four point static bending testing in accordance with (AS/NZS 4063.1, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The average board MOE, MOR, and density obtained from each log, referred to as \u0026lsquo;Avg_Board_MOE\u0026rsquo;, \u0026lsquo;Avg_Board_MOR\u0026rsquo;, and \u0026lsquo;Avg_Board_Den\u0026rsquo; respectively, were calculated by averaging the MOE, MOR, and density values of individual boards from each log.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Statistical analysis\u003c/h2\u003e\u003cp\u003eAll analyses were prepared in R (R Core Team, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022a\u003c/span\u003e) using RMarkdown (Allaire et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) within the RStudio environment (R Core Team, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022b\u003c/span\u003e). The predictive performance of each tool was evaluated using linear regression by comparing predicted values to measured values.\u003c/p\u003e\u003cp\u003ePrecision was defined as the degree of statistical variability, quantified using the coefficient of determination (R\u0026sup2;), which indicates how consistently a tool predicts values relative to the measured data. Accuracy, on the other hand, was assessed as systematic bias, represented by the slope and intercept of the linear regression. This approach follows established metrological definitions, where precision refers to the closeness of repeated measurements to each other, and accuracy refers to the closeness of the predictions to the true (measured) values.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results and discussion","content":"\u003cp\u003eFirst the coefficients of determination (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e) matrix between log MOE, log density and average MOE, MOR and density of boards are presented and discussed. Then the results from regression analysis to investigate the precision and bias of predicting log MOE and average board MOE from the destructive and non-destructive tools are presented. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e serves as an estimate of precision, indicating the proportion of variance explained by independent variable, while the slope and intercept of linear regression line reflect bias, indicating whether the model systematically underpredicts or overpredicts (accuracy or bias). Finally, the overall comparison between the tools regarding their predictive performance, operational capability and efficiency is presented.\u003c/p\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Correlations between all MOE and density measurement\u003c/h2\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the coefficients of determination (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e) between log MOE measured by various destructive and non-destructive tools, density measurement (green log density and estimated extracted log density), average board MOE and MOR. Measured_MOE exhibited moderate to strong positive correlations (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e) with other MOE measurement tools, ranging from \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.57\\)\u003c/span\u003e\u003c/span\u003e to \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.78\\)\u003c/span\u003e\u003c/span\u003e. The highest correlation with Measured_MOE was observed with HM200_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\)\u003c/span\u003e\u003c/span\u003e0.78), likely due to both being resonance-based methods applied to felled logs. This was followed by USMOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.67\\)\u003c/span\u003e\u003c/span\u003e), ST300_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.61\\)\u003c/span\u003e\u003c/span\u003e) and Resi_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.57\\)\u003c/span\u003e\u003c/span\u003e). However, these coefficients of determination (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e) alone do not determine which tools are superior, as multiple additional factors need to be considered for measuring or predicting log MOE. These factors include measurement time, accuracy or bias as defined above, sampling speed, and whether the tools are field-deployable for use in commercial settings (discussed further in section 4.4).\u003c/p\u003e\u003cp\u003eThe average MOE of the boards (Avg_Board_MOE), representing an average performance of the final product (sawn boards), exhibited a strong correlation with Measured_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.84\\)\u003c/span\u003e\u003c/span\u003e), USMOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.71\\)\u003c/span\u003e\u003c/span\u003e), and HM200_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.69\\)\u003c/span\u003e\u003c/span\u003e), indicating these methods can provide reliable predictions of average board quality. ST300_MOE also showed a moderate correlation (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.64\\)\u003c/span\u003e\u003c/span\u003e), while Resi_MOE demonstrated a weaker relationship (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.43\\)\u003c/span\u003e\u003c/span\u003e) with the Avg_Board_MOE. However, this is the average MOE of boards extracted from each log, which does not closely represent the grades recovery. The relatively lower predictability of Resi_MOE may be attributed to its dependence on calibration models and its measurement principle (torque), primarily linked to the wood density, which doesn\u0026rsquo;t provide a direct account of the microfibril angle (MFA) contribution to MOE and is influenced by other physical variables such as resin or water content. Further analysis could explore the underlying causes of variability in prediction accuracy among these tools, particularly for field-deployable methods like ST300 and Resi. These results suggest that as the degree of destructiveness of the tools increases, the proportion of variance explained in predicting log MOE and average board MOE also increases.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe green log density displayed low correlation with all the MOE and density measurements. This result is likely influenced by variations in moisture content, which can fluctuate due to partial drying during transportation, handling, and storage, as well as differences in resin content within the logs. Conversely, extracted log density (Extr_Log_Den_Est) showed strong to moderate coefficients of determination with all log MOE values (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e from 0.39 to 0.63). Extr_Log_Den_Est also exhibited a moderate positive correlation with average board MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.59\\)\u003c/span\u003e\u003c/span\u003e) and average board density (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.60\\)\u003c/span\u003e\u003c/span\u003e). This suggests that extracted log density serves as a more consistent predictor of overall board quality than the green log density for southern pine.\u003c/p\u003e\u003cp\u003eThis indicates that extracted log density is a more reliable metric for predicting both board and log MOE in southern pine species specially where resin and extractive is present, making it a preferable choice over green log density.Resi_Den is as good as Measured_MOE at predicting Extr_Log_Den_Est.\u003c/p\u003e\u003cp\u003eAs expected, there was a strong correlation between the non-destructive measurements Resi_Density and Resi_MOE (R\u0026sup2; = 0.81) as they are not statistically independent. The average board MOR shows a strong relationship with the average board MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.7\\)\u003c/span\u003e\u003c/span\u003e). However, average board MOR exhibited only weak correlations with log MOE measurements (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e values ranging from 0.19 to 0.47), with the strongest correlation observed with Measured_MOE. This suggests that while log MOE can serve as a reliable predictor for average board MOE, it is not an effective predictor for average board MOR, which can be impacted by knots, grain angle and other features such as compression failure.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Comparison of Log MOE via Linear Regression\u003c/h2\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the regression between Measured_MOE and other log MOE measured using various destructive and non-destructive tools differentiated by germplasm.\u003c/p\u003e\u003cp\u003eAs described in previous sections, the destructive method (e.g., HM200_MOE) exhibited the highest precision with Measured_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.78\\)\u003c/span\u003e\u003c/span\u003e), followed by the lab-based non-destructive method (USMOE, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.67\\)\u003c/span\u003e\u003c/span\u003e), and then the field-deployable non-destructive methods (Resi_MOE, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.57\\)\u003c/span\u003e\u003c/span\u003e, and ST300_MOE, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.61\\)\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWhen comparing the regression slopes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), the USMOE demonstrated the lowest bias (slope\u0026thinsp;=\u0026thinsp;0.95, close to 1), indicating a strong predictive power followed by HM200_MOE (a slope of 0.77). ST300_MOE exhibited the lowest accuracy or higher bias (slope\u0026thinsp;=\u0026thinsp;0.47), significantly overpredicting the actual MOE values by approximately 50%, most likely due to sampling of the high-MOE outerwood only. Conversely, Resi-MOE with a slope of 1.22 underpredicted the log MOE by 22%, potentially due to calibration dependencies. The Resi-MOE and the HM200_MOE are biased by about the same degree but in the opposite direction. These results indicate that the destructive and lab-based non-destructive technique outperformed field-deployable tools in predicting log MOE, having both higher precision and lower bias. However, these results are for limited set of data only for 68 trees.\u003c/p\u003e\u003cp\u003eOne of the four sampled genotypes, clonal variety C625 (orange circles in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), exhibited consistently low log MOE values, below 8,000 MPa. This resulted in reduced prediction accuracy and higher deviations from the regression lines, particularly for USMOE and Resi_MOE. The poor prediction performance for this germplasm may be attributed to its unique characteristics, such as growth characteristics (e.g. large juvenile wood proportion). Downes GM, Hogg, and Lee (2017) previously reported similar issues, attributing this to the abnormal behaviour of this specific clone. However, there were only 10 data points from C625. A larger dataset with more trees would provide better insight into these effects and overall comparison of the log MOE predictions.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e4.3 Comparison of average board MOE via Linear Regression\u003c/h2\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows linear regression models predicting average board MOE (Avg_Board_MOE) from destructive (HM200_MOE), lab-based non-destructive (USMOE), and field-deployable non-destructive methods (ST300_MOE and Resi_MOE). Among these, USMOE exhibited the highest precision (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.71\\)\u003c/span\u003e\u003c/span\u003e), followed closely by HM200_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.69\\)\u003c/span\u003e\u003c/span\u003e). A regression analysis was conducted between BING log MOE and average board MOE for comparison, yielding an R\u0026sup2; value of 0.84 and a slope of 0.94. For the field deployable non-destructive testing method measured on standing trees, ST300_MOE, showed a stronger relationship (R\u0026sup2; = 0.64) compared to Resi_MOE (R\u0026sup2; = 0.44). The high coefficients of determination for USMOE and HM200 indicate their strong ability to predict Avg_Board_MOE.\u003c/p\u003e\u003cp\u003eUSMOE demonstrated the least bias, with a slope of 0.97, which is close to 1, indicating that its predictions closely align with actual measured values (better accuracy). This makes USMOE a reliable non-destructive alternative for predicting average board stiffness. Resi_MOE also showed minimal bias (slope\u0026thinsp;=\u0026thinsp;1.04). In contrast, HM200_MOE, a destructive method, exhibited moderate bias (slope\u0026thinsp;=\u0026thinsp;0.71), suggesting a tendency to overestimate MOE, particularly for higher values. The highest bias was observed in ST300_MOE (slope\u0026thinsp;=\u0026thinsp;0.47), indicating that it significantly overestimates actual average board MOE.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSimilar to the log MOE predictions, one of the four sampled genotypes, clonal variety C625 (indicated by orange circles), exhibited the greatest deviation from the regression line, particularly for non-destructive measurements based on USMOE and Resi.\u003c/p\u003e\u003cp\u003eDestructive methods like HM200_MOE remain highly accurate but challenging to implement in field applications. Lab-based USMOE bridges the gap between accuracy and non-destructiveness, though its dependency on laboratory facilities limits its field portability. The results must be interpreted in consideration that this study sampled multiple plantings of varying genetic material and planting stocking. This may have impacted upon the relative usefulness of the different methods. For example, outerwood as sampled by the ST300 may be influenced to a greater extent by competition effects. These tools would typically (but not always) be used in more homogeneous stands. Field-deployable tools like ST300_MOE and Resi_MOE offer practical solutions for operational forestry, but the lower predictive accuracy of Resi highlights the need for improved metrology, calibration techniques and predictive models for MOE. These findings underscore the trade-offs between accuracy and operational efficiency when selecting tools for predicting Avg_Board_MOE.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e4.4 Overall comparison\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e provides a comprehensive comparison of destructive and non-destructive tools based on their measurement efficiency, operational capabilities, and predictive power (precision and bias or accuracy). As highlighted earlier, the destructive methods (BING_MOE and HM200_MOE) require tree felling and are not field-deployable, limiting their use in commercial forestry operations despite their high accuracy.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison of Destructive and Non-Destructive Tools for Log Stiffness Measurement for clearfall age Southern Pine trees\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePoint of Comparison\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBING (Destructive)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHM200 (Destructive)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eUSMOE (Non-destructive)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eResi (Non-destructive)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eST300 (Non-destructive)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTree Felling Required\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes, felling required\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eYes, felling required\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNo, measures standing trees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNo, measures standing trees\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNo, measures standing trees\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eField Deployable\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNot field-deployable\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNot field-deployable\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eField-deployable but lab work needed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eField-deployable, no lab needed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eField-deployable, no lab needed\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eType of MOE Measurement\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDirect measurement, not calibration-based\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDirect measurement, not calibration-based\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eDirect measurement, not calibration-based\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eCalibration-based (requires training samples)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eCalibration-based (requires approximate density)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSampling Rate (contiguous trees per hour)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSlow: approximately 5 trees/hour\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSlow: approximately 5 trees/hour\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSlow: approximately 4 trees/hour\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eFast: approximately 50 trees/hour\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eModerate: approximately 20 trees/hour\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eData Analysis Speed\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFast\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFast\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eFast\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eFast\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRadial Variation Measurement\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eFull Radial Analysis (e.g., Age, Density, Extractives)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eYes, full analysis possible (e.g., cores used for extractives, age, MFA, density)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eYes, partial analysis possible (e.g. density variation)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePrecision of Log MOE Prediction\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBias of Log MOE Prediction\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePrecision of Individual Board MOE\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNot possible\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNot possible\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePossible\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePossible\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNot possible\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePrecision of Average Board MOE Prediction\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate to High\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eModerate to High\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eLow to Moderate*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eBias of Average Board MOE Prediction\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* Resi accuracy may be improved through calibration and improved predictive models for MOE\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn terms of bias for log MOE prediction, USMOE (0.95) demonstrated the lowest bias, followed by HM200_MOE (0.77). Resi_MOE (1.22) exhibited a similar level of bias to HM200 but in the opposite direction, while ST300_MOE (0.47) had the highest bias, significantly overestimating log MOE. For predicting board MOE, USMOE (0.97) and Resi_MOE (1.04) exhibited the lowest bias, whereas ST300_MOE (0.47) had the highest bias, significantly overestimating board MOE. HM200_MOE (0.71) showed moderate bias, slightly underpredicting board MOE.\u003c/p\u003e\u003cp\u003eAs previously noted, ST300 measures the outer wood, and the bias can be corrected provided the physical meaning of stress wave speed is adequately understood and accurately modelled. Additionally, the ST300 is becoming increasingly difficult to service, and its availability from suppliers is gradually declining. However, Resi and ST300 excel in their operational efficiency, with faster sampling rates and no requirement for laboratory work, making them highly suitable for field applications where rapid assessment is prioritised. The Resi tools rely on statistical models which rely on training samples. This approach can introduce uncertainty in the predictive power. Therefore, the type of calibration techniques and the number and quality of training samples are crucial. USMOE, another non-destructive method, requires laboratory measurements of density and ultrasound velocity, making it less practical for large-scale field operations. However, USMOE can provide more accurate estimation (low bias and high precision) of log MOE and average board MOE.\u003c/p\u003e\u003cp\u003eHowever, with ongoing technological advancements, USMOE could become more efficient if density and ultrasonic transmission velocity are measured faster or if alternative methods\u0026mdash;such as segmental MOE measurement using NIR\u0026mdash;are employed. Improvements in coring technology, including increased speed, reduced effort, greater portability, and innovative drilling techniques, would also make this approach better suited for field measurements.\u003c/p\u003e\u003cp\u003eAmong the tools compared, only USMOE allow for the measurement of radial MOE variation. As a direct measurement method, USMOE offers additional capabilities such as the prediction of individual board MOE and enabling detailed analyses such as within-tree variation and growth trends. This makes it particularly valuable for research applications requiring a deeper understanding of wood properties. Meanwhile, Resi calibration could be improved in the future to provide radial MOE profiles for suitable species, potentially expanding its analytical applications. Additionally, the cores obtained from standing trees can also be used for other technical traits or research purposes, such as dendrochronology-related investigations; other useful measurements, such as resin content, chemical composition analysis, and predictions of quality of products derived from those trees, can be made using the cores if necessary. For example, Psaltis et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) used the segment\u0026rsquo;s MOE measured by ultrasound to predict the individual board MOE that can be obtained from a forest. In this study, the USMOE, using a 2D integration approach, can reconstruct an estimate of MOE for individual boards that can be sawn from trees with moderate accuracy (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.53) and low bias (2%).\u003c/p\u003e\u003cp\u003eOverall, the choice of tool depends on the specific application context:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eDestructive tools (BING and HM200) are ideal for research and high-precision applications but are impractical for operational forestry due to their lack of field-deployability.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eUSMOE bridges the gap between accuracy and potential field applicability, with future advancements likely to enhance its practicality.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eResi and ST300 are better suited for large-scale commercial operations, offering efficiency and portability despite their relatively lower accuracy.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eFurther research should focus on improving the calibration models for accuracy of field-deployable tools like Resi and ST300 and on advancing USMOE toward a fully field-compatible solution. These developments could significantly enhance the efficiency and accuracy of non-destructive log quality assessments.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study compared and evaluated the performance of various destructive and non-destructive tools for log MOE and board MOE estimation of plantations grown Southern Pine. The study included log MOE measurement using two destructive tools, BING (Measured_MOE) and HM200, along with three non-destructive tools: one lab-based (USMOE) and two field-deployable (ST300 and Resi).\u003c/p\u003e\u003cp\u003eThe accuracy of the tools increases with the degree of destructiveness. The destructive tool (HM200_MOE) was the most precise in predicting log MOE ((\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.78\\)\u003c/span\u003e\u003c/span\u003e) followed by \u003cem\u003eUSMOE\u003c/em\u003e (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.67\\)\u003c/span\u003e\u003c/span\u003e), \u003cem\u003eST300_MOE\u003c/em\u003e (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.61\\)\u003c/span\u003e\u003c/span\u003e) and \u003cem\u003eResi_MOE\u003c/em\u003e (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.57\\)\u003c/span\u003e\u003c/span\u003e). The prediction of board MOE followed the same trend with higher precision and less bias using USMOE than HM200_MOE. Additionally, Germplasm C625 consistently exhibited low log MOE values (\u0026lt;\u0026thinsp;8,000 MPa), leading to reduced prediction accuracy, particularly for USMOE and Resi_MOE. These deviations align with prior reports attributing this behaviour to unique characteristics of this specific clone. While extracted log density exhibited strong correlations with both log MOE and average board MOE for this species of tree, green log density showed a very weak correlation.\u003c/p\u003e\u003cp\u003eThe choice of tool depends on the specific application requirements. For high accuracy in research or precision-demanding operations, destructive methods remain the most reliable. For non-destructive tools like USMOE, Resi and ST300 offer a practical balance between predictive power and operational efficiency. Future research should focus on improving the calibration models for non-destructive tools to enhance their accuracy and consistency across diverse field conditions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eC.K. drafted the manuscript, conducted data analysis, and prepared figures and tables.V.G. contributed to the manuscript review and editing and the conceptual development of the introduction.I.L. and D.K. reviewed and edited the manuscript and contributed to site selection, data collection, and tree harvesting.G.D. reviewed the manuscript and validated the resistograph data.D.L. contributed to experimental planning, secured funding, data collection, and provided critical review and manuscript editing.H.B. contributed to the experimental method, conceptual development, reviewed and edited the manuscript\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThis study combines information from two projects: \u0026ldquo;Improving returns from Southern Pine plantations through innovative resource characterisation\u0026rdquo; Project PNC361-1415 and \u0026ldquo;Assessing and managing mid-rotation wood quality in Australian softwood plantations to produce fit-for-purpose logs\u0026rdquo;. Project PNB548-2021\u0026rdquo;. The authors gratefully acknowledge the financial support for the project funded by Forest and Wood Products Australia, DAF Forest Industries, the University of the Sunshine Coast and industry partners: HQPlantations, Forest Corporation New South Wales, Hancock Victoria Plantations and Hyne Timber Pty Ltd. We acknowledge the contribution by all staff at DAF Gympie and Salisbury Research Facility including Henri Bailleres, Gary Hopewell, Rhianna Robinson, Rica Minett, Chris Fitzgerald, Marco Lausberg, Tony Burridge, and John Oostenbrink.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAllaire, J., Xie, Y., Dervieux, C., McPherson, J., Luraschi, J., Ushey, K., \u0026amp; Atkins, A. (2023). Rmarkdown: Dynamic Documents for r. https://CRAN.R-project.org/package=rmarkdown. \u003c/li\u003e\n\u003cli\u003eAS/NZS 4063.1. (2010). Characterization of structural timber, Part 1: Test methods. \u003cem\u003eStandards Australia / Standards New Zealand\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eBaill\u0026egrave;res, H., Hopewell, G. P., \u0026amp; Boughton, G. (2009). MOE and MOR assessment technologies for improving graded recovery of exotic pines in Australia. \u003cem\u003eForest \u0026amp; Wood Products Australia,, Project no: PNB040-0708 \u003c/em\u003e\u003c/li\u003e\n\u003cli\u003eBaill\u0026egrave;res, H., Lee, D., Kumar, C., Psaltis, S., Hopewell, G. P., \u0026amp; Brancheriau, L. (2019). Improving returns from southern pine plantations through innovative resource characterisation. \u003cem\u003eForest \u0026amp; Wood Products Australia,, Project no: PNC361-1415\u003c/em\u003e(ISBN 978-1-925213-89-8). \u003c/li\u003e\n\u003cli\u003eBrancheriau, L. (2014). An alternative solution for the determination of elastic parameters in free\u0026ndash;free flexural vibration of a Timoshenko beam. \u003cem\u003eWood Science and Technology, 48\u003c/em\u003e(6), 1269-1279. doi:10.1007/s00226-014-0672-x\u003c/li\u003e\n\u003cli\u003eBuksnowitz, C., M\u0026uuml;ller, U., Evans, R., Teischinger, A., \u0026amp; Grabner, M. (2007). The potential of SilviScan\u0026rsquo;s X-ray diffractometry method for the rapid assessment of spiral grain in softwood, evaluated by goniometric measurements. \u003cem\u003eWood Science and Technology, 42\u003c/em\u003e, 95-102. doi:10.1007/s00226-007-0153-6\u003c/li\u003e\n\u003cli\u003eCown, D. (1978). Comparison of the Pilodyn and torsiometer methods for the rapid assessment of wood density in living trees. \u003cem\u003eNew Zealand Journal of Forestry Science, 8\u003c/em\u003e(3), 384-391. \u003c/li\u003e\n\u003cli\u003eDownes, G., Hudson, I., Raymond, C., Dean, G., Michell, A., Schimleck, L., . . . Muneri, A. (1997). \u003cem\u003eSampling plantation Eucalypts for wood and fibre properties. \u003c/em\u003e. Retrieved from Melbourne: \u003c/li\u003e\n\u003cli\u003eDownes, G., \u0026amp; Lausberg, M. (2016). Evaluation of the RESI software tool for the prediction of HM200 within pine logs sourced from multiple sites across New Zealand and Australia. \u003cem\u003eNZ Solid Wood Innov, 15\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eDownes, G., Lausberg, M., Potts, B., Pilbeam, D., Bird, M., \u0026amp; Bradshaw, B. (2018). Application of the IML Resistograph to the infield assessment of basic density in plantation eucalypts. \u003cem\u003eAustralian Forestry, 81\u003c/em\u003e(3), 177-185. \u003c/li\u003e\n\u003cli\u003eDownes GM, Hogg, B., \u0026amp; Lee, D. (2017). Evaluating the application of the IML Resistograph to the prediction of key wood properties of the Southern Yellow Pine. \u003cem\u003eUniversity of the Sunshine Coast., In Milestone report 5a-d FWPA PNC361-1415\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eEvans, R., Ilic, J., \u0026amp; Matheson, C. (2000). \u003cem\u003eRapid estimation of solid wood stiffness using SilviScan.\u003c/em\u003e Paper presented at the Proceedings of 26th Forest Products Research Conference: Research developments and industrial applications and Wood Waste Forum, Clayton, Victoria, Australia, 19-21 June 2000.\u003c/li\u003e\n\u003cli\u003eFaydi, Y., Brancheriau, L., Pot, G., \u0026amp; Collet, R. (2017). Prediction of Oak Wood Mechanical Properties Based on the Statistical Exploitation of Vibrational Response. \u003cem\u003eBioResources, 12\u003c/em\u003e(3), 5913-5927. \u003c/li\u003e\n\u003cli\u003eGao, S., Wang, X., Wiemann, M. C., Brashaw, B. K., Ross, R. J., \u0026amp; Wang, L. (2017). A critical analysis of methods for rapid and nondestructive determination of wood density in standing trees. \u003cem\u003eAnnals of Forest Science, 74\u003c/em\u003e(2), 27. doi:10.1007/s13595-017-0623-4\u003c/li\u003e\n\u003cli\u003eGendvilas, V., Neyland, M., Rocha-Sep\u0026uacute;lveda, M. F., Downes, G. M., Hunt, M., Jacobs, A., . . . O\u0026rsquo;Reilly-Wapstra, J. (2022). Effects of thinning on the longitudinal and radial variation in wood properties of Eucalyptus nitens. \u003cem\u003eForestry: An International Journal of Forest Research, 95\u003c/em\u003e(4), 504-517. doi:10.1093/forestry/cpac007\u003c/li\u003e\n\u003cli\u003eGrissino-Mayer, H. D. (2003). A Manual and Tutorial for the Proper Use of an Increment Borer. \u003cem\u003eTree-ring research, 59\u003c/em\u003e, 63-79. \u003c/li\u003e\n\u003cli\u003eKumar, C., Psaltis, S., Bailleres, H., Turner, I., Brancheriau, L., Hopewell, G., . . . Lee, D. J. (2021). Accurate estimation of log MOE from non-destructive standing tree measurements. \u003cem\u003eAnnals of Forest Science, 78\u003c/em\u003e(1). doi:10.1007/s13595-021-01031-w\u003c/li\u003e\n\u003cli\u003eLee, J. (2024). Meet DiscBot, our new wood quality detective. \u003cem\u003eScion\u003c/em\u003e. Retrieved from https://www.scionresearch.com/about-us/about-scion/corporate-publications/scion-connections/past-issues-list/issue-17,-september-2015/meet-discbot,-our-new-quality-detective\u003c/li\u003e\n\u003cli\u003eOndrejka, V., Gergeľ, T., Bucha, T., \u0026amp; P\u0026aacute;stor, M. (2021). Innovative methods of non-destructive evaluation of log quality. \u003cem\u003eCentral European Forestry Journal, 67\u003c/em\u003e(1), 3-13. \u003c/li\u003e\n\u003cli\u003eParadis, S., Brancheriau, L., \u0026amp; Baill\u0026egrave;res, H. (2017). \u003cem\u003eBing: Beam Identification by Non destructive Grading\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003ePsaltis, S., Kumar, C., Turner, I., Carr, E. J., Farrell, T., Brancheriau, L., . . . Lee, D. J. (2021). A new approach for predicting board MOE from increment cores. \u003cem\u003eAnnals of Forest Science, 78\u003c/em\u003e(3). doi:10.1007/s13595-021-01093-w\u003c/li\u003e\n\u003cli\u003eR Core Team. (2022a). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. \u003c/li\u003e\n\u003cli\u003eR Core Team. (2022b). RStudio: Integrated development for r. Computer Program, RStudio, PBC, Boston, MA. http://www.rstudio.com/. \u003c/li\u003e\n\u003cli\u003eRinn, F., Schweingruber, F., \u0026amp; Sch\u0026auml;r, E. (1996). RESISTOGRAPH and X-Ray Density Charts of Wood. Comparative Evaluation of Drill Resistance Profiles and X-ray Density Charts of Different Wood Species. \u003cem\u003eHolzforschung, 50\u003c/em\u003e, 303-311. doi:10.1515/hfsg.1996.50.4.303\u003c/li\u003e\n\u003cli\u003eSchimleck, L., Dahlen, J., Apiolaza, L. A., Downes, G., Emms, G., Evans, R., . . . Wang, X. (2019). Non-destructive evaluation techniques and what they tell us about wood property variation. \u003cem\u003eForests, 10\u003c/em\u003e(9), 728. \u003c/li\u003e\n\u003cli\u003eSchimleck, L. R., Evans, R., \u0026amp; Matheson, A. C. (2002). Estimation ofPinus radiata D. Don clear wood properties by near-infrared spectroscopy. \u003cem\u003eJournal of Wood Science, 48\u003c/em\u003e(2), 132-137. doi:10.1007/BF00767290\u003c/li\u003e\n\u003cli\u003eWang, X., \u0026amp; Ross, R. J. (2002). Nondestructive evaluation of green materials\u0026ndash;recent research and development activities. \u003cem\u003eNondestructive evaluation of wood. Forest Products Society, Madison\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eWessels, C. B., Malan, F. S., \u0026amp; Rypstra, T. (2011). A review of measurement methods used on standing trees for the prediction of some mechanical properties of timber. \u003cem\u003eEuropean Journal of Forest Research, 130\u003c/em\u003e(6), 881-893. doi:10.1007/s10342-011-0484-6\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"european-journal-of-wood-and-wood-products","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"harw","sideBox":"Learn more about [European Journal of Wood and Wood Products](http://link.springer.com/journal/107)","snPcode":"107","submissionUrl":"https://submission.nature.com/new-submission/107/3","title":"European Journal of Wood and Wood Products","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7050599/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7050599/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eVarious non-destructive technologies have been employed to assess log quality, facilitating informed decision regarding sorting, segregation and processing decisions. However, there is a lack of comprehensive comparative evaluations of non-destructive tools for MOE assessment in standing trees, specifically regarding their accuracy, efficiency, practical advantages, and their correlation with the quality of sawn boards. Therefore, this study first investigated the correlations between log MOE, log density, and average MOE and MOR of boards obtained from each log. Then, the study evaluated the predictive performance of estimating measured log MOE (Measured_MOE) and average board MOE using a destructive (HM200_MOE), a lab-based non-destructive (USMOE), and two field-deployable non-destructive (ST300_MOE and Resi_MOE) log MOE measurement tools. Finally, a comparison between the destructive and non-destructive tools for log stiffness measurement, focusing on deployability, efficiency, and prediction power was presented. The results showed that as tools become less destructive, their predictive power in estimating log MOE decreases. The destructive method, HM200_MOE, was the most precise (with an 0.78), in estimating log MOE due to its direct measurement on felled logs, followed by the non-destructive m\u003c/p\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:\\)\u003c/span\u003e\u003c/span\u003eethods: USMOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.67\\)\u003c/span\u003e\u003c/span\u003e), ST300_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.61\\)\u003c/span\u003e\u003c/span\u003e), and Resi_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.57\\)\u003c/span\u003e\u003c/span\u003e). USMOE explained the highest variability in the average board MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.71\\)\u003c/span\u003e\u003c/span\u003e), followed by HM200_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.69\\)\u003c/span\u003e\u003c/span\u003e), ST300_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.64\\)\u003c/span\u003e\u003c/span\u003e), and Resi_MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=0.43\\)\u003c/span\u003e\u003c/span\u003e). Green log density showed very weak correlation with log MOE and average board MOE. In contrast, resin extracted log density had moderate to strong correlations with log MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}\\)\u003c/span\u003e\u003c/span\u003e from 0.39 to 0.63) and a high positive correlation with average board MOE (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}^{2}=\\:0.84\\)\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDestructive methods like HM200 are very precise but costly and unsuitable for field applications. Lab-based non-destructive tools such as USMOE can achieve a balance between precision and accuracy and potential field applicability but are slower, more expensive, and not field-deployable like Resi_MOE and ST300_MOE. Field-deployable tools, such as ST300 and Resi, offer practical solutions for operational forestry due to their efficiency and portability. However, their reduced predictive performance underscores the need for improved precision and accuracy through improved modelling or calibration. These findings highlight the trade-offs between predictive performance and operational efficiency, with the choice of tools depending on the required predictive performance, field deployability, sampling needs, and the cost of equipment and measurements.\u003c/p\u003e","manuscriptTitle":"Comparison of Non-Destructive Tools for Measuring MOE of Southern Pine Trees","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-08 10:21:47","doi":"10.21203/rs.3.rs-7050599/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-17T17:03:47+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-17T17:02:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"193204598907144142212349707493016765424","date":"2025-11-17T17:00:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-01T08:00:19+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-20T15:34:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"110795433137995744702183701207012974695","date":"2025-08-06T06:11:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"23769759200668378192211205325240128921","date":"2025-08-05T15:33:45+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-05T12:29:34+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-28T15:11:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-09T12:52:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"European Journal of Wood and Wood Products","date":"2025-07-05T05:09:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"european-journal-of-wood-and-wood-products","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"harw","sideBox":"Learn more about [European Journal of Wood and Wood Products](http://link.springer.com/journal/107)","snPcode":"107","submissionUrl":"https://submission.nature.com/new-submission/107/3","title":"European Journal of Wood and Wood Products","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"283eddee-138b-4f81-ba4f-47881875fa09","owner":[],"postedDate":"August 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-20T15:11:05+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-08 10:21:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7050599","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7050599","identity":"rs-7050599","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}