Difference between revisions of "Version 1.5"
Jump to navigation
Jump to search
| Line 4: | Line 4: | ||
==Summary== | ==Summary== | ||
| − | * In version 1.4, an error was fixed in the calculation of dominant height. This affected the thinning guide so that thinnings were simulated more seldom. However, the number of thinnings were so few in version 1.4 | + | * In version 1.4, an error was fixed in the calculation of dominant height. This affected the thinning guide so that thinnings were simulated more seldom. However, the number of thinnings were so few in version 1.4 so we suspected another error. A bug was found that prevented second and later thinnings to take place, but htis problem had been "overridden" by the error in dominant height. You can say that a sceleton fell out of the closet. |
* Abundance of large trees are important for biodiversity. You can now define your own result variables for volume or number of large trees, and select species and diameter criteria for these variables. Read more [[Result_Variables#User-defined_result_variables | here]]. | * Abundance of large trees are important for biodiversity. You can now define your own result variables for volume or number of large trees, and select species and diameter criteria for these variables. Read more [[Result_Variables#User-defined_result_variables | here]]. | ||
* En error in tree hieght calculations has been fixed. The fixed lead decreased tree heights, and you can expect 1-5 % lower growth rates than in previous versions. | * En error in tree hieght calculations has been fixed. The fixed lead decreased tree heights, and you can expect 1-5 % lower growth rates than in previous versions. | ||
* Prediction of mortality and damages in young stands has been modified to be better adapted to tree group list (where a tree object may represent several trees). Also, some bugs have been fixed. Mortality rates are often slightly higher in version 1.5, and amount of non-lethal damages that cause height growth reductions are slightly lower. The effect on total production is probably less than 1 %. | * Prediction of mortality and damages in young stands has been modified to be better adapted to tree group list (where a tree object may represent several trees). Also, some bugs have been fixed. Mortality rates are often slightly higher in version 1.5, and amount of non-lethal damages that cause height growth reductions are slightly lower. The effect on total production is probably less than 1 %. | ||
| − | * When making a prognosis for a period length less than 5 years, interpolation is applied to each tree object. In previous verioins, linear intepolation was used for tree diameter, stem number and height. In 'an ongoing validation project', it was discovered that the interpolation ''per se'' causes deviations of up to 2 % on stand volume after 20 years, if starting with an interpolation to 2.5 years (to use period midpoints). A new non-linear interpolation algorithm was therefore implemented that minimizes the interpolation error to neglectable values. The algorithm used growth, change of stems, etc. for two growth period ahead to estimate the rate of change (i.e. second derivative) for each tree attribute. This information is then used to compute interpolated values. | + | * When making a prognosis for a period length less than 5 years, interpolation is applied to each tree object. In previous verioins, linear intepolation was used for tree diameter, stem number and height. In 'an ongoing validation project', it was discovered that the interpolation ''per se'' causes deviations of up to 2 % on stand volume after 20 years, if starting with an interpolation to 2.5 years (to use period midpoints). A new non-linear interpolation algorithm was therefore implemented that minimizes the interpolation error to neglectable values. The algorithm used growth, change of stems, etc. for two growth period ahead to estimate the rate of change (i.e. second derivative) for each tree attribute. This information is then used to compute interpolated values. |
| + | |||
==Details== | ==Details== | ||
More details can be found here: | More details can be found here: | ||
Revision as of 12:57, 31 August 2011
 |
Languages: |
[[::Version 1.5|English]] |
Summary
- In version 1.4, an error was fixed in the calculation of dominant height. This affected the thinning guide so that thinnings were simulated more seldom. However, the number of thinnings were so few in version 1.4 so we suspected another error. A bug was found that prevented second and later thinnings to take place, but htis problem had been "overridden" by the error in dominant height. You can say that a sceleton fell out of the closet.
- Abundance of large trees are important for biodiversity. You can now define your own result variables for volume or number of large trees, and select species and diameter criteria for these variables. Read more here.
- En error in tree hieght calculations has been fixed. The fixed lead decreased tree heights, and you can expect 1-5 % lower growth rates than in previous versions.
- Prediction of mortality and damages in young stands has been modified to be better adapted to tree group list (where a tree object may represent several trees). Also, some bugs have been fixed. Mortality rates are often slightly higher in version 1.5, and amount of non-lethal damages that cause height growth reductions are slightly lower. The effect on total production is probably less than 1 %.
- When making a prognosis for a period length less than 5 years, interpolation is applied to each tree object. In previous verioins, linear intepolation was used for tree diameter, stem number and height. In 'an ongoing validation project', it was discovered that the interpolation per se causes deviations of up to 2 % on stand volume after 20 years, if starting with an interpolation to 2.5 years (to use period midpoints). A new non-linear interpolation algorithm was therefore implemented that minimizes the interpolation error to neglectable values. The algorithm used growth, change of stems, etc. for two growth period ahead to estimate the rate of change (i.e. second derivative) for each tree attribute. This information is then used to compute interpolated values.
Details
More details can be found here: Release Notes 1.5 (Click "Cancel" if a login-screen is displayed.)
Next release
Ocotober-november 2011