Big data is trans­form­ing forestry

Australian Forests and Timber - - Foresttech 17 -

For­est ma­chines are ex­pen­sive and hav­ing them stand idle is even more costly. It is dif­fi­cult to pre­dict the con­di­tions of for­est work, be­cause they are af­fected by the weather, sea­son, soil prop­er­ties, el­e­va­tion dif­fer­ences and the de­vel­op­men­tal phases of tree species. If for­est work is done at the wrong time, it causes un­for­tu­nate en­vi­ron­men­tal dam­age. Heavy ma­chines can eas­ily sink into clay soil.

The first step in pre­vent­ing en­vi­ron­men­tal dam­age is to un­der­stand the for­est’s con­di­tions and the changes that af­fect it.

Here, big data and new ways of cre­at­ing prediction mod­els based on big data, are help­ful. Big data means the col­lec­tion and anal­y­sis of ex­tremely large amounts of data. The prediction mod­els be­ing de­vel­oped in an EFFORTE project in­di­cate when and with what kind of equip­ment it is worth go­ing to the for­est with.

“Prediction mod­els are pri­mar­ily a tool for plan­ning for­est work,” says Project Co­or­di­na­tor Jori Uusi­talo.

Go­ing into forests with fine-grained loam and silt soils, as well as peat soils, is done dur­ing the dry time in high sum­mer or win­ter.

Dur­ing the rest of the year, the work is done on ground with high bear­ing ca­pac­ity.

“When the for­est’s soil type and the prop­er­ties of it can be as­sessed, work can be of­fered for for­est ma­chines all year round.”

Big data has al­ready been col­lected pre­vi­ously on the ground’s el­e­va­tion changes, for ex­am­ple. The EFFORTE project aims to com­bine dif­fer­ent big data archives.

“For ex­am­ple, weather data com­bined with changes in el­e­va­tion tells us where rain­wa­ter ac­cu­mu­lates af­ter a shower. For­est work can be planned more pre­cisely with this in­for­ma­tion.”

Big data tells us how the for­est is feel­ing right now. Soon, re­search groups and com­pa­nies do­ing de­vel­op­ment will not have to separately col­lect their data. As mea­sure­ment sys­tems and sen­sors de­velop, the data will ac­crue while the ma­chines work in the for­est.

As mea­sur­ing sys­tems and sen­sors de­velop, sys­tems will learn to pre­dict con­di­tions more pre­cisely. The data that ac­cu­mu­lates along­side for­est work will im­prove and flex­i­bly al­ter prediction mod­els as con­di­tions change.

Prediction mod­els can be cre­ated on, for ex­am­ple, how ‘sinky’ the soil is or the qual­ity of the for­est’s trees.

The tracks left on the ground by a for­est ma­chine tyre are an in­di­ca­tor of how much the heavy ma­chine sinks into the soil. The depth of the track can be mea­sured with the help of the en­gine power.

”The more the tyre sinks, the more power the ma­chine needs to move for­ward,” says Uusi­talo.

The tracks, com­bined with weather mea­sure­ments, let you know when it is a good time to head to the for­est.

The qual­ity of wood can be pre­dicted by ex­am­in­ing other trees that have been cut down in the area. The log­ger takes a photo of the felled tree’s stem. Count­ing the an­nual rings pro­vides in­for­ma­tion on the growth rate of the tree. The har­vester cal­cu­lates the di­am­e­ter and length of the trunk as the prun­ing and cut­ting pro­ceeds. The re­sult is called a trunk pro­file.

“The growth rate and trunk pro­file tell us about, for ex­am­ple, the wood’s den­sity, firm­ness, mois­ture con­tent and rot­ten­ness. The in­for­ma­tion helps us op­ti­mize what kind of prod­ucts to make out of the wood.”

Smart ways of mea­sur­ing and analysing big data will ul­ti­mately lead to the forester get­ting to know the for­est bet­ter. Dif­fer­ent con­di­tions re­quire dif­fer­ent pro­ce­dures.

“A for­est hol­low is hu­mid, but the hill­side is dry. A large for­est is dif­fer­ent in dif­fer­ent cor­ners,” Uusi­talo ex­plains.

Pre­cise in­for­ma­tion on a for­est and its soil helps in tar­get­ing the work. This is called pre­ci­sion forestry. The work can al­ready be op­ti­mized when saplings are planted.

“Dif­fer­ent tree species thrive un­der dif­fer­ent con­di­tions, so the for­est’s soil, dry­ness, hu­mid­ity and el­e­va­tion dif­fer­ences have to be taken into ac­count. If a species of tree thrives in dry sur­round­ings, dry tus­socks are built for its saplings in wa­ter­logged parts of the for­est,”, says Uusi­talo.

Forestry is a chain con­tain­ing many links, where each stage of work af­fects the costs of the next one. Peo­ple may fail to no­tice cer­tain de­tails, which the prediction mod­els are able to in­di­cate in good time.

”When you un­der­stand the soil’s prop­er­ties and growth po­ten­tial, you are able to take into ac­count the to­tal costs and to­tal re­turn.”

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.