Mod­el­ling the per­fect pine plan­ta­tion

Australian Forests and Timber - - FWPA FEATURE - By Matthew Lover­ing

DE­VEL­OP­ING A soft­ware model that pre­dicts the wood qual­ity prop­er­ties (such as wood den­sity, stiff­ness and growth) for how a tree will grow un­der a unique but chang­ing set of en­vi­ron­men­tal con­di­tions is a com­plex un­der­tak­ing. Dr David Drew and his re­search team from CSIRO and Dr Ge­off Downes (now with For­est Qual­ity Pty. Ltd.) have spent the last three years build­ing a model of tree growth in Pi­nus ra­di­ata which pro­vides ex­plicit pre­dic­tions of wood prop­erty vari­a­tion based both on ex­tremely fine level data (such as how wood cells di­vide and grow) and larger scale en­vi­ron­men­tal in­puts such as rain­fall, tem­per­a­ture, soil qual­ity and— per­haps most im­por­tantly to plan­ta­tion man­agers— sil­vi­cul­ture man­age­ment prac­tices.

With sup­port from For­est and Wood Prod­ucts Aus­tralia ( FWPA), Scion, HVP and ForestrySA, David and the re­search team have cre­ated ‘e- Cam­bium’, a pro­cess­based model that takes into ac­count the bi­o­log­i­cal knowl­edge of how a tree uses wa­ter and nu­tri­ents to build wood cells when af­fected by dif­fer­ent cli­mate and en­vi­ron­men­tal con­di­tions. It is de­signed pri­mar­ily as a tool to let for­est man­agers pre­dict tree stem growth as well as wood den­sity (and stiff­ness) re­sponses un­der a range of con­di­tions. The great value of such a process-based model is that the user can ex­plore be­yond ex­ist­ing data and field ex­pe­ri­ence to pre­dict stand growth re­sponses and tree per­for­mance un­der fu­ture con­di­tions for which they may not be any prece­dent (e.g. in­creas­ing av­er­age tem­per­a­tures, or a new sil­vi­cul­tural in­ter­ven­tion). The model can in­cor­po­rate out­puts from two ex­ist­ing tree-growth mod­els: the more com­plex CA­BALA model, which gives very de­tailed re­sults, or the sim­pler to run and use 3PG model (stand­ing for Phys­i­o­log­i­cal Prin­ci­ples Pre­dict­ing Growth), which gives slightly less ac­cu­rate mod­el­ing re­sults.

“This model is valu­able in that, with the length of time it takes to grow a tree, if you do make a mis­take early on you don’t have much of a chance to rec­tify it,” says David. “There­fore some sort of sce­nario-based ex­plo­ration tool be­comes very use­ful. If the tool says a par­tic­u­lar sil­vi­cul­ture man­age­ment is the worst thing you could ever do then it pro­vides a ba­sis for reeval­u­at­ing a given ap­proach and ex­plor­ing al­ter­na­tives. That’s where this model has the po­ten­tial to be a real dol­lar saver.”

The team mea­sured cli­mate data and tree growth at six sites: four in Aus­tralia and two in New Zealand. At the four Aus­tralian sites about 40 trees (aged 16 to 18 years old) were mea­sured ev­ery day for two and a half years us­ing au­to­mated den­drom­e­ters. These pre­ci­sion tools were at­tached to each tree to mea­sure the rate the tree grew in 15 minute in­ter­vals, while also record­ing cli­mate con­di­tions. The data was sent wire­lessly back to the re­search team, which was prob­a­bly just as well as at one site the den­drom­e­ter ca­bles were chewed by wom­bats re­sult­ing in data not be­ing recorded. At an­other site the so­lar panel that recharged the bat­tery of a den­drom­e­ter was dam­aged by a speed­ing kan­ga­roo!

At the end of the two and a half year mon­i­tor­ing pe­riod pith-to- bark core sam­ples were taken from each tree at, or as close as pos­si­ble to, where the den­drom­e­ter had been mea­sur­ing growth. These core sam­ples were an­a­lysed us­ing Sil­viS­can tech­nol­ogy (de­vel­oped by Dr Robert Evans) to mea­sure wood prop­er­ties such as den­sity. This al­lowed David to match wood prop­er­ties with daily pat­terns of growth and en­vi­ron­men­tal im­pacts (such as tem­per­a­ture and rain­fall).

While e- Cam­bium may have been based on pre- ex­ist­ing the­o­ret­i­cal mod­els of tree growth it evolved sig­nif­i­cantly as the vol­ume of in­cred­i­bly de­tailed data grew.

‘The model started off highly the­o­ret­i­cal but as the data came in, it gave us the op­por­tu­nity to re­fine things and im­prove the qual­ity of the frame­work. The model changed more than I ex­pected as the data started to stream in, and there were some ma­jor changes we had to make to our ear­lier the­o­ret­i­cal con­cepts to ac­com­mo­date it, but that was a very good thing. It was ex­actly what we wanted from all that data!’ says David. ‘It’s great that the forestry in­dus­try ap­pre­ci­ates that the mod­el­ing process is a back­wards–for­wards process.’

The e- Cam­bium model, with its al­go­rithms de­rived from data from the six mea­sure­ment sites, was cal­i­brated against wood prop­erty data from an­other 10 sites. Ex­cit­ingly, the model was able to pre­dict around 80 per cent (us­ing CA­BALA in­puts) of the vari­abil­ity seen in wood den­sity in trees from these sites, and around 70 per cent ac­cu­racy us­ing 3PG. Although pre­dict­ing fi­nal stand vol­ume wasn’t the pri­or­ity of the project, David says ‘The in­di­ca­tions are strong that the model would have a sim­i­lar high level of ac­cu­racy in pre­dict­ing wood vol­umes from a site.’

‘In the “bet­ter/ faster/ cheaper” par­a­digm for busi­ness this re­search is help­ing build up the “bet­ter”,’ says David. ‘The stage we’re at now is that we’ve built a testable model that in­dus­try can ex­per­i­ment with and help mod­ify. Cur­rently HVP are ex­per­i­ment­ing with the model and get­ting rea­son­able re­sults. They can see the real po­ten­tial for it in their fu­ture de­ci­sion mak­ing.’

The cur­rent e- Cam­bium model is avail­able to all FWPA mem­bers for test­ing us­ing their pine plan­ta­tions and to ex­plore the ef­fects of dif­fer­ent sites, sil­vi­cul­tural regime and weather con­di­tions on both tree growth and wood prop­er­ties. Although e- Cam­bium is not a fully oper­a­tional model yet, David hopes that in­dus­try will em­brace such process-mod­els to get the best from Aus­tralia’s tim­ber re­source. ‘For in­dus­try to be will­ing to take this to the next level they need to be con­vinced of its com­mer­cial value,’ says David. “Per­haps one of the great ad­van­tages of these mod­els is they help grow­ers ex­plore the ma­jor im­pli­ca­tions of de­ci­sions so that prob­lems can be avoided in the fu­ture.”

Den­drom­e­ter on Ra­di­ata Pine.

Den­drom­e­ter data col­lect­ing.

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.