Synergies and Trade-Offs in the Production of NWFPs Predicted in Boreal Forests

A icle Syne gies and T ade-O s in he P oduc ion o NWFPs P edic ed in Bo eal Fo es s Mikko Ku ila 1,*, Timo Pukkala 2ID and Ja i Miina 1 1Na u al Resou ces Ins i u e Finland (Luke), Yliopis oka u 6, FI-80100 Joensuu, Finland; ja i[email p o ec ed] 2Facul y o Science and Fo es y, School o Fo es Sciences, Uni e si y o Eas e n Finland, P.O. Box 111, FI-80101 Joensuu, Finland; [email p o ec ed] *Co espondence: [email p o ec ed]; Tel.: +358-29-532-3262 Recei ed: 4 June 2018; Accep ed: 9 July 2018; Published: 11 July 2018   Abs ac : The global g ow h o he consump ion o non-wood o es p oduc s (NWFPs) is e iden due o he cu en ends in li es yle and consump ion. Alongside he inc eased popula i y and comme cial use o NWFPs, hei yields a e also mo e o en aken in o accoun in o es managemen planning. Empi ical yield models ecen ly de eloped o di e en NWFPs enable o es manage s o include hei p edic ed yields in mul i-objec i e op imiza ion. Howe e , knowledge on he syne gies and ade-o s be ween imbe p oduc ion and di e en NWFPs is scan y. In his s udy, we analyzed hese ela ionships h ough co ela ion ma ices and p oduc ion possibili y on ie s in wo case s udy o es holdings om Finland. A la ge numbe o Pa e o op imal o es holding le el plans we e p oduced by mul i-objec i e op imiza ion and used o analyze ade-o s and syne gies. Empi ical yield models o 12 NWFPs, ep esen ing di e en be ies, mush ooms, and ee-based p oduc s, we e u ilized in he analyses. The esul s e ealed syne gies and possibili ies o join -p oduc ion o NWFPs, bu also ade-o s be ween NWFPs and imbe p oduc ion. NWFPs o en had a nega i e co ela ion wi h cu ing emo als, wi h he only excep ion being cowbe y. Despi e he o e all nega i e co ela ion, he maximum yields o NWFPs called o some cu ings. Nega i e co ela ions wi h he ne p esen alue o imbe p oduc ion we e weake . The esul s a e aluable when he aim is o di e si y he use o bo eal o es s and open a enues o uly mul i-objec i e decision suppo se ices o acili a e he decision making o o es owne s. Keywo ds: mul i-objec i e o es managemen ; non-wood o es p oduc s; p oduc ion possibili y on ie 1. In oduc ion Bo eal o es ecosys ems p o ide a b oad ange o di e en angible and in angible goods and se ices wi h di e en uses and economic alues. Fo es s a e o en managed o he e icien p oduc ion o a limi ed numbe o p oduc s. The ob ious jus i ica ion o managing o es s o imbe p oduc ion is ha unc ional ma ke s exis o imbe , and o es owne s ge mone a y bene i s om imbe sales. The lack o ma ke s o he o he p oduc s and di icul ies in measu ing he bene i s ob ained om hem has led o a si ua ion in which many o es goods a e ega ded as side-p oduc s when decisions on o es managemen a e made, which is di ec ly indica ed by he ac ha he alue o hand-picked na u e p oduc s is only abou 1% o he alue o he wood om o es s [ 1 ]. A ypical example o hese kinds o p oduc s is a ious non-wood o es p oduc s (NWFPs) o non- imbe o es p oduc s (NTFPs), such as be ies and mush ooms, ha in bo eal condi ions, a e eely collec ed because o b oad e e yman’s igh s. Sus ainable mul i-p oduc o es managemen is a di e se and complex ma e . In he global con ex , Shackle on e al. [ 2 ] ha e e iewed he his o y and po en ial in ol emen in managing bo h imbe and non- imbe p oduc s om o es s, especially in opical a eas. In Eu ope, consume s’ Fo es s 2018,9, 417; doi:10.3390/ 9070417 www.mdpi.com/jou nal/ o es s Fo es s 2018,9, 417 2 o 15 in e es s owa ds he u iliza ion o NWFPs ha e inc eased [ 3 ], which has led o a emp s o s a managing o es s o he p oduc ion o NWFPs. As a esul , he e ha e been e o s o de elop yield models o di e en NWFPs and link hem in o es simula o s (see, e.g., Ku ila and Tah anainen [ 4 ] o a Eu opean e iew). Ea lie s udies indica e ha aluing NWFPs can inc ease he p o i abili y o s and managemen and modi y he imbe -o ien ed o es managemen o p omo ing NWFPs [ 5 – 10 ]. Fo example, acco ding o Palahíe al. [ 5 ], he addi ional incomes om mush oom picking made he managemen o Pinus syl es is L. and P. nig a A n. s ands in Ca alonia p o i able due o a posi i e e ec o hinnings on mush oom p oduc ion. In Pinus b u ia Ten. s ands in he eas e n Medi e anean a ea, he join -p oduc ion o pine honey and imbe led o longe o a ion pe iods [ 6 ]. In Medi e anean Pinus pinea L. s ands, op imizing he p oduc ion o imbe and edible pine seeds simul aneously esul ed in delayed hinnings [ 8 ]. Tah anainen e al. [ 10 ] obse ed ha he p oduc ion o imbe and ma ke ed mush ooms in Picea abies (L.) Ka s . s ands in eas e n Finland was mos ly in syne gy, wi h mush oom ha es ing c ea ing signi ican addi ional incomes om o es s. The majo i y o he abo e-men ioned s udies ha e been implemen ed a he indi idual s and le el, and only a single NWFP was included in hese calcula ions. An example o mul i-p oduc o es managemen has been gi en by Miina e al. [ 7 ], who op imized he s and managemen o he join p oduc ion o imbe , bilbe y (Vaccinium my illus L.), and cowbe y (V. i is-idaea L.). S and le el calcula ions a e bene icial o inding ou how he s ands could be e icien ly managed o he join p oduc ion o NWFPs and imbe . Howe e , holding le el examina ions a e also needed o e eal s ands whe e he p oduc ion should be di ec ed owa ds NWFPs when he goals o he o es landowne a e maximized unde he p oduc ion possibili ies o he whole o es holding. In holding le el planning, he examina ion s a s om he cu en o es s uc u e o he holding, and hus mo e closely esembles hose decision-making si ua ions ha he o es owne s a e acing. In addi ion, a majo di e ence be ween holding- and s and-le el examina ions is ha he o es plan is ypically p epa ed o a ixed pe iod o one o a ew decades, whe eas he s and-le el calcula ions commonly co e he whole o a ion pe iod. Gam eld e al. [ 11 ] ound ha , in bo eal and empe a e o es s, he e was a ade-o be ween ee biomass p oduc ion and bilbe y p oduc ion. A nega i e ela ionship be ween ee biomass p oduc ion and he p oduc ion o bilbe y was calcula ed o be due o excessi e shading in closed p oduc i e o es s. Gam eld e al. [ 11 ] concluded ha , due o he ade-o s, i would be di icul o maximize all ecosys em se ices simul aneously a he s and scale. Thei analyses we e based on he Swedish na ional o es in en o y da a, and hus hey did no conside he u u e p oduc ion possibili ies o he o es s in Sweden. Pohjanmies e al. [ 12 ] also ound ha s ong con lic s exis be ween imbe p oduc ion and non- imbe bene i s. Con lic s among imbe and NWFPs p oduc ion could be mi iga ed, o example, by modi ying o es managemen in s ands ha a e he mos sui able o NWFPs and applying imbe -p oduc ion-o ien ed managemen in s ands ha do no p oduce NWFPs. In addi ion, Pukkala [ 13 ] used da a en elopmen o analyze he e iciency o he simul aneous p oduc ion o se e al ecosys em se ices in e en-aged and con inuous co e o es y a he holding le el. The NWFPs included in his s udy we e bilbe y and cowbe y yields. I was ound ha inc easing he discoun a e, leading o inc eased cu ings and lowe g owing s ock olumes, dec eased he p oduc ion o o he ecosys em se ices in bo h sil icul u al sys ems, bu he e ec was s onge in e en-aged o es y. This di e ence was mainly explained by he use o clea - ellings in e en-aged o es y, which had a long-las ing nega i e impac on se e al ecosys em se ices. In a p oduc ion sys em like o es holding, he ma ginal a e o ans o ma ion is ypically no cons an h oughou he ange o he holding’s p oduc ion possibili ies. S a ing om a low ini ial le el, a small inc ease in he p oduc ion o ce ain NWFP may ha e no e ec on he p oduc ion o o he goods. When mo e is demanded om he p oduc ion sys em, ade-o s be ween di e en p oduc s s a o eme ge. Fo hese pu poses, he e icien join -p oduc ion possibili ies o gi en ou pu s can be illus a ed and analyzed by using p oduc ion possibili y cu es o on ie s [ 14 , 15 ]. These on ie s indica e he echnically e icien combina ions o he selec ed ou pu s wi h gi en Fo es s 2018,9, 417 3 o 15 inpu s; any poin inside he cu e implies ha he esou ces a ailable a e no ully used o a e being u ilized ine icien ly [ 15 ]. A conca e shape o he p oduc ion possibili y on ie indica es a compe ing ela ionship, and he cu a u e o he cu e indica es he deg ee o compe i i eness be ween he ou pu s, i.e., he ma ginal a e o ans o ma ion [15]. Some o es planning- ela ed s udies ha e used app oaches based on p oduc ion possibili y on ie s. Fo example, To h and McDill [ 16 ] ex ended ou mul iobjec i e p og amming me hods o gene a e spa ially explici o es plan al e na i es and illus a ed he esul s using Pa e o on ie s. In addi ion, Kä kkäinen e al. [ 17 ] applied a p oduc ion possibili y on ie -based app oach when hey e alua ed he e ec s o ene gy wood p oduc ion on imbe p oduc ion po en ial and o es biodi e si y. Bo ges e al. [ 18 ] p esen ed a decision suppo sys em ha included in e ac i e isualiza ion o he Pa e o on ie and e alua ed he ade-o s a a egional le el be ween co k and wood p oduc s. They also s udied he adeo s o di e en ecosys em se ices a a landscape le el wi h mul iple owne ship [19]. Su p isingly li le is known abou he syne gies and ade-o s be ween di e en o es p oduc s. I he join -p oduc ion o a ious NWFPs is conside ed, i is impo an o know i i is possible o achie e syne gies whe e an inc easing p oduc ion o a ce ain NWFP also esul s in an inc easing p oduc ion o some o he NWFPs. Al e na i ely, i is also impo an o know he deg ee o ade-o s be ween NWFPs, i.e., how much an inc ease in he p oduc ion o ce ain NWFPs dec eases he p oduc ion o o he o es p oduc s. The ela ionships be ween imbe p oduc ion and NWFPs a e c i ically impo an o mos o es landowne s and many o hem a e in e es ed in knowing whe he i would be possible o achie e highe incomes om NWFPs han om imbe . The objec i e o his s udy was o examine he ela ionships among a ious NWFPs and imbe p oduc ion a he o es holding scale. The ela ionships we e s udied wi h co ela ion analyses ha we e based on a la ge numbe o e icien o es plans c ea ed o wo o es holdings. In addi ion, he ela ionships we e s udied wi h p oduc ion possibili y on ie s. The posi i e o nega i e co ela ions be ween di e en p oduc s e eal, espec i ely, he syne gis ic o compe i i e ela ion be ween NWFPs and imbe , whe eas he p oduc ion possibili y on ie s gi e in o ma ion on how he ela ionship de elops when he p oduc ion o a gi en NWFP is inc eased wi hin he planning a ea s udied. In his s udy, empi ical yield models we e used o p edic he p oduc ion o he selec ed NWFPs and imbe wi hin wo o es holdings ha ing di e en g owing s ocks and g owing si e condi ions, bu ypical o Finnish o es s. 2. Ma e ials and Me hods 2.1. Desc ip ion o Yield Models o Selec ed NWFPs and Timbe 2.1.1. Bilbe y Shoo s The esh weigh o bilbe y shoo s was p edic ed as a unc ion o si e and s and cha ac e is ics by using a bilbe y co e age model. The ollowing model (wi hou s and age as a p edic o ) was i ed based on he same da a ha we e used by Tu iainen e al. [20]: Co e age =100 ×(1+exp{− (×)})−1(1) (×)=−4.3070 −2.3938 ×OMaT −0.5244 ×OMT −0.4912 ×VT −1.6174 ×CT −1.3006 ×RockySandySoil −1.3240 ×Sp uceMi eI −II −0.7178 ×PineMi eI −III−0.5407 ×PineMi eIV −1.2717 ×PineMi eV +1.0298 ×Poo lyP odLand −0.5150 ×A Regen −1.9340 ×Fo me Ag Land +0.2327 ×Pine −0.5928 ×Bi ch ×(OMaT o OMT)−0.9948 ×Bi ch ×Sp uceMi e +0.0045 ×Al −0.0049 ×Al 2/1000 +0.1237 ×G−0.0023 ×G2 Fo es s 2018,9, 417 4 o 15 whe e Co e age is he mean co e age o bilbe y in he s and (%); Gis he s and basal a ea (m 2 ha −1 ); Pine and Bi ch a e indica o a iables o Sco s pine (Pinus syl es is L.) and deciduous ees as he dominan ee species, espec i ely (o he wise he dominan species is No way sp uce, Picea abies (L.) Ka s .); and Al is he al i ude (m). The e ec o he e ili y o he mine al soil si e is desc ibed by indica o a iables as ollows [ 21 ]: OMaT =Oxalis-Maian hemum ype, OMT =Oxalis-My illus ype, VT =Vaccinium ype, CT =Calluna ype, and Poo lyP odLand = poo ly p oduc i e land and unp oduc i e land (acco ding o [ 22 ]) (o he wise he si e is MT =My illus ype). The pea land si es o sp uce (Sp uceMi e) and pine (PineMi e) mi e a e de ined as ollows [ 23 ]: I = eu ophic, II = he b- ich (meso ophic), III = Vaccinium my illus and all-sedge (meso-oligo ophic), IV = Vaccinium i is-idaea and small-sedge (oligo ophic), and V = co on g ass and dwa -sh ub (poo omb o-oligo ophic bogs). A Regen and Fo me Ag Land a e indica o a iables o a i icial egene a ion me hods (o he wise na u al egene a ion is used) and o me ag icul u al land (o he wise he si e is o me o es ), espec i ely. In a mixed s and, he co e age o bilbe y was p edic ed i s o each ee species using he o al s and basal a ea as a p edic o in he models, and hen he co e age o bilbe y was calcula ed as he weigh ed a e age o he species-speci ic p edic ions. The p opo ions o ee species o s and basal a ea we e used as weigh s. The esh weigh o bilbe y shoo s pe hec a e was calcula ed by assuming ha shoo ha es ing emo es on a e age 36 g d y weigh pe m 2 co e ed by bilbe y [ 24 ] and shoo d y ma e con en is one hi d o esh weigh . 2.1.2. Bilbe ies and Cowbe ies Se e al models o bilbe y and cowbe y yields ha e al eady been published in Finland (see [ 21 ]). Since he p edic ions o di e en models do no always co ela e s ongly [ 25 ] and i is di icul o say which model is he mos eliable, new me a models we e i ed in he cu en s udy o bilbe y and cowbe y yields. Fi e p e ious models o bilbe y [ 26 – 30 ] and i e models o cowbe y [ 26 – 29 , 31 ] we e used o calcula e p edic ions o 12,441 s ands g owing on mine al soils si es. The s ands belonged o ou la ge o es holdings, loca ed in di e en pa s o Finland. The empe a u e sum o he ou holdings anged om 800 d.d. (d.d. s ands o “deg ee days”, which is he accumula ed sum o mean daily empe a u e minus 5 ◦ C on hose days when he mean empe a u e is o e 5 ◦ C) o 1300 d.d. Since each s and had i e di e en p edic ions o bo h be y species, he numbe o obse a ions o i ing he me a models was 62,205 (5 ×12,441) o bo h wild be y species. Since he uni o he “yield” was di e en in di e en models (sco e, p io i y, o yield in kg ha−1), he be y yield p edic ions o each model we e s anda dized by sub ac ing he mean om he p edic ion and di iding he di e ence by he s anda d de ia ion o he p edic ions. A e his s ep, he mean p edic ion was equal o ze o wi h s anda d de ia ion equal o one o all models. Then, he s anda dized p edic ions we e de-s anda dized based on he a e age p edic ion and s anda d de ia ion o he models o Miina e al. [ 30 ] and Tu iainen e al. [ 25 ]. A e his s ep, all yields we e exp essed in kg ha−1. Va iables used as po en ial p edic o s we e si e and g owing s ock a iables ha a e no mally measu ed o assessed in o es in en o ies: o es si e ype, empe a u e sum, mean ee heigh , mean diame e , s and age, s and basal a ea, and p opo ions o di e en ee species o s and basal a ea. The ollowing be y yield models we e i ed using he p edic ions o ea lie be y yield models as obse a ions: Bilbe y =expn1.723 +0.376 ×√D−0.543 ×PBi ch −0.288 ×PSp uce −0.083 ×G/ ln(D+1)+0.000519 ×D×G−0.000263 ×G2 −1.542 ×OMT −0.626 ×VT −1.511 ×CT} R2=0.598 (2) Fo es s 2018,9, 417 5 o 15 Cowbe y =exp1.323 +0.073 ×ln(D+1)+0.166 ×PPine −0.072 ×PSp uce −0.040 ×G/ ln(D+2)+0.509 ×MT +1.601 ×(VT o CT)} R2=0.563 (3) whe e Bilbe y and Cowbe y a e, espec i ely, he yield o bilbe y and cowbe y in he s and (kg ha−1yea −1) ;Dis he mean b eas heigh diame e (cm); and P Bi ch ,P Sp uce , and P Pine a e he p opo ions o b oadlea es, sp uce, and pine o s and basal a ea, espec i ely. The me a models p edic ha bo h bilbe y and cowbe y yields a e he highes on sub-xe ic si es (VT) and in pine-domina ed s ands. A low s and densi y is associa ed wi h high cowbe y yields, bu low bilbe y yields. 2.1.3. Mush ooms The yields o comme cial mush ooms (Mush ooms), Bole us edulis Bull. and Lac a ius spp., we e p edic ed as a unc ion o s and olume inc emen as i has been ound ha he yields o myco hizal ungi co ela e posi i ely wi h he g ow h a e o ees [ 32 ]. The ollowing models we e i ed using he esh weigh (kg ha −1 ) o mush ooms as a dependen a iable and he olume inc emen p edic ed o he g owing s ocks o he sample plo s in he da a se o Tah anainen e al. [ 33 ] as an independen a iable: Mush ooms =exp{0.969 +1.718 ×ln(g o +0.001)−0.144 ×g o}(4) Bole us =exp{−3.624 +5.408 ×ln(g o +0.001)−0.661 ×g o}(5) Lac a ius =exp{1.107 +2.138 ×ln(g o +0.001)−0.220 ×g o}(6) whe e g o = i e-yea s and olume inc emen (m 3 ha −1 ) p edic ed by Monsu o es planning so wa e [34]. 2.1.4. Sp uce Shoo s Sp uce shoo s we e collec ed om young sp uces wi h a heigh o 3–6 m. The esh weigh o sp uce shoo s (g ee−1) was calcula ed using he model o Miina e al. [35]: Sp uceShoo s =exp{4.594 +0.274 ×dbh}(7) whe e dbh = ee diame e a b eas heigh (cm). 2.1.5. Bi ch Lea es Bi ch lea es we e collec ed om sil e bi ches (Be ula pendula Ro h) wi h a heigh o 0.5–6 m. The d y weigh o bi ch lea es (g ee −1 ) was calcula ed using he ollowing biomass model o Repola [36]: Bi chLea es =exp{−29.528 +33.372 ×ds/(ds +2.0)}(8) whe e ds = ee diame e a s ump heigh (cm); ds = 2.0 + 1.25 · dbh. The esh weigh o bi ch lea es was calcula ed by mul iplying he model p edic ion by i e since he d y ma e con en o he lea es was assumed o be app oxima ely 20%. 2.1.6. Pine Ba k Thin pine ba k was peeled o he boles o pines ha we e emo ed in cu ings and ha e a dbh o 8–18 cm. Based on he biomass models o pine [ 37 ], i was assumed ha he esh weigh o ba k peeled o one cubic me e o such pines is 25 kg. Fo es s 2018,9, 417 6 o 15 2.1.7. Resin Sp uce and pine esin was collec ed om ees wi h a dbh o a leas 22 cm and g owing on mo e e ile o ypical mine al and pea land si e ypes o sp uce (My illus ype and sp uce mi e III) and pine (Vaccinium ype and pine mi e IV). The esh weigh o pine and sp uce esin (g ee −1 ) was calcula ed using he models o Kie ä äinen and Salminen [38] as ollows: PineResin =50.837 +13.693 ×dbh (9) Sp uceResin =23.422 +4.364 ×dbh (10) 2.1.8. Bi ch Sap Bi ch sap was collec ed om ma u e sil e bi ches s ands wi h a mean diame e o a leas 20 cm and g owing on My illus ype o mo e e ile si es. The sap model was modi ied om he sap yield equa ions p esen ed by Mahe [ 39 ], assuming ha a bi ch ha ing a dbh o 30 cm yields 75 li e s o sap yea −1 [ 40 ]. The olume o sap (li e s ee −1 ) was calcula ed o sil e bi ches wi h a dbh o a leas 10 cm using he ollowing model: Bi chSap =−35.0 +20.0 ×√dbh (11) 2.2. Case S udy Holdings We used wo case s udy o es holdings in ou analyses. The majo di e ence be ween he case s udy holdings was he si e e ili y and he esul ing dominance o ee species (Table 1). The mo e e ile o es holding was sp uce domina ed and i also included a a he high olume o bi ch. The o he holding was less e ile, he o al olume o ees was clea ly lowe , and he o es s we e pine domina ed. The jus i ica ion o including hese wo holdings o ou analyses was ha di e en NWFPs can be ypically ound in o es s wi h di e en e ili y a es and hey a e also ela ed o di e en ee species. Table 1. The cha ac e is ics o wo case s udy o es holdings. Sp uce-Domina ed Fo es Pine-Domina ed Fo es A ea, ha 210 542 Numbe o s ands 275 415 Tempe a u e sum, d.d. > 5 ◦C 1200 1100 G owing si es (%) He b- ich o be e 41.6 6.3 Mesic 38.2 42.8 Sub-xe ic 16.4 33.2 Poo e han sub-xe ic 3.8 17.8 T ee species ( olume, m3ha−1) Pine 48.6 58.6 Sp uce 92.0 15.8 B oadlea 47.4 24.3 Asso men s ( olume, m3ha−1) Saw log 101.9 24.4 Small log 6.5 8.9 Pulpwood 79.9 66.3 S age o s and de elopmen (%) Seedling & sapling s and 2.1 14.2 Seed & shel e ee s and 8.0 0 Pulpwood-sized o es 11.9 23.2 Saw-log sized hinning o es 41.5 56.5 Ma u e o es 36.5 6.0 All s ands ep esen ed mine al soil si es. Fo es s 2018,9, 417 7 o 15 2.3. Use o he Monsu Planning Sys em The s and in en o y da a was used in he Monsu o es planning so wa e [ 34 ]. The leng h o he planning pe iod applied in ou calcula ions was 30 yea s, and i was di ided in o h ee 10-yea sub-pe iods. The au oma ic e en simula o o Monsu simula ed al e na i e ea men schedules o all s ands o he holding. The simula ions included bo h e en-aged managemen al e na i es and con inuous co e o es managemen al e na i es. In he la e managemen s a egy, he o es s a e managed by di e en kinds o hinnings, mainly om abo e. In e en-aged managemen , he hinning is mo e o en done om below. A he end o o a ion, he o es s a e egene a ed (wi h clea - elling o seed- ee cu ing). Se e al al e na i e managemen schedules we e simula ed o he s ands by changing he iming o cu ings. A no-cu ing schedule was also simula ed o e e y s and. In e en-aged o es y, he egene a ion me hod was clea - elling and plan ing in mesic and be e si es, clea - elling and sowing pine seed in sub-xe ic si es, and na u al egene a ion o pine ia seed ees in xe ic si es. S and dynamics we e simula ed by using he models o Pukkala e al. [ 41 ] o ee su i al, diame e inc emen , and ing ow h (g adual ad ance egene a ion). The asso men olumes o ha es ed ees we e calcula ed wi h he ape models o Laasasenaho [ 42 ]. Quali y deduc ions we e made on saw log olume, based on he models o Meh ä alo [ 43 ] and he esul s o Malinen e al. [ 44 ]. The deduc ion was implemen ed by mo ing a pa o saw log olume in o pulpwood asso men . The ne income om a imbe sale was calcula ed by sub ac ing he ha es ing cos s om he oadside alue o he ha es ed ees. Ha es ing cos s we e calcula ed wi h he models o Rummukainen e al. [45]. S and de elopmen was simula ed in i e-yea s eps, and cu ings and o he ea men s we e simula ed in he middle o each o he h ee 10-yea pe iods, i.e., in yea s 5, 15, and 25. The yields o NWFPs we e calcula ed o he s a ing yea and a he end o each 10-yea pe iod, i.e., in yea s 10, 20, and 30. These yields desc ibe he amoun s o biological p oduc ion o he a ious NWFPs in hese pa icula yea s. The p edic ions can be ega ded as NWFP po en ials. Ac ual ha es ing o some p oduc s (e.g., esin) has an impac on he ees and canno be epea ed e e y yea . Some o he NWFPs, such as be ies and mush ooms, ha e no impac on he ee s ock, and ha es ing can be epea ed e e y yea . The managemen schedules simula ed o he s ands o med a decision space om whe e a high numbe o plans we e gene a ed using mul i-objec i e op imiza ion. The maximized objec i e unc ion was an addi i e u ili y unc ion [ 46 ], which included all he NWFPs analyzed in his s udy, he olume o ha es ed imbe , and he NPV o imbe p oduc ion. NPV was calcula ed om he incomes and cos s o he ea men s simula ed du ing he 30-yea pe iod, and he alue (NPV) o he inal g owing s ock. The NPV o he inal g owing s ock was p edic ed wi h models as explained by Pukkala [ 13 ]. NPV was calcula ed wi h a s anda d eal discoun a e o 3%. The amoun o each p oduc was di ided by he maximum possible amoun o he p oduc . The same was done o NPV. This con e sion emo ed he e ec o he uni s o he p oduc s and NPV, i.e., con e ed he yields and NPVs in o ela i e alues. The maximum possible amoun s we e ound by single-objec i e op imiza ions. The yield o he NWFP in he las simula ion yea (yea 30) was used as he objec i e a iable in op imiza ion. The weigh s o di e en p oduc s in he u ili y unc ion we e d awn om uni o m dis ibu ion and scaled so ha he sum o he weigh s was equal o one. The numbe o andom combina ions o weigh s was 200. Toge he wi h he single-op imum solu ions, he numbe o solu ions used o analyze he co ela ions among NPV, ha es ed olume, and NWFPs was 214. All hese solu ions a e pa e o-op imal when all NWFPs a e assumed o be managemen objec i es in addi ion o ha es ed olume and NPV. The amoun s o di e en ou pu s in he 214 solu ions we e used as he da ase in co ela ion analysis. Then, some example NWFPs we e selec ed o a mo e de ailed analysis h ough p oduc ion possibili y on ie s be ween ha es ed imbe olume and p oduc ion po en ial o he NWFP. The esul s o co ela ion analysis we e used o pick he NWFPs o his analysis: one NWFP co ela ed Fo es s 2018,9, 417 8 o 15 posi i ely wi h ha es ed olume, he o he showed no co ela ion, and he hi d and ou h had a weak o s ong nega i e co ela ion wi h he olume o ha es ed imbe . Then, p oduc ion possibili y on ie s we e also c ea ed o selec ed pai s o NWFPs. These pai s we e also selec ed based on co ela ion coe icien s: posi i e co ela ion, no co ela ion, weak nega i e co ela ion, and s ong nega i e co ela ion. When he p oduc ion possibili y on ie s we e c ea ed be ween wo p oduc s, no o he p oduc s in addi ion o hese wo we e included in he objec i e unc ion. 3. Resul s 3.1. P oduc ion Possibili ies The p oduc ion possibili ies o di e en NWFPs and imbe in sp uce- and pine-domina ed o es holdings a e p esen ed in Table 2. The 30-yea d ain anged om 0 o 66,661 m 3 in he sp uce-domina ed o es and om 0 o 122,261 m 3 in he pine-domina ed o es . The ange o ne p esen alue (3%) was 1,101,936–1,583,157 € in he sp uce-domina ed o es and 2,081,733–2,584,536 € in he pine-domina ed o es . The p edic ed yields o he NWFPs a ied a lo among he 200 al e na i e plans wi h andom weigh s o objec i es. Compa isons o he maximum yields o he 200 al e na i e plans o he single-objec i e maximums e ealed ha he al e na i e plans cap u ed he o al ange o a ia ion in p oduc ion possibili ies well. When conside ing he minimum alues o NWFPs, i mus be no iced ha achie ing a ze o p oduc ion in he biological yields o , e.g., be ies, is p ac ically impossible. Table 2. Ini ial yield and minimum, mean, and maximum yields o non-wood o es p oduc s a he end o he 30-yea pe iod ob ained by ollowing a o al o 200 holding-le el o es plans in which he weigh s o he objec i e a iables we e assigned andomly, o by se ing only a single objec i e. Objec i e Va iable Ini ial Yield 200 Fo es Plans wi h Random Weigh s o Objec i es Single-Objec i e Maximum Minimum Mean Maximum Sp uce-domina ed o es Bilbe y (kg ha−1)14.2 4.0 13.0 20.1 20.9 Cowbe y (kg ha−1)6.3 5.5 6.2 6.9 7.0 Mush oom (kg ha–1)23.2 9.5 21.9 24.0 24.4 Bole us (kg ha–1)7.2 2.4 6.6 7.6 7.8 Lac a ius (kg ha–1)15.3 6.1 14.6 16.0 16.2 Bilbe y shoo (kg ha–1)63.4 21.6 56.6 72.1 74.5 Sp uce shoo (kg ha–1)52.3 9.1 87.7 152.7 155.5 Bi ch lea es (kg ha–1)0.3 0.4 3.0 3.9 3.9 Pine ba k (kg ha–1)68.9 17.5 41.9 51.4 52.0 Pine esin (kg ha–1)15.5 0.3 24.5 43.5 43.7 Sp uce esin (kg ha–1)9.3 0.2 9.4 21.1 21.6 Bi ch sap (li e ha–1)2593.4 389.3 5019.6 7519.0 7653.9 Pine-domina ed o es Bilbe y (kg ha–1)12.2 6.8 15.6 17.4 19.4 Cowbe y (kg ha–1)10.6 9.6 10.2 11.5 11.6 Mush oom (kg ha–1)17.4 9.4 17.6 18.9 19.3 Bole us (kg ha–1)5.1 2.2 4.8 5.4 5.5 Lac a ius (kg ha–1)11.6 6.1 11.8 12.7 13.0 Bilbe y shoo (kg ha–1)62.9 34.7 66.6 77.9 78.2 Sp uce shoo (kg ha–1)33.9 12.0 41.0 66.4 67.3 Bi ch lea es (kg ha–1)4.5 0.4 2.9 3.8 3.8 Pine ba k (kg ha–1)155.9 25.0 145.1 180.6 180.7 Pine esin (kg ha–1)6.1 1.4 50.1 70.6 70.6 Sp uce esin (kg ha–1)1.3 0.0 2.1 4.2 4.3 Bi ch sap (li e ha–1)546.0 1711.9 3707.7 4285.0 4285.0 3.2. Co ela ions among Di e en P oduc s The co ela ions among imbe d ain and yields o NWFPs we e mainly nega i e (Tables 3and 4). This is due o he ac ha many NWFPs a e ei he di ec ly collec ed om ees ( esin, sap, and ba k), a e in a symbio ic ela ionship wi h ees (mush ooms), o a e collec ed indi ec ly depending on he exis ence o ees (be ies). Sp uce shoo s and bi ch lea es a e collec ed om young s ands, and hus he yields o hese NWFPs depend on he age-class s uc u e o holdings. In he sp uce-domina ed o es holding, conside able amoun s o egene a ion cu ings we e done in he beginning o he Fo es s 2018,9, 417 9 o 15 30-yea pe iod in some o he 214 solu ions. As a esul o ha , s ands egene a ed o sp uce we e no mo e sui able o sp uce shoo ha es ing a he end o he 30-yea pe iod and he co ela ion be ween imbe d ain (and NPV) and sp uce shoo s was nega i e. The nega i e co ela ions we e s onge in he pine-domina ed o es han in he sp uce-domina ed o es . Table 3. Co ela ion ma ix among imbe d ain and ne p esen alue (NPV) and yields o non-wood o es p oduc s ob ained by ollowing a o al o 214 holding-le el o es plans in sp uce-domina ed o es holding. NPV Bilbe y Cowbe y Mush oom Bole us Lac a ius Bilbe y Shoo s Sp uce Shoo s Bi ch Lea es Pine Ba k Pine Resin Sp uce Resin Bi ch Sap 0.656 ** −0.607 ** 0.814 ** −0.494 ** −0.536 ** −0.504 ** −0.487 ** 0.014 0.028 −0.384 ** −0.824 ** −0.879 ** −0.732 ** D ain −0.344 ** 0.605 ** 0.231 ** 0.161 * 0.222 ** −0.014 0.071 0.446 ** -0.046 –0.487 ** –0.615 ** –0.393 ** NPV −0.169 * 0.183 ** 0.143 * 0.197 ** 0.905 ** −0.611 ** 0.276 ** –0.044 0.375 ** 0.807 ** 0.240 ** Bilbe y −0.393 ** −0.498 ** −0.394 ** −0.083 −0.184 ** 0.129 –0.573 ** –0.866 ** –0.557 ** –0.620 ** Cowbe y 0.985 ** 0.999 ** 0.357 ** 0.285 ** 0.175 ** 0.452 ** 0.483 ** 0.291 ** 0.413 ** Mush oom 0.983 ** 0.305 ** 0.268 ** 0.127 0.479 ** 0.533 ** 0.325 ** 0.466 ** Bole us 0.369 ** 0.285 ** 0.184 ** 0.448 ** 0.492 ** 0.302 ** 0.423 ** Lac a ius –0.523 ** 0.539 ** –0.038 0.329 ** 0.674 ** 0.210 ** Bilbe y shoo s –0.163 * 0.291 ** 0.167 * –0.388 ** 0.216 ** Sp uce shoo s 0.089 0.098 0.044 0.093 Bi ch lea es 0.498 ** 0.119 0.398 ** Pine ba k 0.570 ** 0.585 ** Pine esin 0.552 ** Sp uce esin Posi i e co ela ions a e displayed in blue ( he ou ones co esponding o co ela ions (0.8, 1), (0.6, 0.8), (0.4, 0.6) and (0.2, 0.4), espec i ely; da ke colo s deno e s onge co ela ions) and nega i e co ela ions in a ed colo ( he ou ones co esponding o co ela ions ( − 0.8, − 1), ( − 0.6, − 0.8), ( − 0.4, − 0.6) and ( − 0.2, − 0.4), espec i ely; da ke colo s deno e s onge co ela ions). S a is ical signi icance is deno ed by as e isks (* p< 0.1; ** p< 0.05). Table 4. Co ela ion ma ix among imbe d ain and ne p esen alue (NPV) and yields o non-wood o es p oduc s ob ained by ollowing a o al o 214 holding-le el o es plans in pine-domina ed o es holding. NPV Bilbe y Cowbe y Mush oom Bole us Lac a ius Bilbe y Shoo s Sp uce Shoo s Bi ch Lea es Pine Ba k Pine Resin Sp uce Resin Bi ch Sap 0.550 ** –0.511 ** 0.888 ** –0.865 ** –0.898 ** –0.868 ** –0.557 ** –0.395 ** 0.108 –0.509 ** –0.875 ** –0.721 ** –0.505 ** D ain –0.231 ** 0.509 ** –0.111 –0.191 ** –0.119 –0.116 –0.222 ** 0.545 ** 0.180 ** –0.587 ** –0.626 ** –0.248 ** NPV –0.138 * 0.522 ** 0.399 ** 0.521 ** 0.881 ** –0.064 0.461 ** 0.028 0.614** 0.557 ** 0.214 ** Bilbe y –0.707 ** –0.816 ** –0.712 ** –0.251 ** –0.336 ** 0.263 ** –0.601 ** –0.712** –0.561 ** –0.364 ** Cowbe y 0.977 ** 1.000 ** 0.649 ** 0.399 ** 0.155* 0.619 ** 0.713** 0.525 ** 0.514 ** Mush oom 0.979 ** 0.551 ** 0.388 ** 0.022 0.646 ** 0.729** 0.553 ** 0.525 ** Bole us 0.650 ** 0.397 ** 0.149* 0.615 ** 0.718** 0.531 ** 0.517 ** Lac a ius 0.071 0.527 ** 0.078 0.642** 0.523 ** 0.331 ** Bilbe y shoo s –0.170 * 0.255 ** 0.362** 0.008 0.315 ** Sp uce shoo s 0.089 0.035 –0.176 ** –0.107 Bi ch lea es 0.208** 0.126 0.312 ** Pine ba k 0.616 ** 0.348 ** Pine esin 0.364 ** Sp uce esin Posi i e co ela ions a e displayed in blue and nega i e co ela ions in a ed colo (da ke colo s deno e s onge co ela ions). S a is ical signi icance is deno ed by as e isks (* p< 0.1; ** p< 0.05). Timbe d ain only co ela ed posi i ely wi h cowbe y and bi ch lea p oduc ion (Tables 3and 4). This is due o he ac ha good cowbe y yields a e ound also a e egene a ion cu ings [ 7 ] and bi ch egene a es na u ally on cu ing a eas [47,48]. The signs and magni udes o co ela ions we e e y simila in bo h sp uce and pine-domina ed o es holdings. Howe e , one clea di e ence be ween holdings was ound: he co ela ion be ween he p oduc ion po en ials o bilbe y (shoo s and be ies) and sp uce shoo s was only s a is ically signi ican and nega i e in sp uce-domina ed o es holding (Tables 3and 4). Bilbe y is abundan on si es which a e usually egene a ed o sp uce. Regene a ion cu ings a e ha m ul o bilbe y, bu a he same ime, egene a ion o sp uce will inc ease he p oduc ion o sp uce shoo s. 3.3. P oduc ion Possibili y F on ie s The imbe d ain and cowbe y yield had a s ong posi i e co ela ion (Tables 3and 4). The shape o he p oduc ion possibili y on ie e eals ha when only hese wo p oduc s we e conside ed,