The Long-Term Consequences of Forest Fires on the Carbon Fluxes of a Tropical Forest in Africa

applied sciences

Article The Long-Term Consequences of Forest Fires on the Carbon Fluxes of a Tropical Forest in Africa

Rico Fischer

Helmholtz Centre for Environmental Research-UFZ Leipzig, Department of Ecological Modelling, Permoserstr. 15, 04318 Leipzig, Germany; rico.ﬁ[email protected]; Tel.: +49-341-235-482-267

Abstract: Tropical forests are an important component of the global carbon cycle, as they store large amounts of carbon. In some tropical regions, the forests are increasingly influenced by disturbances such as fires, which lead to structural changes but also alter species composition, forest succession, and carbon balance. However, the long-term consequences on forest functioning are difficult to assess. The majority of all global forest fires are found in Africa. In this study, a forest model was extended by a fire model to investigate the long-term effects of forest fires on biomass, carbon fluxes, and species composition of tropical forests at Mt. Kilimanjaro (Tanzania). According to this modeling study, forest biomass was reduced by 46% by fires and even by 80% when fires reoccur. Forest regeneration lasted more than 100 years to recover to pre-fire state. Productivity and respiration were up to 4 times higher after the fire than before the fire, which was mainly due to pioneer species in the regeneration phase. Considering the full carbon balance of the regrowing forest, it takes more than 150 years to compensate for the carbon emissions caused by the forest fire. However, functional diversity increases after a fire, as fire-tolerant tree species and pioneer species dominate a fire-affected forest area and thus alter the forest succession. This study shows that forest models can be suitable tools to simulate the dynamics of tropical forests and to assess the long-term consequences of fires. Keywords: tropical forest; fire; forest model; Africa; carbon Citation: Fischer, R. The Long-Term Consequences of Forest Fires on the Carbon Fluxes of a Tropical Forest in Africa. Appl. Sci. 2021, 11, 4696. https://doi.org/10.3390/app11104696 1. Introduction Tropical forests are important for the climate of the Earth because they represent a Academic Editor: Rubén Díaz-Sierra relevant carbon sink [1–3]. Tropical forests account for about 50% of the total carbon stored in global vegetation [4,5]. However, these forests are threatened by various environmental Received: 4 May 2021 and anthropogenic hazards such as logging and fire [6–9]. Between 2000 and 2012, tropical Accepted: 19 May 2021 rainforests accounted for 32% of global forest area loss [10]. Extreme weather events Published: 20 May 2021 and agricultural practices have already intensified forest fire regimes, which have led to accelerated losses of tropical forest [11,12]. Global fire carbon emissions were 2.0 Gt C per Publisher’s Note: MDPI stays neutral year (=109 t of carbon), with the majority of carbon emissions occurring in savannas and with regard to jurisdictional claims in tropical forests [13]. Annual carbon emissions from tropical deforestation and degradation published maps and institutional affil- were on average 0.5 Gt C, whereby it is debated whether these carbon emissions from fires iations. may not be compensated by forest regrowth after the fire [13]. It is also assumed that the number of fires in tropical forests will continue to increase in the future [14]. Almost 70% of all global forest fires are found in Africa [15,16]. These fires are rarely caused by natural processes such as lightning strikes; instead, human actions are often Copyright: © 2021 by the author. the cause [17]. This can happen through conversion to agricultural land but also through Licensee MDPI, Basel, Switzerland. poachers or gatherers [17,18]. Forest fires can have an impact on the species composition, This article is an open access article the living biomass, and the carbon balance of tropical forests [19]. distributed under the terms and These fires are also an influential ecological factor in the forests around Mt. Kilimanjaro conditions of the Creative Commons in Africa [15,18,20]. In the last century, the impact of fire on the vegetation has increased Attribution (CC BY) license (https:// due to climatic changes as well as the consequences of the growing population in the creativecommons.org/licenses/by/ Kilimanjaro area. The upper tree line has already decreased by up to 800 m in altitude [15]. 4.0/).

Appl. Sci. 2021, 11, 4696. https://doi.org/10.3390/app11104696 https://www.mdpi.com/journal/applsci Appl. Sci. 2021, 11, 4696 2 of 17

Consequently, about 15% of the forest area at the Kilimanjaro region has been destroyed in the last 40 years [15], with negative consequences for the water supply of lowland farmland. However, the long-term consequences of fires on tropical forest dynamics are difficult to assess, as forest fires create heterogeneous landscapes with long-lasting impacts. A fire regime can be described by characteristics such as frequency, size, and severity. All these factors and also their interactions form a complex dynamic system. The long tradition in fire ecology has produced a broad understanding of these processes. Numerous models have been developed to simulate damages caused by fires in forests. These fire models are mainly developed as extensions of vegetation models [21–23]. Many fire models simulate the spread of fire as a function of topographic features. They were mainly developed for geoinformation systems and work on a scale of several kilometers [24]. Dynamic Global Vegetation Models (DGVM) have also been extended to include fire models [9,21]. Since these types of models mostly operate at a coarse scale of several kilometers, they use a top-down approach. Instead, high-resolution forest models operate at a much finer spatial resolution of up to 10–20 m [25,26]. In order to understand the effects of fire events on forest structure and dynamics, it is therefore necessary to develop an adapted fire model for this fine scale. For the analysis of the effects of fire events on the dynamics of tropical forests at Mt. Kilimanjaro, the forest model FORMIND [27,28] was extended by a fire model. Specifically, the following question was answered: How strong is the influence of forest fires on the aboveground biomass, productivity, and carbon balance of sub-montane tropical forests at Mt. Kilimanjaro? To answer this question, the forest model FORMIND was used to simulate typical fire patterns for this region and to look at long-term forest dynamics. For this purpose, first the forest dynamics after a fire event were investigated, i.e., the long-term regeneration of forest attributes such as biomass after the fire. In a second step, forest dynamics influenced by permanent recurrent forest fires were examined. To carry out this research, a new fire model was developed and integrated in the forest model.

2. Materials and Methods 2.1. Study Region Mt. Kilimanjaro The tropical sub-montane forest area examined in this study is located within the forest belt at the base of Mt. Kilimanjaro (S 3.260150◦, E 37.417458◦; Figure1). The climate is wet tropical with annual precipitation reaching 2.700 mm [29]. In this area, several forest research sites were established. Five forest sites (between 50 × 50 m and 100 × 100 m; in total 2 ha) were distributed along the southeastern slope of Mt. Kilimanjaro in the sub- montane and lower montane “Newtonia” rainforest (∼1150–2050 m a.s.l.). The dominant tree species are Heinsenia diervilleoides, Strombosia schefﬂera, Entandophrangma excelsum, and Garcinia tansaniensis. Forest inventories were conducted in each site in 2012 (by the German Research Foundation DFG within the Research Unit FOR1246) in cooperation with the Kilimanjaro National Park [30,31]. For this forest, a parameterization for the FORMIND forest model had already been developed and initial investigations about the carbon balance under natural conditions had been carried out [32,33].

2.2. The Forest Model FORMIND FORMIND is an individual- and process-based forest model designed to simulate the dynamics of tropical forests [27,28]. For each tree in a stand, the processes of growth, mortality, and competition were calculated. Tree growth depends on a biomass balance, taking photosynthesis and respiration into account. At each time step (here: yearly), the biomass increment of each tree is calculated. Photosynthesis is derived on the basis of incoming radiation and shading by other trees. This forest model also includes a carbon module, which allows estimation of the full carbon balance of a forest stand, including the carbon ﬂuxes in the soil and in the deadwood [34]. Hence, it is possible to determine the net ecosystem exchange (NEE), which is the difference between carbon uptake (i.e., Appl. Sci. 2021, 11, 4696 3 of 17

Appl. Sci. 2021, 11, x FOR PEER REVIEWphotosynthesis) and carbon release (respiration and decomposition). The FORMIND model3 of 18 has already been used to study the carbon ﬂuxes of tropical forests [34,35], and it is also able to simulate the effects of logging [36–38], but the simulation of forest ﬁres is still missing.

Figure 1. Map showing the location of the study region atat Mt.Mt. Kilimanjaro in Tanzania,Tanzania, Africa.Africa. TheThe red dotdot representsrepresents the the location location of of the the study study site; site the; the red red rectangle rectangle indicates indicates the areathe investigatedarea investigated with thewith satellite the satellite data (MODISdata (MODIS burned burned area, seearea, Section see Section 2.5). Photo: 2.5). Photo: Stephan Stephan Getzin. Getzin. A complete parameterization has already been created for the sub-montane forests at 2.2. The Forest Model FORMIND Mt. Kilimanjaro [32,33]. In addition, all occurring tree species have already been classified into sixFORMIND plant functional is an individual groups [-32 and] (for process an overview-based forest of the model grouping, designed see Section to simulate 2.5 and the Appendixdynamics Aof). tropical In this study,forests no [27,28]. changes For were each madetree in to a thestand established, the processes parameterization. of growth, mortality, and competition were calculated. Tree growth depends on a biomass balance, tak- 2.3.ing photosynthesis Review of Fire Models and respiration into account. At each time step (here: yearly), the bio- massHere, increment three establishedof each tree fireis calculated. models are Photosynthesis reviewed, and theis derived way in whichon the theirbasis respective of incom- advantagesing radiation can and be shading used for by a other new fire trees. model Thisin forest FORMIND model also are discussed.includes a carbon module, which allows estimation of the full carbon balance of a forest stand, including the carbon 2.3.1.fluxes Fire in the Model soil and by Drossel in the deadwood and Schwabl [34]. Hence, it is possible to determine the net eco- systemThe exchange Drossel–Schwabl (NEE), which fire model is the is difference based on a between cellular automaton carbon uptake [22,39 (i.e.]. A, cellphotosyn- can be empty,thesis) coveredand carbon with release vegetation, (respiration or burning. and decomposition). The status of a cell The change FORMIND is rule-based: model has (1) An al- emptyready been cell is used populated to study with the vegetationcarbon fluxes (here: of tropical tree) with forests probability [34,35] p, and in the it is next also step; able (2) to asimulate cell with the vegetation effects of starts logging to burn[36–38 with], but the the inflammation simulation of rate forest f; (3) fires a cell is withstill missing. vegetation startsA to complete burn if at parameterization least one neighboring has already cell is burning; been created (4) a burningfor the sub cell-montane becomes emptyforests inat theMt. nextKilimanjaro step. [32,33]. In addition, all occurring tree species have already been classified intoThe six main plant advantage functional of thisgroups fire model[32] (for is an the overview adequate of modelling the grouping of the, see spread Section of fire,2.5. startingand Appendix from the A) ignition. In this sourcestudy, (Figureno changes2). However, were made the to dynamical the established spread parameteriza- of fire events doestion. not play a major role for this study. The disadvantage is the high computational effort (daily time step) and that each burning cell has completely no vegetation after the fire. In addition,2.3. Review it of is Fire not Models applicable for the individual-based forest model used, as it aggregates the complexHere, three forest established structure intofire models binary informationare reviewed of, and forest the or way non-forest. in which their respective advantages can be used for a new fire model in FORMIND are discussed. 2.3.2. Fire Model by Green 2.3.1.The Fire landscape Model by modelDrossel MOSAIC and Schwabl was extended by a fire model [40]. This model is also based on a cellular automaton, where a cell consists of a single plant. Since fire events are The Drossel–Schwabl fire model is based on a cellular automaton [22,39]. A cell can rare events, they are described in this model with a Poisson process. If a fire breaks out, a be empty, covered with vegetation, or burning. The status of a cell change is rule-based: randomly selected rectangular area burns (Figure3). The size of the fire area is modelled as (1) An empty cell is populated with vegetation (here: tree) with probability p in the next exponentially distributed. Once an area has burned down, it is immediately repopulated. step; (2) a cell with vegetation starts to burn with the inflammation rate f; (3) a cell with vegetation starts to burn if at least one neighboring cell is burning; (4) a burning cell becomes empty in the next step. The main advantage of this fire model is the adequate modelling of the spread of fire, starting from the ignition source (Figure 2). However, the dynamical spread of fire events does not play a major role for this study. The disadvantage is the high computational effort (daily time step) and that each burning cell has completely no vegetation after the fire. Appl. Sci. 2021, 11, x FOR PEER REVIEW 4 of 18

In addition, it is not applicable for the individual-based forest model used, as it aggregates the complex forest structure into binary information of forest or non-forest.

Appl. Sci. 2021, 11, x FOR PEER REVIEW 4 of 18

Appl. Sci. 2021, 11, 4696 4 of 17 In addition, it is not applicable for the individual-based forest model used, as it aggregates the complex forest structure into binary information of forest or non -forest. Figure 2. Fire model according to Drossel–Schwabl. On the left is a representation of the cellular automaton, consisting of vegetation (green) and empty cells (white). On the right is the spread of a fire event (red). In this schematic visualization of a virtual landscape, one cell has the size of 20 m, and the total area is 30 × 30 cells.

2.3.2. Fire Model by Green The landscape model MOSAIC was extended by a fire model [40]. This model is also based on a cellular automaton, where a cell consists of a single plant. Since fire events are rare events, they are described in this model with a Poisson process. If a fire breaks out, a Figure 2.2. Fire model according to Drossel–Schwabl.Drossel–Schwabl. On the left is a representationrepresentation ofof thethe cellularcellular randomly selected rectangular area burns (Figure 3). The size of the fire area is modelled automaton, consisting of vegetation (green) and emptyempty cellscells (white).(white). On the right is thethe spreadspread ofof aa as exponentially distributed. Once an area has burned down, it is immediately repopu- ﬁrefire eventevent (red). In this schematic visualization of a virtual landscape, one cell has the size of 2 200 m, lated. and the total area isis 3030 × 3030 cells. cells.

FigureFigure 3.3. FireFire modelmodel accordingaccording toto Green.Green. OnOn thethe leftleft isis thethe visualizationvisualization ofof thethe cellularcellular automaton.automaton. InIn aa cell, the the biomass biomass of of the the vegetation vegetation is isshown shown—the—the darker darker the the cell cell,, the themore more biomass. biomass. On the On right the rightis a random is a random fire area ﬁre area(red). (red). In this In thisarea area all trees all trees burn burn and and die. die. The The burned burned area area is immediately is immediately repopulated, starting with a low biomass value. In this schematic visualization of a virtual landscape, repopulated, starting with a low biomass value. In this schematic visualization of a virtual landscape, one cell corresponds to a single plant (e.g., tree), and the total area is 30 × 30 cells. one cell corresponds to a single plant (e.g., tree), and the total area is 30 × 30 cells.

TheThe strengthstrength ofof thisthis modelmodel isis thethe modellingmodelling ofof temporaltemporal dynamicsdynamics betweenbetween twotwo firefire events and the spatial variability in the fire affected area. The disadvantage is that all trees events and the spatial variability in the fire affected area. The disadvantage is that all Appl. Sci. 2021, 11, x FOR PEER REVIEWtreesin the in fire the area fire areaalways always burn burn and anddie. die.In an In extension an extension of the of model, the model, this thisweakness weakness5 wasof 18 Figure 3. Fire model according to Green. On the left is the visualization of the cellular automaton. waspartially partially corrected corrected by allowing by allowing only only pioneer pioneer species species to burn, to burn, while while climax climax species species sur- Invived a cell, [40]. the biomass of the vegetation is shown—the darker the cell, the more biomass. On the right survivedis a random [40 fire]. area (red). In this area all trees burn and die. The burned area is immediately re-

populated,2.3.3. Fire startingModel bywith Busing a low biomass value. In this schematic visualization of a virtual landscape, 2.3.3.one cell Fire corresponds Model by to Busing a single plant (e.g., tree), and the total area is 30 × 30 cells. ToTo investigate the the effects effects of of fire fire on on the the forests forests of the of theUSA, USA, the forest the forest model model FORCLIM FOR- CLIM[41,42]The [ 41was strength,42 extended] was of extended this by modela fire by model a is fire the model [43,44].modelling [43 The,44 of]. influence Thetemporal influence of dynamics fire of on fire vegetation between on vegetation twois mod- fire is modelledeventselled in and such in the such a spatial way a way that varia that firebility fire-tolerant-tolerant in the treefire tree affectedspecies species harea. haveave Thea agreater greater disadvantage probability probability is that of survivalsurvivalall trees thaninthan the fire-intolerant fire fire-intolerant area always species, species burn and, andand larger largerdie. trees In trees an survive extension survive more moreof than the than smaller model, smaller trees. this weaknesstrees. The frequency The wasfre- ofpartiallyquency fire events of corrected fire but events also by but the allowing severityalso the only severity of a pioneer fire isof realizeda speciesfire is randomlyrealized to burn, randomly (Figurewhile climax4 ).(Figure species 4). sur- vived [40].

FigureFigure 4.4. Fire model after after Busing. Busing. On On the the left left is is the the representation representation of of the the cellular cellular automaton automaton with with no nofire. fire. In a In cell, a cell, the the biomass biomass of the of the vegetation vegetation is shown is shown—the—the darker darker the the cell cell,, thethe more more biomass. biomass. On Onthe right, fire has burned the entire area, but depending on fire tolerance, some trees survive. In this the right, fire has burned the entire area, but depending on fire tolerance, some trees survive. In schematic visualization of a virtual landscape, one cell has the size of 20 m, and the total area is 30 this schematic visualization of a virtual landscape, one cell has the size of 20 m, and the total area is × 30 cells. 30 × 30 cells. The disadvantage of this fire model is that there is always fire in the entire forest area. However, not all trees die, as this depends on the fire tolerance of the species and the size of the tree.

2.4. The New Fire Model ForFire For the simulation of fire events with the individual-based forest model FORMIND, a separate fire model ForFire (Forest Fire model) was developed (Figure 5), which combines the advantages of the abovementioned fire models. The detailed dynamics of the fire spread are not relevant for this study, as the time step of the forest model is one year. For the fire model ForFire, it is important that the appearance of forest fires, the total spread of the fire area, and the generation of heterogeneous landscapes can be represented.

Figure 5. Schematic representation of the fire model ForFire. This fire model was integrated in the forest model FORMIND. In each time step (here: 1 year) the random number of fires in this time step is determined. Then, an exponential distribution is used to specify the size of the forest fire. To Appl. Sci. 2021, 11, x FOR PEER REVIEW 5 of 18

2.3.3. Fire Model by Busing To investigate the effects of fire on the forests of the USA, the forest model FORCLIM [41,42] was extended by a fire model [43,44]. The influence of fire on vegetation is modelled in such a way that fire-tolerant tree species have a greater probability of survival than fire-intolerant species, and larger trees survive more than smaller trees. The frequency of fire events but also the severity of a fire is realized randomly (Figure 4).

Figure 4. Fire model after Busing. On the left is the representation of the cellular automaton with no fire. In a cell, the biomass of the vegetation is shown—the darker the cell, the more biomass. On the right, fire has burned the entire area, but depending on fire tolerance, some trees survive. In this schematic visualization of a virtual landscape, one cell has the size of 20 m, and the total area is 30 Appl. Sci. 2021, 11, 4696 × 30 cells. 5 of 17

The disadvantage of this fire model is that there is always fire in the entire forest area. However, not all trees die, as this depends on the fire tolerance of the species and the size of theThe tree. disadvantage of this fire model is that there is always fire in the entire forest area. However, not all trees die, as this depends on the fire tolerance of the species and the size of2.4. the The tree. New Fire Model ForFire 2.4. TheFor New the Firesimulation Model ForFire of fire events with the individual-based forest model FORMIND, a separate fire model ForFire (Forest Fire model) was developed (Figure 5), which com- For the simulation of fire events with the individual-based forest model FORMIND, a bines the advantages of the abovementioned fire models. The detailed dynamics of the separate fire model ForFire (Forest Fire model) was developed (Figure5), which combines fire spread are not relevant for this study, as the time step of the forest model is one year. the advantages of the abovementioned fire models. The detailed dynamics of the fire For the fire model ForFire, it is important that the appearance of forest fires, the total spread are not relevant for this study, as the time step of the forest model is one year. For thespread fire modelof the ForFire,fire area it, isand important the generation that the appearance of heterogeneous of forest landscapes fires, the total can spread be repre- of thesented. fire area, and the generation of heterogeneous landscapes can be represented.

FigureFigure 5.5.Schematic Schematic representationrepresentation ofof thethe firefire model model ForFire. ForFire. ThisThis firefire model model was was integrated integrated inin the the forestforest modelmodel FORMIND.FORMIND. InIn eacheach timetime stepstep (here:(here: 11 year)year) thethe randomrandom numbernumber ofof firesfires inin thisthis time time step is determined. Then, an exponential distribution is used to specify the size of the forest fire. To step is determined. Then, an exponential distribution is used to specify the size of the forest fire. To ensure that the fire can break out, each sub-area (20 m plot) of the total simulation area is checked for whether there is enough biomass and for whether the top layer of soil is not too wet. After that, the type of fire is determined randomly, either a strong crown fire or a lighter ground fire. However, not all trees in the fire area burn and die. The probability of a tree dying from fire depends on the strength of the fire, the size of the tree, and the fire tolerance depending on the species.

First, a stochastic process is introduced that counts the number of ﬁres in a year (F in yr−1). This number of annual forest ﬁres is assumed to be Poisson distributed:

1 F(t) ∼ Poisson (1) λ

where λ (yr) is the mean time interval between two fire events. If a fire breaks out in one year, the fire center is randomly determined on the entire simulated forest area. The size of the fire area (S in % of total forest area) is realized via an exponential distribution:

1 S(t) ∼ Exponential (2) β

where β (%) is the relative size of the mean fire area with respect to the total simulated forest area. The area of the forest fire is chosen in such a way that, from a randomly chosen fire Appl. Sci. 2021, 11, x FOR PEER REVIEW 6 of 18

ensure that the fire can break out, each sub-area (20 m plot) of the total simulation area is checked for whether there is enough biomass and for whether the top layer of soil is not too wet. After that, the type of fire is determined randomly, either a strong crown fire or a lighter ground fire. However, not all trees in the fire area burn and die. The probability of a tree dying from fire depends on the strength of the fire, the size of the tree, and the fire tolerance depending on the species.

First, a stochastic process is introduced that counts the number of fires in a year (F in yr−1). This number of annual forest fires is assumed to be Poisson distributed: 1 () ∼ (1) where λ (yr) is the mean time interval between two fire events. If a fire breaks out in one year, the fire center is randomly determined on the entire simulated forest area. The size of the fire area (S in % of total forest area) is realized via an exponential distribution: 1 () ∼ (2) Appl. Sci. 2021, 11, 4696 6 of 17 where β (%) is the relative size of the mean fire area with respect to the total simulated forest area. The area of the forest fire is chosen in such a way that, from a randomly chosen firecenter, center, fire fire breaks breaks out onout randomly on randomly neighboring neighboring areas. areas. This createsThis creates an irregular an irregular contiguous con- tiguousarea around area around the fire the center. fire center. However,However, fire fire can can only only break out onon aa forestforest plotplot if if there there is is enough enough flammable flammable material. material.If there If there is less is less than than 200 200 g/m g/m2 of2 of biomass biomass in in the the area, area, the the possibility possibility for for a a fire fire to to spread spread is isalmost almost non-existent non-existent [45 [45].]. In In addition, addition, the the possibility possibility of a of fire a breakingfire breaking out is out also is dependent also de- pendenton climatic on climatic conditions, conditions, such as soilsuch moisture, as soil moisture cf. SPITFIRE, cf. SPITFIRE in LPJ [46 in]. ThisLPJ [ is46]. integrated This is integratedin the fire in model the fire ForFire model byForFire only by starting only starting a fire with a fire a with certain a certain probability probability B on aB ploton awithin plot within the fire the area, fire area, depending depending on the on soil the water soil water content content (θ in ( Vol%)θ in Vol%) in the in topsoil the topsoil layer. layer.For this For purpose, this purpose, the relative the relative water water saturation saturation m (%) m is(%) calculated, is calculat whiched, which indicates indicates how howwater-saturated water-saturated the topsoilthe topsoil layer layer is [21 is]: [21]: () = − (3) m(θ) = θ − MSW/Por− −MSW (3) where MSW and Por are parameters of the soil [47]. For the ignition probability B (Figure 6)where we get:MSW and Por are parameters of the soil [47]. For the ignition probability B (Figure6 ) we get: − (m m ) ()B (=m )= e π /+e 0+.20.2 (4 (4))

wherewhere mme e(%)(%) is is the the saturation saturation of of the the upper upper soil soil layer layer with with water, water, above above which which it it is is very very unlikelyunlikely that that a afire ﬁre will will start start [21]. [21].

FigureFigure 6. 6. ModelledModelled probability probability of of a afire fire outbreak. outbreak. Probability Probability B B that that a afire fire will will spread spread on on a aforest forest plot plot inin the the fire fire area, area, depending depending on on the the water water saturation saturation m m in in the the topsoil topsoil layer layer ( (herehere mme e== 0.4 0.4=40% = 40%).).

After determining the forest area in which the fire occurs, the next step considers all individual trees in the fire area. Every tree in the fire area is potentially at risk of burning and dying. The probability that a tree will burn depends on its species-specific fire tolerance and its trunk diameter [43]. Fire tolerance can be divided into four groups: Species with fire tolerance group 1 burn in any fire, whereas fire tolerance group 4 is very resistant to fire. The probability PFire that a tree with a stem diameter DBH [cm] will burn in a fire event is calculated according to [43] as follows, depending on the fire tolerance group (Figure7):

1. PFire (DBH) = 1 ((−(1− f sev)·0.00202)−0.00053) · DBH 2. PFire (DBH) = e ((−(1− f sev)·0.02745)−0.00255) · DBH 3. PFire (DBH) = e −0.00053·DBH 4. PFire (DBH) = e − 0.5 − (1 − fsev) · 0.5

where fsev (value 0–1) is an indicator of the severity and type of fire event. A ground fire (fsev ~ 0.2) is not as dangerous for the trees as a crown fire (fsev ~ 0.7). These probabilities were originally developed for the forest model CLIMACS, applied to forests in North America [48]. A distinction is made between tree species that always burn regardless of their size and the intensity of the fire event (fire tolerance group 1) and between tree species that have a certain probability of surviving a fire event if the trees are larger/older (fire tolerance group 2–4). Appl. Sci. 2021, 11, x FOR PEER REVIEW 7 of 18

After determining the forest area in which the fire occurs, the next step considers all individual trees in the fire area. Every tree in the fire area is potentially at risk of burning and dying. The probability that a tree will burn depends on its species-specific fire tolerance and its trunk diameter [43]. Fire tolerance can be divided into four groups: Species with fire tolerance group 1 burn in any fire, whereas fire tolerance group 4 is very resistant to fire. The probability PFire that a tree with a stem diameter DBH [cm] will burn in a fire event is calculated according to [43] as follows, depending on the fire tolerance group (Figure 7):

1. () = 1 (()⋅.). ⋅ 2. () = (()⋅.). ⋅ 3. () = .⋅ 4. () = − 0.5 − (1 − ) ⋅ 0.5

where fsev (value 0–1) is an indicator of the severity and type of fire event. A ground fire (fsev ~ 0.2) is not as dangerous for the trees as a crown fire (fsev ~ 0.7). These probabilities were originally developed for the forest model CLIMACS, applied to forests in North America [48]. A distinction is made between tree species that always burn regardless of Appl. Sci. 2021, 11, 4696 their size and the intensity of the fire event (fire tolerance group 1) and between tree7 spe- of 17 cies that have a certain probability of surviving a fire event if the trees are larger/older (fire tolerance group 2–4).

Figure 7. Probability of death of a tree in a fire fire event event,, after [43].43]. This probability depends o onn the fire fire tolerance of of the the tree species, the stem diameter in breast height (DBH) (DBH),, and the severity of the firefire (fsev). (a) The probability of death in a weak fire (fsev = 0.2). (b) The probability of death in a stronger (fsev). (a) The probability of death in a weak fire (fsev = 0.2). (b) The probability of death in a stronger fire (fsev = 0.7). fire (fsev = 0.7).

The described firefire modelmodel ForFireForFire combines combines the the ideas ideas of of various various established established fire fire models. models.It has It thehas advantagesthe advantages that thethat firethe area fire changesarea changes spatially, spatially, that fire that only fire breaks only breaks out under out undergiven vegetationgiven vegetation conditions, conditions and that, and trees that only trees die only with die a certainwith a certain probability, probability, depending de- pendingon a species-specific on a species-specific tolerance tolerance and the and tree the age tree (or age stem (or diameter).stem diameter). The influenceThe influence of a oflarge a large forest forest fire onfire forest on forest dynamics dynamics and and carbon carbon balance balance can can be investigated, be investigated, as well as well as the as thelong-term long-term consequences consequences of recurrent of recurrent fire fire events. events.

2.5. Simulation Settings To analyze the effects of a firefire event on forest dynamics, a forest area of nine hectares was simulated over a period of 500 years with a yearly time step. The fire fire model has four parameters: the the mean mean fire fire frequency, frequency, the mean fire fire size, the mean fire fire intensi intensity,ty, and the firefire tolerance of the individual species groups. To To determine fire fire frequency in the study area, the satellite datasets “ “MODISMODIS Burned Area Product MCD45” from 2000 to 2012 were analyzedanalyzed.. ( (InIn total 144 images werewere analyzed;analyzed;see see https://modis-fire.umd.edu/index.html https://modis-fire.umd.edu/index.html,, accessed on 12 Febru Februaryary 2021, for more information about MODIS Burned Area Product Product;; for user manualuser and downloadingmanual see and https://modis-fire.umd.edu/files/MODIS_Burned_ downloading see https://modis- Area_Collection51_User_Guide_3.1.0.pdf, accessed on 12 February 2021). These maps show the monthly burned area in an area through spectral variations and changes in vegetation (see, for example, [49]). In the savannah area of the Kilimanjaro region, almost every year a fire can be observed; in the tropical forests of this region, a fire can be observed somewhat less frequently [15]. The analysis of the available satellite data for the entire Kilimanjaro area (region investigated: 2◦29–3◦72 South and 37◦12–37◦57 East) showed that a large-scale fire occurs on average every three to four years and is between 1 and 16 pixels in size (pixel resolution 500 m) [50]. The larger fires were observed in 2001, 2004, 2007, and 2009, and they mostly occur in the dry seasons, which are characterized by the El Nino Southern Oscillation [20]. In the forested area at Mt. Kilimanjaro, the satellite data show that on average a fire is about 25 hectares in size (=1 image pixel, with the size of 500 × 500 m). However, smaller fire areas cannot be detected due to the spatial resolution (i.e., 500 m) of the satellite product, and thus this mean size seems too high. Since the simulated forest area in the model has a size of 9 hectares, the mean fire area was set to 20% of the considered forest area—which corresponds to experiences from other areas [7,40,51]. Fire intensity indicates the difference between a light ground fire (fire intensity = 0.1) and a strong crown fire (fire intensity = 1.0). For this study, a mean value of 0.55 was set [43,48]. All parameter values for the fire model ForFire can be found in Table1. Appl. Sci. 2021, 11, 4696 8 of 17

Table 1. Parameter of the fire model ForFire and parameter values for the study region for the scenario where repeated fire events are simulated. For the scenario with only one fire event, the parameters are S = 60% and fsev = 0.67).

Parameter Description Parameter Value for This Study Fire frequency F mean time between two fire events 4 years average size of a fire (here: 2 ha) in Fire size S relation to the total size of the 20% simulated forest (here: 9ha) indicator (0–1) for the strength and Fire severity fsev type of fire (light ground fire or 0.55 strong crown fire)

The mortality of a tree in a fire event is influenced by its fire tolerance. There are four groups of fire tolerance in the fire model: Species with fire tolerance group 1 burn in any fire, whereas fire tolerance group 4 is very resistant to fire. For each tree species in the study area, the fire tolerance was determined by the literature or expert knowledge (for a complete list, see AppendixA). The thickness of the bark is the most important factor [ 52]. The thicker a bark is, the less heat is generated in the cambium layer of the tree (responsible for the growth of the tree). In the study area, for example, the species Agarstia salicifolia and Morella salicifolia have particularly thick bark and are therefore considered to be fire tolerant [15,53,54]. The species Alagium chinense is a deciduous species, which is also considered to be adapted to the occurrence of fire [53,55]. The three fire-tolerant species mentioned belong to plant functional type 4. This species group was thus classified as fire tolerant (i.e., fire tolerance 4). The species Ilex mitis, on the other hand, is considered fire intolerant, as all trees of this species died in a fire experiment [56]. This species belongs to plant functional type 2, which was thus classified as fire intolerant (i.e., fire tolerance 1). Fire tolerance was determined for each plant functional type, depending on the fire tolerance of the individual tree species in the corresponding species group (Table2, AppendixA).

Table 2. Species grouping with ﬁre tolerance at Mt. Kilimanjaro region. Grouping of all tree species in the study area into six species groups according to maximum height and shade tolerance (for full species list, see AppendixA). This species grouping into plant functional types (PFT) was already investigated in former studies [32,33]. In this study, only a ﬁre tolerance was assigned to each species group.

PFT Maximum Height [m] Shade Tolerance Exemplary Tree Species Fire Tolerance 1 >33 shade tolerant Strombosia schefﬂeri 2 2 16–33 shade tolerant Heinsenia diervilleoides 1 3 16–33 medium shade tolerant Ficus sur 1 4 16–33 shade intolerant (pioneer) Polyscias albersiana 4 5 <16 shade tolerant Leptonychia usambarensis 2 6 <16 shade intolerant (pioneer) Cyathea manniana 2

3. Results 3.1. The Consequences of a Fire Event on the Carbon Fluxes in a Tropical Forest In order to understand the effects of a large-scale fire on tropical forest dynamics, a single fire was simulated with the forest model FORMIND (simulation area 9 ha, fire size 60%, fire severity 0.67; see Figure8). After the simulated fire event, the total aboveground biomass was reduced by 46% (Figure9). Overall, it takes up to 150 years after the fire event for the forest biomass to reach the level of an undisturbed forest and to stabilize—which is much longer than the natural succession (100 years, cf. Figure9). Compared with the succession in the undisturbed forest, this is mainly due to the climax species, which need up to 2 times longer to reach the biomass level of the undisturbed forest. Appl. Sci. 2021, 11, x FOR PEER REVIEW 9 of 18

Table 2. Species grouping with fire tolerance at Mt. Kilimanjaro region. Grouping of all tree species in the study area into six species groups according to maximum height and shade tolerance (for full species list, see Appendix A). This species grouping into plant functional types (PFT) was already investigated in former studies [32,33]. In this study, only a fire tolerance was assigned to each species group.

Maximum Fire Tol- PFT Shade Tolerance Exemplary Tree Species Height [m] erance 1 >33 shade tolerant Strombosia scheffleri 2 2 16–33 shade tolerant Heinsenia diervilleoides 1 3 16–33 medium shade tolerant Ficus sur 1 4 16–33 shade intolerant (pioneer) Polyscias albersiana 4 5 <16 shade tolerant Leptonychia usambarensis 2 6 <16 shade intolerant (pioneer) Cyathea manniana 2

3. Results 3.1. The Consequences of a Fire Event on the Carbon Fluxes in a Tropical Forest In order to understand the effects of a large-scale fire on tropical forest dynamics, a single fire was simulated with the forest model FORMIND (simulation area 9 ha, fire size 60%, fire severity 0.67; see Figure 8). After the simulated fire event, the total aboveground biomass was reduced by 46% (Figure 9). Overall, it takes up to 150 years after the fire event for the forest biomass to reach the level of an undisturbed forest and to stabilize— Appl. Sci. 2021, 11, 4696 which is much longer than the natural succession (100 years, cf. Figure 9). Compared9 with of 17 the succession in the undisturbed forest, this is mainly due to the climax species, which need up to 2 times longer to reach the biomass level of the undisturbed forest.

FigureFigure 8.8. TheThe heterogeneityheterogeneity ofof biomassbiomass inin aa virtualvirtual landscapelandscapebefore beforethe thesimulated simulated fire fire ( left(left),), during during thethe firefire ( (middle)middle),, andand afterafter thethe firefire ((right).right). Here, the green level indicates thethe amountamount ofof abovegroundaboveground biomass;biomass; brownbrown correspondscorresponds toto anan areaarea withoutwithout biomassbiomass inin thethe forest.forest. TheThe spreadspread ofof thethe simulatedsimulated fire is shown in red. Until the fire event, the forest was simulated undisturbed on an area of 9 hec- fire is shown in red. Until the fire event, the forest was simulated undisturbed on an area of 9 hectares tares (300 m × 300 m; one pixel has the size of 20 m × 20 m). After 250 years of simulation, a large- (300 m × 300 m; one pixel has the size of 20 m × 20 m). After 250 years of simulation, a large-scale scale fire was realized (here: fire size 60%, intensity 0.67). Not all trees die in the fire area, and thus fireparts was of realizedthe fire (here:area still fire sizeshow 60%, biomass intensity values 0.67). after Not the all treesfire event. die in Please the fire note, area, andin this thus schemati parts ofc Appl. Sci. 2021, 11, x FOR PEER REVIEWthe fire area still show biomass values after the fire event. Please note, in this schematic visualization10 of 18 visualization only a virtual landscape is shown. only a virtual landscape is shown. Due to the different fire tolerances, but also due to the different tree size distributions of the individual tree species, the fire affects the species groups differently (Figure 9a). The biomass of the dominant climax species was reduced by 46%; the biomass of the pioneer species was reduced by about 53%. However, since the forest was dominated by climax species before the forest fire, this led to the largest absolute biomass losses (−158 t/ha) compared with the other species groups. After the fire, secondary succession starts at the burnt areas. First, the pioneer species colonize these areas. The previously absent fire-tolerant and shade-intolerant tree species dominate the forest in this regeneration phase. Af- ter about 50 years, the biomass of trees with high shade tolerance then also increases again.

FiFiguregure 9. 9. SimulatedSimulated aboveground aboveground biomass biomass ( (aa)) and and carbon carbon fluxes fluxes ( (b)) after after a a fire fire event. A A large large-scale-scale firefire was simulated in thethe yearyear 250.250. TheThe developmentdevelopment of of aboveground aboveground biomass biomass per per shade-tolerance shade-tolerance is isshown shown on on the the left-hand left-hand side. side. In In addition, addition, the the gross gross primary primary production production (GPP), (GPP), tree tree respiration, respiration, net netprimary primary production production (NPP), (NPP) and, and mortality mortality for thefor simulatedthe simulated forest forest area area are shown are shown on the on right-hand the right- hand side. Mortality is without burnt trees. side. Mortality is without burnt trees.

ForestDue to productivity the different two fire tolerances,years after butthe alsofire is due up toto the4 times different larger tree than size befo distributionsre the fire (upof the to individual92 tC/(ha yr), tree Figure species, 9b). the The fire higher affects productivity the species groupsis mainly differently caused by (Figure growing9a). Thepio- neerbiomass species of the on dominantthe burnt climaxarea. The species respiration was reduced of the by living 46%; vegetation the biomass increases of the pioneer 5-fold afterspecies the wasforest reduced fire. However, by about the 53%. resulting However, net sinceprimary the production forest was dominatedremains rather by climaxstable afterspecies the beforefire at the value forest of fire, about this 3 led–4 t toC/(ha the yr). largest Both absolute productivity biomass and lossesrespiration (−158 do t/ha) not returncompared to pre with-fire the levels other until species more groups.than 100 After years the after fire, the secondary fire (Figure succession 9b). Biomass starts loss at duethe burntto mortality areas. is First, very the high pioneer during species the fire, colonize but already these 10 areas. years Theafter previously the fire it almost absent reachesfire-tolerant the level and of shade-intolerant 3.5 tC per hectare tree and species year again dominate (Figure the 9b). forest in this regeneration phase.During After the about fire event 50 years,, a negative the biomass carbon of balance trees with(NEE) high was shadeobserved, tolerance as up to then 75 tons also ofincreases carbon again.per hectare are emitted (Figure 10a). In the ten years after the fire, the NEE was positiveForest due productivity to the high productivity two years after of thepioneer fire is species. up to 4 However, times larger for than the next before 20 the years, fire the(up carbon to 92 t Cbalance/(ha yr), was Figure negative9b)., with The higheran average productivity annual carb is mainlyon source caused of 0.8 by tC/ha. growing In the followingpioneer species 50 years, on thethe burntforestarea. absorbs The more respiration carbon of than the livingit emits, vegetation on average increases 0.6 tC/ha 5-fold per year.after theFor forestthe cumulative fire. However, carbon the balance, resulting including net primary the carbon production emission remains released rather during stable theafter fire, the the fire forest at the needs value ofmany about decades 3–4 tC /(hato sequ yr).ester Both enough productivity carbon and to respirationreturn to pre do-fire not levels.return The to pre-fire overall levels balance until of the more carbon than 100fluxes years remains after thenegative fire (Figure for more9b). than Biomass 150 years loss afterdue tothe mortality fire (Figure is very 10b). high during the fire, but already 10 years after the fire it almost reaches the level of 3.5 tC per hectare and year again (Figure9b).

Figure 10. Simulated net ecosystem exchange after a fire event. Positive values correspond to net carbon sequestration in the forest. In the year 250, a large-scale fire was simulated, which led to a strong carbon source. (a) Annual carbon flux (NEE) and (b) cumulative carbon flux. The red area in (b) indicates the cumulative carbon flux before the forest fire.

Appl. Sci. 2021, 11, x FOR PEER REVIEW 10 of 18

Figure 9. Simulated aboveground biomass (a) and carbon fluxes (b) after a fire event. A large-scale fire was simulated in the year 250. The development of aboveground biomass per shade-tolerance is shown on the left-hand side. In addition, the gross primary production (GPP), tree respiration, net primary production (NPP), and mortality for the simulated forest area are shown on the right- hand side. Mortality is without burnt trees.

Forest productivity two years after the fire is up to 4 times larger than before the fire (up to 92 tC/(ha yr), Figure 9b). The higher productivity is mainly caused by growing pioneer species on the burnt area. The respiration of the living vegetation increases 5-fold after the forest fire. However, the resulting net primary production remains rather stable

Appl. Sci. 2021, 11, 4696 after the fire at the value of about 3–4 tC/(ha yr). Both productivity and respiration do10 ofnot 17 return to pre-fire levels until more than 100 years after the fire (Figure 9b). Biomass loss due to mortality is very high during the fire, but already 10 years after the fire it almost reaches the level of 3.5 tC per hectare and year again (Figure 9b). DuringDuring the the fire fire event event,, a a negative negative carbon carbon balance (NEE) was observed, as u upp to 75 tons tons ofof carbon carbon per per hectare hectare are are emitted emitted ( (FigureFigure 10a).10a). In In the the ten ten years years after after the the fire fire,, the NEE was positivepositive due due to to the the high high productivity productivity of of pioneer pioneer species. species. However, However, for for the the next next 20 20 years, years, thethe carbon carbon balance balance was negative,negative, with anan averageaverage annualannual carboncarbon sourcesource of of 0.8 0.8 t CtC/ha./ha. In the followingfollowing 50 50 years, years, the forest absorbs more carboncarbon thanthan itit emits,emits, onon averageaverage 0.60.6 t CtC/ha/ha per year.year. For the cumulative carbon balance, including the carbon emission released during thethe fire, fire, the the forest forest needs many many decades decades to to sequ sequesterester enough enough carbon carbon to to return to pre pre-fire-fire levels.levels. The The overall overall balance balance of of the the carbon carbon fluxes fluxes remains remains negative negative for for more more than than 150 150 years afterafter the the fire fire ( (FigureFigure 10b).10b).

FigureFigure 10. SimulatedSimulated net net ecosystem exchange exchange after after a a fire fire event. Positive Positive values values correspond correspond to to net carboncarbon sequestration sequestration in in the the forest. forest. In In the the year year 250, 250, a a large large-scale-scale fire fire was simulated simulated,, which led to a strongstrong carbon carbon source. source. ( (aa)) Annual Annual carbon carbon flux flux (NEE) (NEE) and and (b)) cumulative cumulative carbon carbon flux. flux. The The red red area area in (b) indicates the cumulative carbon flux before the forest fire. (b) indicates the cumulative carbon flux before the forest fire.

3.2. The Influence of Recurrent Fire Events on Forest Dynamics In order to estimate the consequences of regularly occurring fire events on the forest dynamics, numerous fires were simulated with the forest model FORMIND. On average, a fire of the size of 20% of the simulation area (9 hectares) was simulated every 4 years (cf. Table1 ). This scenario corresponds to results of satellite data analysis on fire areas at the Kilimanjaro region. Compared with the undisturbed scenario without fire, the dynamics in the fire- disturbed forest are significantly altered (Figure 11). Total aboveground biomass is reduced by 80%; the forest thus has a deficit of 300 tons of biomass per hectare. The first species group consisting of large climax species has the largest reduction, as this species group loses 95% of its biomass. The pioneer species (PFT 4 and 6), on the other hand, can increase their biomass considerably. Both species groups account in total for 70% of the total biomass in the fire scenario, whereas they played no role in the undisturbed scenario. The disturbed forest does not reach a state of equilibrium in terms of species composition. The climax species can increase in biomass after a fire, but it loses a lot of biomass during the next fire event. Overall, the shade-tolerant tree species (PFT1 and 2) account for only 8% of the total biomass, which varies largely over time (1–45%). Appl. Sci. 2021, 11, x FOR PEER REVIEW 11 of 18

3.2. The Influence of Recurrent Fire Events on Forest Dynamics In order to estimate the consequences of regularly occurring fire events on the forest dynamics, numerous fires were simulated with the forest model FORMIND. On average, a fire of the size of 20% of the simulation area (9 hectares) was simulated every 4 years (cf. Table 1). This scenario corresponds to results of satellite data analysis on fire areas at the Kilimanjaro region. Compared with the undisturbed scenario without fire, the dynamics in the fire-disturbed forest are significantly altered (Figure 11). Total aboveground biomass is reduced by 80%; the forest thus has a deficit of 300 tons of biomass per hectare. The first species group consisting of large climax species has the largest reduction, as this species group loses 95% of its biomass. The pioneer species (PFT 4 and 6), on the other hand, can increase their biomass considerably. Both species groups account in total for 70% of the total biomass in the fire scenario, whereas they played no role in the undisturbed scenario. The disturbed forest does not reach a state of equilibrium in terms of species composition. The Appl. Sci. 2021, 11, 4696 climax species can increase in biomass after a fire, but it loses a lot of biomass during11 of the 17 next fire event. Overall, the shade-tolerant tree species (PFT1 and 2) account for only 8% of the total biomass, which varies largely over time (1–45%).

FigureFigure 11. 11.Simulated Simulated aboveground aboveground biomass biomass for for each each plant plant functional functional type type (PFT; (PFT; climax climax green, green, pioneer pioneer red, red, mid-tolerance mid-tolerance purple)purple) without without fire fire events events ( a(a)) and and after after numerous numerous fire fire events events ( b(b).). On On average, average, therethere isis aa firefire in in this this region region every every 44 years.years. (c()c) Comparison Comparison of of the the species species composition composition for for an undisturbedan undisturbed and and disturbed disturbed forest forest (here, (here, the biomass the biomass fraction fraction of each of plant each functionalplant functional type serves type asserves a proxy as a forproxy diversity). for diversity). odm = odm organic = organic dry matter. dry matter.

DueDue toto thethe highhigh carbon emissions during a a fire fire event event and and the the carbon carbon sequestration sequestration in inthe the regeneration regeneration phase phase after after a fire, a fire, the the variation variation of the of thecarbon carbon balance balance (NEE) (NEE) is very is very high high(Figure (Figure 12b). 12Inb). the Inundisturbed the undisturbed forest, the forest, standard the standard deviation deviation of carbon of flux carbon is 0.4 flux tC ha is−1 −1 −1 −1 0.4yr tC, whereasha yr in, whereasthe scenario in the with scenario regular with forest regular fires, forest the standard fires, the deviation standard is deviation 10 times ishigher. 10 times However, higher. the However, mean carbon the mean flux is carbon balanced flux at iszero, balanced just as atin the zero, undisturbed just as in thesce- Appl. Sci. 2021, 11, x FOR PEER REVIEWundisturbednario (Figure scenario 12a,b). Considering (Figure 12a,b). the Considering cumulative carbon the cumulative balance of carbon both scenarios balance of (Fig12 both ofure 18 scenarios12c), both (Figure reach a12 statec), both of equilibrium reach a state after of equilibrium about 100– after150 years. about However, 100–150 years. the equilibrium However, −1 −1 thestate equilibrium in the scen stateario without in the scenario fire is 150 without tC ha fire higher. is 150 tC ha higher.

Figure 12.12. Simulated carbon balance (net ecosystem exchange) over time in a tropical forest without fires ﬁres (a)) andand withwith recurrent ﬁresfires onon averageaverage everyevery 44 yearsyears ((bb).). PositivePositive valuesvalues correspondcorrespond toto carboncarbon sinksink behavior.behavior. ((cc)) CumulativeCumulative carboncarbon balance of both scenarios.

Investigating thethe speciesspecies compositioncomposition with with increasing increasing fire fire intensity, intensity, the the simulation simulation re- resultssults show show the the typical typical picture picture of a of hump-shaped a hump-shaped curve curve (Figure (Figure 13). If 13). the ecosystemIf the ecosystem is undis- is undisturbedturbed (i.e., fire (i.e. frequency, fire frequency is zero), is the zero), normalized the normalized Shannon Shannon Index is lowIndex at 30%is low (= at standard 30% (= standardShannon IndexShannon divided Index by divided ln(6)). If by a fireln(6)). occurs If a annuallyfire occurs (equivalent annually (equivalent to 10 fires per to decade), 10 fires perthe Shannondecade), Indexthe Shannon increases Index to 50%. increases Fires that to occur50%. lessFires frequently that occur than less annually, frequently however, than produceannually, an however, even higher produce species an even diversity higher (Figure species 13 diversity). This results (Figure in 13). a maximum This results of in the a maximumShannon Index of the for Shannon 2–3 fire eventsIndex for per 2 decade.–3 fire events Analysis per of decade. satellite Analysis data for theof satellite Kilimanjaro data region showed between one to four forest fires per decade. Thus, high species diversity can for the Kilimanjaro region showed between one to four forest fires per decade. Thus, high be expected in disturbed areas around Mt. Kilimanjaro. If future fire frequency increases species diversity can be expected in disturbed areas around Mt. Kilimanjaro. If future fire due to higher temperatures and less precipitation [14,57], there might be a reduction in frequency increases due to higher temperatures and less precipitation [14,57], there might functional diversity in the study area, according to the simulation results (Figure 13). be a reduction in functional diversity in the study area, according to the simulation results (Figure 13).

Figure 13. Simulated species diversity for different fire frequencies. In each fire scenario, forest fires were simulated with a frequency of 1 to 10 fires per decade. The parameters are the same as in the standard fire scenario (see Table 1). The species diversity is described with the normalized Shannon Index (standard Shannon Index divided by ln(6)). For this index, the relative biomass fraction of each plant functional type is calculated over the last 300 years of simulation (in total, 6 plant functional types).

Appl. Sci. 2021, 11, x FOR PEER REVIEW 12 of 18

Figure 12. Simulated carbon balance (net ecosystem exchange) over time in a tropical forest without fires (a) and with recurrent fires on average every 4 years (b). Positive values correspond to carbon sink behavior. (c) Cumulative carbon balance of both scenarios.

Investigating the species composition with increasing fire intensity, the simulation results show the typical picture of a hump-shaped curve (Figure 13). If the ecosystem is undisturbed (i.e., fire frequency is zero), the normalized Shannon Index is low at 30% (= standard Shannon Index divided by ln(6)). If a fire occurs annually (equivalent to 10 fires per decade), the Shannon Index increases to 50%. Fires that occur less frequently than annually, however, produce an even higher species diversity (Figure 13). This results in a maximum of the Shannon Index for 2–3 fire events per decade. Analysis of satellite data for the Kilimanjaro region showed between one to four forest fires per decade. Thus, high species diversity can be expected in disturbed areas around Mt. Kilimanjaro. If future fire frequency increases due to higher temperatures and less precipitation [14,57], there might be a reduction in functional diversity in the study area, according to the simulation results Appl. Sci. 2021, 11, 4696 12 of 17 (Figure 13).

FigureFigure 13. 13. SimulatedSimulated species species diversity diversity for fordifferent different fire firefrequencies. frequencies. In each In eachfire scenario, fire scenario, forest forestfires werefires simulated were simulated with a withfrequency a frequency of 1 to of10 1fires to 10 per fires decade. per decade. The parameters The parameters are the aresame the as same in the as standardin the standard fire scenario fire scenario (see Table (see 1). Table The species1). The diversity species diversity is described is described with the normalized with the normalized Shannon IndexShannon (standard Index Shannon (standard Index Shannon divided Index by dividedln(6)). For by this ln(6)). index, For the this relative index, thebiomass relative fraction biomass of each plant functional type is calculated over the last 300 years of simulation (in total, 6 plant func- fraction of each plant functional type is calculated over the last 300 years of simulation (in total, tional types). 6 plant functional types).

4. Discussion Fires represent an important disturbance type when considering the terrestrial bio- sphere [14,57,58]. In a forest, fire can lead to strong changes, especially in species composition, forest succession, and carbon dynamics [19]. As African forests in particular are strongly affected by fire, it is important to study the different consequences of fire, as this can have an impact on forest dynamics, as well as on the function of tropical forests as a carbon sink. Two different scenarios were investigated in this modelling study. The first scenario analyzed the changes in aboveground biomass, productivity, and carbon balance of a tropical forest at Mt. Kilimanjaro after a large-scale fire. A second scenario analyzed changes after regularly recurring fires. It turned out that the forest needs more than 100 years to recover to its pre-fire state. The climax species lose the largest amount of biomass, as they dominated the forest before the fire. Productivity and respiration are up to 4 times higher after the fire than before the fire. This is mainly due to the regrowing pioneer species. After the fire, trees of fire-tolerant species will dominate a forest stand. This alters the succession in tropical forest, as it is more species-rich after a fire and the fire-tolerant species prolong the succession. This effect also has an influence on the carbon balance in a tropical forest. Please note that the results presented apply specifically to the parameterized forest area at the Mt. Kilimanjaro region. In addition, both the parameters of the forest model and the parameters of the fire model include uncertainties. Although monthly satellite data were available, it is challenging to make statements about the type of fire (ground fire vs. crown fire) and the area of spread. Important in the analysis of the fire model was the consideration of the feedback of a fire event with the vegetation. This means that a fire has an influence on the vegetation, and the vegetation influences the characteristics of a fire. This was implemented in such a way that a fire only breaks out if combustible material is present and the soil or litter layer is relatively dry. Furthermore, a tree burns depending on its fire tolerance and its age. What is not considered are the long-term effects of fire on soil properties, such as nutrient inputs. Since the growth of trees in the forest model has so far been described independent of the nutrient balance in the soil, nutrient inputs have not been considered in this study. Appl. Sci. 2021, 11, 4696 13 of 17

In fact, no real fire experiment was conducted in the study area at Mt. Kilimanjaro. The fire experiments available in the literature often refer to the effects on grasslands [59–61]. These experiments involve little cost, the time horizon is manageable, and the damage to nature is not so high (compared to a fire experiment in a forest). Experiments concerning forest fires refer mostly to temperate forests or to the tropical forest in the Amazon. The influence of fire on forest structure, biomass, and species composition was investigated in such an experiment [7,62]. Losses in living biomass after fire ranged from 10% to 90%, depending on how often fire occurred in the area studied. Our simulations indicate that aboveground biomass can be reduced by between 40% and 80%, and thus the results are comparable to the analyses in the Amazon. This is also confirmed by a second study in the Amazon, which found 42% to 57% losses in biomass after a fire [63]. Some studies related to fire consider not only the loss of biomass but also the change in species composition [64,65]. In this simulation study, a significant change in species composition was also observed. The forest without disturbance was dominated by climax species; in the simulated forest with fire, the proportion of pioneer species increased. Thus, the species composition looked much more diverse. This was evaluated by calculating the normalized Shannon Index: in the undisturbed forest it was 30%, and in the fire-disturbed forest (fire events every 4 years), it reaches up to 80%. This means that fire events can increase functional diversity. In this study the relationship between fire frequency and species diversity was also investigated, which is also a topic of the intermediate disturbance hypothesis (IDH) [66]: “The highest diversity of tropical rainforest trees should occur ... with smaller disturbances that are neither very frequent nor infrequent”. According to this theory, ecosystems respond to light but regular disturbances with increasing species diversity. Species diversity shows a maximum at medium disturbance intensity. According to the IDH, however, species diversity declines again when disturbances are intense. This study shows a maximum of species diversity when 2–3 forest fires per decade occur in a region. Empirical studies in a forest-grassland ecotone showed similar results with a maximum of heterogeneity at five fire events per decade [65]. This simulation study at Mt. Kilimanjaro revealed that the investigated forest shows responses to disturbances similar to the IDH. Physiological trade-offs between different tree species—here the trade-off between growth and mortality—might be an important factor for this phenomenon. However, it must also be noted that with 2–3 fires per decade, the tropical forest has a much lower biomass level than an undisturbed forest, with negative consequences for its carbon balance.

5. Conclusions Fires affect millions of hectares of tropical forest, generating billions of USD in damage annually [14]. Science should pay more attention to fire events in all tropical areas. This should not be done by simply transferring knowledge from fire experiments of temperate forests to the species-rich complex tropical forests. Instead, studies and experiments on the effects of forest fires in tropical areas should be intensified. This modelling study shows how a forest model can support such studies by extrapolating local findings in both time and space.

Funding: This study was conducted within the framework of the research unit FOR1246 (Kilimanjaro ecosystems under global change: linking biodiversity, biotic interactions and biogeochemical ecosystem processes) funded by the Deutsche Forschungsgemeinschaft (DFG). I acknowledge funding from the Helmholtz Alliance “Remote Sensing and Earth System Dynamics”. Data Availability Statement: The forest model FORMIND, the source code, and the parametrization are freely available and can be downloaded from www.formind.org, 19 May 2021. Acknowledgments: I want to thank Andreas Huth for helpful discussions and constructive comments on the manuscript and Ulrike Hiltner and Martina Alrutz for initial investigations of burned Appl. Sci. 2021, 11, 4696 14 of 17

areas with satellite data. I would like to thank the reviewers for their thoughtful comments and efforts towards improving the manuscript. Conﬂicts of Interest: The author declares no conﬂict of interest.

Appendix A Species List of the Study Area at Mt. Kilimanjaro List of tree species found in the tropical forest at Mt. Kilimanjaro with maximum growth height and light requirement (tol = shade tolerant; intol = shade intolerant; med = medium light requirement). All tree species were grouped into six plant functional types (PFT). For details of the species grouping, see [32,33]. In addition, the fire tolerance is indicated for each species (1 = fire intolerant to 4 = fire tolerant).

Species Max. Height (m) Shade Tolerance PFT Fire Tolerance Agarista salicifolia 20 intol 4 4 Alangium chinense 25 intol 4 1 Albizia gummifera 30 tol 2 1 Aningeria adolﬁ friedericii 50 tol 1 1 Aphloia theiformis 15 med 3 1 Bersama abyssinica 20 med 3 1 Canthium oligocarpum ssp captum 10 tol 5 1 Casearia battiscombei 30 tol 1 1 Celtis durandii 25 med 3 1 Clausena anisata 20 med 3 1 Cornus volkensii 30 med 3 1 Cyathea manniana 15 intol 6 1 Dracaena afromontana 10 tol 5 1 Dracaena laxissima 5 tol 5 1 Eckebergia capensis 25 tol 2 1 Embelia schimperi 30 tol 2 1 Entandrophragma excelsum 70 tol 1 1 Erica excelsa 28 intol 4 4 Erythrococca polyandra 10 tol 5 1 Ficus sur 25 med 3 1 Galiniera saxifraga 15 tol 5 1 Garcinia tansaniensis 40 tol 1 1 Garcinia volkensii 20 tol 2 1 Hagenia abyssinica 25 intol 4 3 Hallea rubrostipulata 33 med 3 1 Heinsenia diervilleoides 25 tol 2 1 Ilex mitis 30 tol 2 1 Lasianthus kilinandscharicus 10 tol 5 1 Lepidotrichilia volkensii 16 med 3 1 Leptonychia usambarensis 15 tol 5 1 Macaranga capensis var kilimandscharica 30 med 3 1 Maesa lanceolata 20 intol 4 1 Maytenus acuminata 15 tol 5 1 Morella salicifolia 15 intol 4 4 Myrica salicifolia 15 intol 4 4 Newtonia buchananii 40 tol 1 1 Ocotea usambarensis 45 tol 1 2 Olinia rochetiana 25 med 3 1 Pauridiantha paucinervis 10 tol 5 1 Pavetta abyssinica 15 tol 5 1 Appl. Sci. 2021, 11, 4696 15 of 17

Species Max. Height (m) Shade Tolerance PFT Fire Tolerance Peddiea fischeri 15 tol 5 1 Pittosporum spec lanatum 10 med 3 1 Podocarpus latifolius 35 tol 1 1 Polyscias fulva 25 intol 4 1 Polyscias albersiana 15 intol 4 1 Psychotria cyathicalyx 15 tol 5 1 Rapanea melanophloeos 30 med 3 1 Rawsonia lucida 25 tol 2 1 Rothmannia urcelliformis 10 tol 2 1 Schefflera myriantha 30 tol 2 1 Schefflera volkensii 25 tol 2 1 Strombosia scheffleri 35 tol 1 1 Syzygium guineense 30 tol 2 1 Tabernaemontana stapfiana 25 med 3 1 Teclea nobilis 20 tol 2 1 Xymalos monospora 25 tol 2 1

References 1. Ciais, P.; Sabine, C.; Bala, G.; Bopp, L.; Brovkin, V.; Canadell, J.; Chhabra, A.; DeFries, R.; Galloway, J.; Heimann, M.; et al. Carbon and other biogeochemical cycles. In Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, I., Nauels, A., Xia, Y., Bex, V., Midgley, P.M., Eds.; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2013; pp. 465–570. [CrossRef] 2. Friedlingstein, P.; O’Sullivan, M.; Jones, M.W.; Andrew, R.M.; Hauck, J.; Olsen, A.; Peters, G.P.; Peters, W.; Pongratz, J.; Sitch, S.; et al. Global Carbon Budget 2020. Earth Syst. Sci. Data 2020, 12, 3269–3340. [CrossRef] 3. Harris, N.L.; Gibbs, D.A.; Baccini, A.; Birdsey, R.A.; de Bruin, S.; Farina, M.; Fatoyinbo, L.; Hansen, M.C.; Herold, M.; Houghton, R.A.; et al. Global maps of twenty-first century forest carbon fluxes. Nat. Clim. Chang. 2021, 11, 234–240. [CrossRef] 4. Houghton, R.A.; Hall, F.; Goetz, S.J. Importance of biomass in the global carbon cycle. J. Geophys. Res. Biogeosci. 2009, 114. [CrossRef] 5. Pan, Y.D.; Birdsey, R.A.; Fang, J.Y.; Houghton, R.; Kauppi, P.E.; Kurz, W.A.; Phillips, O.L.; Shvidenko, A.; Lewis, S.L.; Canadell, J.G.; et al. A large and persistent carbon sink in the world’s forests. Science 2011, 333, 988–993. [CrossRef][PubMed] 6. Pugh, T.A.M.; Arneth, A.; Kautz, M.; Poulter, B.; Smith, B. Important role of forest disturbances in the global biomass turnover and carbon sinks. Nat. Geosci. 2019, 12, 730–735. [CrossRef] 7. Cochrane, M.A.; Alencar, A.; Schulze, M.D.; Souza, C.M.; Nepstad, D.C.; Lefebvre, P.; Davidson, E.A. Positive feedbacks in the fire dynamic of closed canopy tropical forests. Science 1999, 284, 1832–1835. [CrossRef][PubMed] 8. Nepstad, D.C.; Verissimo, A.; Alencar, A.; Nobre, C.; Lima, E.; Lefebvre, P.; Schlesinger, P.; Potter, C.; Moutinho, P.; Mendoza, E.; et al. Large-scale impoverishment of Amazonian forests by logging and fire. Nature 1999, 398, 505–508. [CrossRef] 9. Andela, N.; Morton, D.C.; Giglio, L.; Chen, Y.; van der Werf, G.R.; Kasibhatla, P.S.; DeFries, R.S.; Collatz, G.J.; Hantson, S.; Kloster, S.; et al. A human-driven decline in global burned area. Science 2017, 356, 1356–1362. [CrossRef] 10. Hansen, M.C.; Potapov, P.V.; Moore, R.; Hancher, M.; Turubanova, S.A.; Tyukavina, A.; Thau, D.; Stehman, S.V.; Goetz, S.J.; Loveland, T.R.; et al. High-resolution global maps of 21st-century forest cover change. Science 2013, 342, 850–853. [CrossRef] 11. Brando, P.M.; Silverio, D.; Maracahipes-Santos, L.; Oliveira-Santos, C.; Levick, S.R.; Coe, M.T.; Migliavacca, M.; Balch, J.K.; Macedo, M.N.; Nepstad, D.C.; et al. Prolonged tropical forest degradation due to compounding disturbances: Implications for CO2 and H2O fluxes. Glob. Chang. Biol. 2019, 25, 2855–2868. [CrossRef][PubMed] 12. Morton, D.C.; DeFries, R.S.; Randerson, J.T.; Giglio, L.; Schroeder, W.; van der Werf, G.R. Agricultural intensification increases deforestation fire activity in Amazonia. Glob. Chang. Biol. 2008, 14, 2262–2275. [CrossRef] 13. van der Werf, G.R.; Randerson, J.T.; Giglio, L.; Collatz, G.J.; Mu, M.; Kasibhatla, P.S.; Morton, D.C.; DeFries, R.S.; Jin, Y.; van Leeuwen, T.T. Global fire emissions and the contribution of deforestation, savanna, forest, agricultural, and peat fires (1997–2009). Atmos. Chem. Phys. 2010, 10, 11707–11735. [CrossRef] 14. Cochrane, M.A. Fire science for rainforests. Nature 2003, 421, 913–919. [CrossRef][PubMed] 15. Hemp, A. The impact of fire on diversity, structure, and composition of the vegetation on Mt. Kilimanjaro. In Land Use Change and Mountain Biodiversity; Spehn, E.M., Liberman, M., Körner, C., Eds.; CRC Press: Boca Raton, FL, USA, 2006; pp. 51–68. 16. Simon, M.; Plummer, S.; Fierens, F.; Hoelzemann, J.J.; Arino, O. Burnt area detection at global scale using ATSR-2: The GLOBSCAR products and their qualification. J. Geophys. Res. Atmos. 2004, 109, D14S02. [CrossRef] 17. Hirschberger, P. Wälder in Flammen. Ursachen und Folgen der Weltweiten Waldbrände; WWF Deutschland: Berlin, Germany, 2012; p. 90. 18. Hemp, A. Climate change-driven forest fires marginalize the impact of ice cap wasting on Kilimanjaro. Glob. Chang. Biol. 2005, 11, 1013–1023. [CrossRef] Appl. Sci. 2021, 11, 4696 16 of 17

19. Whelan, R.J. The Ecology of Fire; Cambridge University Press: Cambridge, UK, 1995. [CrossRef] 20. Axmacher, J.C.; Scheuermann, L.; Schrumpf, M.; Lyaruu, H.V.M.; Fiedler, K.; Müller-Hohenstein, K. Effects of fire on the diversity of geometrid moths on Mt. Kilimanjaro. In Land Use Change and Mountain Biodiversity; Spehn, E.M., Liberman, M., Körner, C., Eds.; CRC Press: Boca Raton, FL, USA, 2006; pp. 69–76. 21. Thonicke, K.; Venevsky, S.; Sitch, S.; Cramer, W. The role of fire disturbance for global vegetation dynamics: Coupling fire into a Dynamic Global Vegetation Model. Glob. Ecol. Biogeogr. 2001, 10, 661–677. [CrossRef] 22. Zinck, R.; Grimm, V. More realistic than anticipated: A classical forest-fire model from statistical physics captures real fire shapes. Open Ecol. J. 2008, 1, 8–13. [CrossRef] 23. Gardner, R.H.; Romme, W.H.; Turner, M.G. Predicting forest fire effects at landscape scales. In Spatial Modeling of Forest LANDSCAPE Change. Approaches and Applications; Mladenoff, D.J., Baker, W.L., Eds.; Cambridge University Press: Cambridge, UK, 1999; pp. 163–185. 24. Keane, R.E.; Cary, G.J.; Davies, I.D.; Flannigan, M.D.; Gardner, R.H.; Lavorel, S.; Lenihan, J.M.; Li, C.; Rupp, T.S. A classification of landscape fire succession models: Spatial simulations of fire and vegetation dynamics. Ecol. Model. 2004, 179, 3–27. [CrossRef] 25. Bugmann, H. A review of forest gap models. Clim. Chang. 2001, 51, 259–305. [CrossRef] 26. Shugart, H.H.; Wang, B.; Fischer, R.; Ma, J.; Fang, J.; Yan, X.; Huth, A.; Armstrong, A.H. Gap models and their individual-based relatives in the assessment of the consequences of global change. Environ. Res. Lett. 2018, 13, 033001. [CrossRef] 27. Fischer, R.; Bohn, F.; de Paula, M.D.; Dislich, C.; Groeneveld, J.; Gutiérrez, A.G.; Kazmierczak, M.; Knapp, N.; Lehmann, S.; Paulick, S.; et al. Lessons learned from applying a forest gap model to understand ecosystem and carbon dynamics of complex tropical forests. Ecol. Model. 2016, 326, 124–133. [CrossRef] 28. Köhler, P.; Huth, A. Simulating growth dynamics in a South-East Asian rainforest threatened by recruitment shortage and tree harvesting. Clim. Chang. 2004, 67, 95–117. [CrossRef] 29. Hemp, A. Continuum or zonation? Altitudinal gradients in the forest vegetation of Mt. Kilimanjaro. Plant Ecol. 2006, 184, 27–42. [CrossRef] 30. Peters, M.K.; Hemp, A.; Appelhans, T.; Behler, C.; Classen, A.T.; Detsch, F.; Ensslin, A.; Ferger, S.W.; Frederiksen, S.B.; Gebert, F.; et al. Predictors of elevational biodiversity gradients change from single taxa to the multi-taxa community level. Nat. Commun. 2016, 7, 13736. [CrossRef] 31. Ensslin, A.; Rutten, G.; Pommer, U.; Zimmermann, R.; Hemp, A.; Fischer, M. Effects of elevation and land use on the biomass of trees, shrubs and herbs at Mount Kilimanjaro. Ecosphere 2015, 6, art45. [CrossRef] 32. Fischer, R.; Ensslin, A.; Rutten, G.; Fischer, M.; Schellenberger Costa, D.; Kleyer, M.; Hemp, A.; Paulick, S.; Huth, A. Simulating carbon stocks and fluxes of an African tropical montane forest with an individual-based forest model. PLoS ONE 2015, 10, e0123300. [CrossRef] 33. Fischer, R.; Rödig, E.; Huth, A. Consequences of a reduced number of plant functional types for the simulation of forest productivity. Forests 2018, 9, 460. [CrossRef] 34. Paulick, S.; Dislich, C.; Homeier, J.; Fischer, R.; Huth, A. The carbon fluxes in different successional stages: Modelling the dynamics of tropical montane forests in South Ecuador. For. Ecosyst. 2017, 4. [CrossRef] 35. Armstrong, A.H.; Huth, A.; Osmanoglu, B.; Sun, G.; Ranson, K.J.; Fischer, R. A multi-scaled analysis of forest structure using individual-based modeling in a costa rican rainforest. Ecol. Model. 2020, 433, 109226. [CrossRef] 36. Hiltner, U.; Huth, A.; Hérault, B.; Holtmann, A.; Bräuning, A.; Fischer, R. Climate change alters the ability of neotropical forests to provide timber and sequester carbon. For. Ecol. Manag. 2021, 492, 119166. [CrossRef] 37. Kammesheidt, L.; Köhler, P.; Huth, A. Sustainable timber harvesting in Venezuela: A modelling approach. J. Appl. Ecol. 2001, 38, 756–770. [CrossRef] 38. Huth, A.; Drechsler, M.; Köhler, P. Multicriteria evaluation of simulated logging scenarios in a tropical rain forest. J. Environ. Manag. 2004, 71, 321–333. [CrossRef][PubMed] 39. Drossel, B.; Schwabl, F. Self-organized critical forest-fire model. Phys. Rev. Lett. 1992, 69, 1629–1632. [CrossRef][PubMed] 40. Green, D.G. Simulated effects of fire, dispersal and spatial pattern on competition within forest mosaics. Vegetatio 1989, 82, 139–153. [CrossRef] 41. Bugmann, H.K.M.; Solomon, A.M. The use of a European forest model in North America: A study of ecosystem response to climate gradients. J. Biogeogr. 1995, 22, 477–484. [CrossRef] 42. Bugmann, H.K.M. A simplified forest model to study species composition along climate gradients. Ecology 1996, 77, 2055–2074. [CrossRef] 43. Busing, R.T.; Solomon, A.M. Modeling the Effects of Fire Frequency and Severity on Forests in the Northwestern United States; Scientific Investigations Report 2006-5061; USGS: Reston, Virginia, 2007; p. 16. 44. Busing, R.T.; Solomon, A.M.; McKane, R.B.; Burdick, C.A. Forest dynamics in oregon landscapes: Evaluation and application of an individual-based model. Ecol. Appl. 2007, 17, 1967–1981. [CrossRef] 45. Schultz, J. Die Ökozonen der Erde-Die Ökologische Gliederung der Geosphäre, 1st ed.; Eugen Ulmer: Stuttgart, Germany, 1988; p. 488. 46. Thonicke, K.; Spessa, A.; Prentice, I.C.; Harrison, S.P.; Dong, L.; Carmona-Moreno, C. The influence of vegetation, fire spread and fire behaviour on biomass burning and trace gas emissions: Results from a process-based model. Biogeosciences 2010, 7, 1991–2011. [CrossRef] Appl. Sci. 2021, 11, 4696 17 of 17

47. Fischer, R.; Armstrong, A.; Shugart, H.H.; Huth, A. Simulating the impacts of reduced rainfall on carbon stocks and net ecosystem exchange in a tropical forest. Environ. Model. Softw. 2014, 52, 200–206. [CrossRef] 48. Dale, V.H.; Hemstrom, M.A. CLIMACS: A Computer Model of Forest Stand Development for Western Oregon and Washington; PNW-327; United States Department of Agriculture: Washington, DC, USA, 1984. 49. Archibald, S.; Roy, D.P. Identifying individual fires from satellite-derived burned area data. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Cape Town, South Africa, 12–17 July 2009; pp. 160–163. 50. Alrutz, M. Auswirkungen von Feuer auf den Tropischen Regenwald am Kilimandscharo, Tansania. Eine Modell-Studie; Universität Leipzig, Fakultät für Physik und Geowissenschaften, Institut für Geographie: Leipzig, Germany, 2013. 51. Goldammer, J.G. Feuer in Waldökosystemen der Tropen und Subtropen; Birkhäuser: Basel, Switzerland; Berlin, Germany, 1993. 52. Bond, W.J.; van Wilgen, B.W. Fire and Plants; Chapman & Hall: London, UK, 1996; Volume 14. 53. Hemp, A. Vegetation of Kilimanjaro: Hidden endemics and missing bamboo. Afr. J. Ecol. 2006, 44, 305–328. [CrossRef] 54. Hemp, A. Climate change and its impact on the forests of Kilimanjaro. Afr. J. Ecol. 2009, 47, 3–10. [CrossRef] 55. Tutul, E.; Uddin, M.; Rahman, M.; Hassan, M. Angiospermic flora of Runctia sal forest, Bangladesh. II. Magnoliopsida (Dicots). Bangladesh J. Plant Taxon. 2010, 17, 33–53. [CrossRef] 56. Rasolonandrasana, B.P.N.; Goodman, S.M. The influence of fire on mountain sclerophyllous forests and their small-mammal communities in Madagascar. In Land Use Change and Mountain Biodiversity; Spehn, E.M., Liberman, M., Körner, C., Eds.; CRC Press/Taylor & Francis: Boca Raton, FL, USA, 2006. [CrossRef] 57. Bowman, D.M.J.S.; Balch, J.K.; Artaxo, P.; Bond, W.J.; Carlson, J.M.; Cochrane, M.A.; D’Antonio, C.M.; DeFries, R.S.; Doyle, J.C.; Harrison, S.P.; et al. Fire in the Earth System. Science 2009, 324, 481–484. [CrossRef][PubMed] 58. Pfeiffer, M.; Spessa, A.; Kaplan, J.O. A model for global biomass burning in preindustrial time: LPJ-LMfire (v1.0). Geosci. Model Dev. 2013, 6, 643–685. [CrossRef] 59. Andersen, A.N.; Cook, G.D.; Corbett, L.K.; Douglas, M.M.; Eager, R.W.; Russell-Smith, J.; Setterfield, S.A.; Williams, R.J.; Woinarski, J.C.Z. Fire frequency and biodiversity conservation in Australian tropical savannas: Implications from the Kapalga fire experiment. Austral Ecol. 2005, 30, 155–167. [CrossRef] 60. Aragón, R.; Cristóbal, L.; Carilla, J. Fire, plant species richness, and aerial biomass distribution in mountain grasslands of northwest Argentina. In Land Use Change and Mountain Biodiversity; Spehn, E.M., Liberman, M., Körner, C., Eds.; CRC Press/Taylor & Francis: Boca Raton, FL, USA, 2006; pp. 89–99. [CrossRef] 61. Vogl, R.J. Effects of fire on grasslands. In Fire and Ecosystems; Kozlowski, T.T., Ahlgren, C.E., Eds.; Academic Press: New York, NY, USA, 1974; pp. 139–194. [CrossRef] 62. Cochrane, M.A.; Schulze, M.D. Fire as a recurrent event in tropical forests of the eastern Amazon: Effects on forest structure, biomass, and species composition. Biotropica 1999, 31, 2–16. [CrossRef] 63. Kauffman, J.B.; Cummings, D.L.; Ward, D.E.; Babbitt, R. Fire in the Brazilian Amazon: 1. Biomass, nutrient pools, and losses in slashed primary forests. Oecologia 1995, 104, 397–408. [CrossRef][PubMed] 64. Oliveras, I.; Malhi, Y.; Salinas, N.; Huaman, V.; Urquiaga-Flores, E.; Kala-Mamani, J.; Quintano-Loaiza, J.A.; Cuba-Torres, I.; Lizarraga-Morales, N.; Roman-Cuesta, R.M. Changes in forest structure and composition after fire in tropical montane cloud forests near the Andean treeline. Plant Ecol. Divers. 2014, 7, 329–340. [CrossRef] 65. Peterson, D.W.; Reich, P.B. Fire frequency and tree canopy structure influence plant species diversity in a forest-grassland ecotone. Plant Ecol. 2008, 194, 5–16. [CrossRef] 66. Connell, J.H. Diversity in tropical rain forests and coral reefs. Science 1978, 199, 1302–1310. [CrossRef][PubMed]