Fire research abstracts and reviews: Volume 13, 1971 (1971)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

306 F I R E R E S E A R C H Statistics and Economics Buildmg Development and Regulations Structural and Material Fire Tests Extinguishmg Materials and Eqmpment Detection and Electrotechmcs Industrial and Toxic Hazards Technical Inquiries Overseas Activities Library and Pubhcations Lectures and Exhibitions Symposium General Committees Available from Her Majesty's Stationery Office, 49 High Holborn, London, England, Price 95 p TRANSLATION Trabaud, Louis (Phyto-^cologiste, docteiu- en ficologie, C E P.E, MontpeUier, France) "Some Measurements and Observation on the Phytodynamics of Burnt Areas m Bas-Languedoc (Southern France)," Naturalm mmspehenstaf sir. Bot., 21 231 (1970)* In French. Subject: Burnt area phytodynamics Translated by G. L. Ordway Summary In the Mediterranean region fire can be considered as an important factor of the dynamics of vegetation With this ID nund we have made some observa- tion on the regeneration of burnt areas in the Bas-Languedoc (Southern France). Observations were earned out m situ on a burnt-out zone comprising side by side a thicket of kermte oak (Quercus cocafera) and a stand of Aleppo pine (PinMS halepensts) In this zone, soil samples were taken m order to study the contents of seeds that were able to germmate Under natural conditions the area is immediately occupied by preexistmg species endowed with an important capacity of resproutmg from the stump In the samples numerous species, mainly annuals, appear that were not m- ventoned tn sttu * The results presented m this paper were obtamed as part of a study of the action of fires on vegetation in Baa-Languedoc This program is a part of the research of the dynamics of a thicket of Quercus eoccifera planted by the Division of Phytoecology of the Centre d'Etudes phytosocio- logiques et ^cologiques of Montpellier

ABSTRACTS AND R E V I E W S 307 INTRODUCTION Fire IS a defimtive factor m the dynamics of plant life m the woods of the Mediter- ranean region In France numerous authors have studied the fire factor and its devastating effects, notably, for the Midi, Ribbe (1866, 1869), Jacquemet (1907), Ducamp (1932), Laurent (1937) Nevertheless, none of these made precise observations on the development of vegetation after a fire And unfortunately there are few ecological data available on the causes and effects of fires Only a few authors, such as Kuhnholtz-Lordat (1938, 1952, 1958), Braun-Blanquet (1935, 1936), Kornas (1958), have approached the problem with somewhat more rigor Braun-Blanquet descnbed several stages of vegetative degradation, but he did not analyze what determmes the regressive sequence According to him, a number of phenomena, which are generally associated with each other (cuttmg, fire, and pasturage) have brought vegetation to an advanced state of degradation such as can currently be seen m certam areas Kornas mdicates the causes of the retrogressive sequence Comparmg parcels subjected to fires, he tries to describe the series of successive stages, but he does not analyze the actual process of plant dynanucs after the fire He mdicates that with the dominance of Quercus ilex in the forest (which would be accordmg to him the chmax) the cycle of progressive evolution would be closed Kuhnholtz-Lordat, who is one of our best specialists on the subject, describes the various stages of progressive or retrogressive sequences He estimates the approximate time required to reconstitute a stand of Q ilex He names the species (Q. coccifera, Arbutus unedo, Cistus sp , Ptnus sp ) currently best recogmzed as pyrophytes, and indicates the plant characteristics that perimt them to survive and to spread over burnt areas Before proceedmg, we should like to make a brief parenthetical remark on the meamng of the term "pyrophyte," which is actually ambiguous and covers several meamngs. Herewith is a critical review Grand Larousse encyclopidtque (1963) "said of a plant that easily repairs injuries brought about by fire thanks to its mode of reproduction by rhizome " Grand Littri (1962) does not include the term. Carpenter (An Ecological Glossary 1966, New York, Hafner, 306 pp ) • "a tree having a thick fire-resistant bark, which thus escapes damage caused by forest fires" P. Gray (Dictionary of Biological Sciences, 1967, Remhold) "a plant havmg a bark resistant to forest fires " Kuhnholtz-Lordat (L'ecran vert, Mem. Mus Nat Hist Nat, I X , 1958) This m our opmion is an excellent definition of pyrophytes—"There are plants that are more or less resistant to flame, there are even those whose propagation or reproduc- tion appears to be stimulated by fire, these are pyrophytes " He refers to numerous examples of pyrophytes havmg different kinds of resistance One could thus have two large types of pyrophytes. pyrophytes resistmg fire, and others that are not fiire-resistant but whose regeneration would be promoted by fire (e g , the rockroses). The resistant pyrophytes could also either be favored, thanks to their underground organs of survival (kermfes oak), or else not favored.

308 F I B E R E S E A R C H but having two different survival systems, resistant underground (live oak) or resistant aboveground (cork oak). We shall use the defimtion of "pyrophyte" proposed by Kuhnholtz-Lordat. However, we shall add that a true pyrophyte should be one that at the same time is resistant to fire and whose regeneration is promoted by fire In the works cited the authors describe the effects of fires recurrmg at more or less regular mtervals, but they do not follow the development. The descriptions are made startmg with stages that are supposed a prion to belong to the pyrophyte sequence I t would appear from our observations and will be seen later on that burnt vegetation, at least in certain natural situations, rather rapidly regains its structure and composition almost as they were mitially Furthermore, these authors provide no mformation on the variation of intensity of the fire, physical mechanisms, degradation due to heat, or the biological condition of the plants. Abroad, however, studies of this type are rather numerous Among others, Ahlgren (1960) describes the effects of fire on the reproduction and growth of plants m forests of Ptnus banksiana m Minnesota He cites species that arise and their development after the fire, mdicatmg the ecological characteristics and the means of reproduction for each of them Floyd (1966) in Austraha describes the effect of fire on the seeds of pioneer species, particularly woody plants, that germi- nate en masse after a fire. Each species behaves differently with respect to the mechanism of the fire and the temperatures reached. Oppenheimer (1961) re- ferrmg to the studies of Went in the Umted States, and Specht m Austraha, states that raised temperatures stimulate the germination potential of many seeds, which in fact are akeady very resistant to heat. I n France the absence of exact data on plant regeneration immediately after the passage of a fire has encouraged us to undertake some preUmmary observations on a recently burned area. Description of the Burnt Area On June 16, 1967, a fierce fire burned 15 hectares of oak thicket and of pine stands in the commune of Buzignargues (H^rault) at a place called Pic du Prieur The fire startmg from the edge of a vmeyard, grew rapidly worse and chmbed to the top of the hill. The grades, variable but fairly regular, are of the order of 7 to 15 percent The exposure is S and SW. The substrate is marl and hmestone. The soil is sparse of a brown hmestone type We did not make a survey of the vegetation before the fire. Two types of plant formation cover the hill—at the foot on the marl a stand of pme of Pmu^ halepensis,* which rmged the hill, and on the hmestone at the top a thicket of Qttercus cocafera. The principal species in the pine stand that were recogmzable after the fire were Pmus halepensis, Rosmarinus officinalis, Lavandula latifolia, Juniperus oxycedrus; those at the top were Quercus coccifera, Quercm ilex, Juniperus oxycedrus, Pmus halepmsts The most frequently found grass in the two groups was Brachypodium ramomm In the pmes the top stratum was formed of Aleppo pmes alone. These were about 8 m high and had a coverage of 60% (crowns). Thiii undergrowth oc- cupied about 40% of the surface. The thicket of kermfes oak was dotted here and * The specific names are those given in Les Qucdre Flares de France by P. Foumier (LechevaUier, 1961, 405 pp).

ABSTRACTS AND R E V I E W S 309 there with Aleppo pines about 4 m tall but of low coverage smce the kermfes oak occupied 95% of the surface Methods of Observing the Regeneration of Vegetation We used two methods to follow the process and regeneration (1) the first consisted of periodic observations made directly on the burnt areas, (2) the second consisted of laboratory observations of the germmation of seeds contained in samples from layers of surface soil taken from the burnt plots and put under glass In order to follow the conditions of regeneration on the ground, plant censuses were made on three successive dates June 30,1967 (14 days after the fire), Septem- ber 18, 1867 (three months after the fire), and April 23, 1968 (ten months after the fire). The surveys were made at sites immediately adjacent to those from which the samples had been removed. On June 20, 1967 (four days after the fire) we gathered m each mitial population two soil samples of a quarter of a square meter (50 cmX50 cm) and 5 cm thick, these samples were immediately put m a greenhouse m large plastic trays on a bed of fine sterile sand. The trays were misted regularly with ordmary water. The first gerimnations took place June 26. Few m number at first, they became very numerous m the course of the first year and then decreased Towards the end of the penod of observation the soil was turned to let the last seeds germmate. The last germmation took place August 6,1968 (i e , 14 months after the samplmg). The experiment was ended the end of October 1968. The count of the germmated seeds was made m total for the two samples from each location I t is important to emphasize that only plants coming from seeds could develop m the trays, because we took the precaution of not transplanting roots or rhizomes; this would have allowed certam perenmals to develop by vegetative reproduction. T A B L E 1 Survey of species of the pinewood and oak thicket Type regen- Pinewood Oak thicket eration after 9/18/67 4/23/68 9/18/67 4/23/68 Type fire Phillyrea angustifolia X X X X V R Dorycnium suffrultcosum X X X X V R S Brachypodtwn ramosum X X X V R s Rubia peregnna X X X V R s Asparagus acuhfohtis X X X X V R Ptstacia lentiacua X X X X V R Quercus coccifera X X X X V R Fwnana condifolia X X V R s Leuzea comfera X X X V R s Buphurum ngtdum X X X X V R s Smilax aspera X X X X V R Genista scorpiua X X X V R Rhamnua alatemits X X X V R

310 F I R E R E S E A R C H T A B L E 1 (CorUtnued) Pmewood Oak thicket 9/18/67 4/23/68 9/18/67 4/23/68 Type Type regen- eration after fire Carex hallenana X X X X V R S AphyUanthes monspeliensts X X X X V R Quercus ilex X X X V R Lontcera tmplexa X X V R Aristolochia pistolochta X X X V R S Sonchus oleraceus X X A s Scorpiurus subtnllosus X X A s Argyrolobium hnnaeanum X X V s Thymus vulgaris X X V R s Crstus monspeliensis X X V s Psoralea httumtnosa X X V R Brachypodium phoenicoides X X V R Euphorbia serrata X V s Rubus vlmtfoltus X X V R Tonlis arvensis X A s Ononis mtnuhssirrui X V R s Rosmarinus officinalis X V s Pinus haleperms X V s Anagallts arvensis X A s Lagosens sancta X A s Eryngium campesire X V s Shoenus nigricans X X V R Sangmsorba minor X V s Crepis laraxacifoha X V s Gahum sp. X A s Seseh elalum X V s Muscan racemosum X V R s Bromus erectus X V R Hteracium murorum X X V R Cons monspeliensis X V R Andropogon ischaemum X V R s Daphne gnidium X X V R Euphorbia exigua X A s Carex glauca X V R Clematis flammvla X X V R Sonchus ienemmus X A s Hieracium piloseUa X V R Narcissus juncifolius X V R s Seseli monlanum X V R Juniperus oxycedrus X V R Festuca omna X V R Teucnum chanuiedrys X V R Staehelina dubia X V R Allium sp X V R s The species are listed in order of decreasmg frequency Type V=peremual, A=aimual. Type of regeneration after the fire R=regrowth m general from the stump, S = seeds

ABSTRACTS AND R E V I E W S 311 OBSERVATIONS MADE AT THE SAMPLING SITES—TABLE 1 June 30, 1967 (14 days after the fire) no plants had resprouted or gerrmnated September 18, 1967, only a few perenmals—those among the most abundant— had resprouted I t is noteworthy that i t was the perenmals alone that reoccupied the free space On April 23, 1968, the pmewood had 50 species of perennials and the oak thicket only 30 species In Table 1 the symbols R and S represent the type of regeneration after the fire R, regrowth, in general from the stalk, S, seed I t appears of mterest to compare the two regions by keeping track of the type of regeneration of the species after the passage of the fire Three classes have been considered (a) perenmal species in the strict sense that they are able to regenerate by regrowth but also by seed, (b) perenmal species that can only regenerate by seed, (c) annual species in the strict sense Of the 50 species m the pmewood, 33 are perenmals in the strict sense (66%), 9 are seeding perenmals (18%), and 8 are true annuals (16%) The perenmals of either kind are thus very numerous (more than half of the species present), but the proportion of species able to regenerate by seed alone, both annuals and peren- mals, 34% IS nonetheless higher for the pme than for the kermfes oak thicket In fact, the oak has for the 30 species present, 26 true perenmals (86 6%), 2 seedmg perenmals (6 7%) and 2 annuals (6 7%). There are thus very few annuals and seedmg perenmals This is probably due to the competition caused especially by the kermis oak itself, which has exceptional regenerative powers and which re- sprouts qmte vigorously. The large proportion of species resproutmg should be noted 70% of the total of the species encountered m the two regions Nevertheless only 66% of the species m the pmewood resprouted as opposed to 86% of species regrowmg from stumps m the oak thicket Most of the perennial, woody, and suffrutescent species can resprout. Of these only Argyrolobium hnnaeanum, Cistus monspeliensts, Pmus halepensis, and Ros- marinus officinalis reproduce by seeds alone On this account, they appear only m the spring In general, the perenmal species that reseed produce a large number of seeds, which permits them to fill m bare spaces rapidly. This does not appear to be the case here, on the other hand m an open stand of kermfes oak and rockrose on gravel and marl at Castelnau-de-Guers, we were able to count up to 4,000 seedlmgs of Cistus monspeliensis per m'. Perennial species have the capabihty of reproducmg after a fire either by re- sproutmg or by seedmg Thus some species that can regenerate immediately by resproutmg rapidly colomze the area and ehrmnate by competition foreign volun- teer species that would attempt to occupy the space The case of the kermfes oak thicket IS typical 86% of the species counted were able to regenerate immediately by resproutmg, the thicket thus reconstituted itself just as i t was before the fire This IS a very important observation for an exact understandmg of plant dynamics.

312 F I R E R E S E A R C H T A B L E 2 Number of seedhngs by species and by type (under glass) V=perennial, A=annual, and species normally absent in the plant population at the site Pinewood Oak thicket Ptnus Quercus Species halepensus coccifera Type Engeron naudim* 91 146 A Sonchus lenemmm* 36 43 A Engeron canadense* 8 21 A Cistus monspehensts 6 27 V Rosmarinus officinalts 6 19 V Crepis taraxacifoha* 1 1 V Cenlaunum pulchellum* 63 A Fumana condijolia 54 V CKUrra perfoltata 51 A Thymus wlgans 3 V Sondius oleraceus* 3 A Andropogon xschaemum 2 V Brachypodium ramosum 6 V Carex haUenana 5 V AUtum sp. 4 V Scorpiurus submllosus 2 A Dorymtum suffruhcosum 2 V Pmus halepensis 2 V Argyrolobium linnaeanum 1 V Rubta peregnna 1 V Epilohium tetragonum'*' 1 V Total 323 281 OBSERVATIONS MADE ON THE SOIL SAMPLE PUT UNDER GLASS- TABLE 2. Table 2 gives a list of the species enumerated and of the number of plants per species accordmg to region. Twelve species germinated for the pinewood sample and fifteen for the oak However, if one considers the number of seedhngs, the pme had many more seeds that gernunated (323) than the oak (281). Thus, though havmg fewer species, the pinewood had a higher number of viable seeds. Apart from some annuals (Erigeron naudini, E canadense, Sonchus tenernmus) and biennials (Crepis taraxacifolia) common to the two groups, which are species not normally found at these sites, but which are foimd m large numbers, i t is said, after fires (cf Floyd, 1966), there are only two common perennials Cistus mon- speliensis and Rosmarinus officinalis I t would appear that Rosmarinus officinalis is not peculiar to one of the two groupmgs rather than the other, on the other hand, Cistus monspeliensis is more frequent m the thicket of Querciis coccifera.

ABSTRACTS AND R E V I E W S 313 The remaining species, which are not common to the two groups, show that we have here two different plant communities. The coefficient of similarity of the two groups, determined by considering the species of seedhngs produced, is 28.6%, i t is thus fairly small The two lists of species produced from the soil samples do not have much in common In order to explam the germination, two possibihties can be considered The seeds of annuals as well as perennials could have been carried m durmg the four days between the date of the fire and the date of samplmg, or what is more likely, these seeds could have been stored in the soil (cf Guyot, 1968) and been un- affected by the passage of the fire In the wood of Ptnus halepensis, the total number of species that germinated in the course of the experiment is twelve, of which six were annuals and six peren- nials, giving a ratio of A/P = 6/6 = 1, the annuals represent 50% of the germinatmg species and there are as many annuals as perenmals I f one weights the calculation by taking mto account the number of seedhngs of each species, the variation of the numbers is more important The total number of seedhngs is 323, of which 252 are annuals and 71 perenmals, givmg a ratio of the number of germinations (?VGp = 252/71=3 54. The percentage of annual seedhngs is 78 02%, while that of perennials is only 21 98% One can thus see the preponderance of annuals This is not surprising, since the fire leaves open spaces for plants to grow, and annual species have a great abihty to produce seeds with a high germinating capacity. Furthermore, the pinewood is a population of low stabihty In addition, the greenhouse species were not subjected to intraspecies and interspecies competition, so that the number of annual species is very important, since as has been seen elsewhere, in a natural medium there is competition that reduces the expansion of the annuals to the benefit of the perennials, which are apt to develop rapidly by resprouting from the stumps. In the oak thicket the number of species germinatmg is larger than m the pine- wood—namely, 15, of which four were annuals and eleven perenmals. The ratio 4 / P = 4 / l l = 0 . 3 6 , and the respective percentages are 26 6% annuals and 73 3% perennials In distmction to the pmewood, the perenmal species are predominant. However, the number of seedhngs though different from the pmewood, is even higher for the annuals. Total number of seedlmgs, 281, number of annual seedhngs, 212, number of perenmals, 69 The coefficient (?4/(?, = 212/69=3 07, or 75 44% of annuals against 24.55% of perenmals. Comparison of the Results Obtained from Experiments under Glass and the Surveys Made at the Samphng Sites I f we compare the coefficients of similarity of the results obtamed by the ex- periments under glass and the surveys made on the sites, in which we not only account for those species able to reproduce from seed, we can say that there is a rather large connection between the samples and the surveys from each group. In other words, the similarity is greater between the samples coming from the pinewood and the survey of the pmewood than between the samples from the pines and the survey of the oak, and the similarity is greater between the samples

314 F I R E R E S E A R C H from the oak and the survey of the oak than between the samples of the oak and the survey of the pme (Table 3) T A B L E 3 Values of similarity coefficients between the experiments under glass and the surveys (Species reproducing by seed) Study under glass of pine- wood samples Study under glass of oak- thicket samples Survey of oak thicket Survey of pinewood 24 2 (8) 33 3 (11) 41 9 (13) Survey of oak thicket 17 3 (4) 36 8 (8) Study under glass of oak thicket 28 6 (6) samples The numbers m parentheses are the numbers of species m common However, the coefficient of similarity of the oak samples and the pine survey (33 3) appears to be abnormally high In addition, the coefficient of similarity of the pine survey and the oak survey (all species) is 40 3 This is rather high, in fact the two surveys have 23 species in common, and, what is more, if one com- pares the surveys m each case, only takmg account of species that reproduce by seed, the coefficient becomes 41 9 I t thus appears, even more strongly for the burnt areas (surveys) than for the samples under glass, that the connection is rather large between the two groups From the plant population view, the two groups belong to different formations, but they are neighbors ecologically, whence the presence of certam species com- mon to the two. The pmewood pertains to the Rosmarmeto-Lithospermetum pmetosum B r - B l 1924 and the oak thicket to the Cocciferetum rosmaretosum Br - B l . 1924 In addition the pinewood is less stable and has a tendency to evolve towards the chmax of the Quercetum ilicis (cf the species of shrubs Phtllyrea angusttfolm, Quercus ilex, Rubia •peregrina, etc ) and on this account has species that will be the same as those of the oak thicket Discussion The two plant groups are not too greatly different from each other, but they react differently to the action of fires The number of species appearmg on burnt areas is much greater for the pme than for the oak In addition, the proportion of annuals is much greater for the pme than for the oak This is connected with the fact that the competition between species and between individuals is not as strong m the pmewood, because there are fewer species that resprout from stumps and that thus very quickly fill the area On account of this, a rather significant amount of space is available for the growth of foreign species In fact, the propor- tion of species that propagate only by seed is higher for the pinewood (34%) than for the oak (13%)

ABSTRACTS AND R E V I E W S 315 However, the proportion of perennial species reproducing vegetatively is so important in the two groups that a series of intermediate transition stages is not observed in the hwrni areas, although this would often be the case (Ahlgren, 1960, Floyd, 1966). Here, the species present before the fire are the ones repopu- latmg the space left open after the passage of the fire. This can be attributed to mterspecies competition and to the immediate occu- pation of free space by species that send forth a large number of shoots In fact, in the absence of this competition and repopulation by shoots (the case of the samples removed to a greenhouse), one finds that the number of annuals, mcludmg species that do not belong to the two groups, can be qmte high. But also the favoring of the original species is due to the fact that the soil is a buffer agent that protects the underground parts of the plants found there (rhi- zomes, roots, tubers) In fact, when prehminary experiments were made to study the action of fire on the kermfes oak thicket, we used thermocouples to measure the temperatures reached durmg fires that were set, at the level of the plant mass, temperatures reached 750''C or even 800''C, on the other hand, the heat released by the burning had httle effect on thermocouples buried in the soil at a depth of 10 cm, there was no rise m temperature, while at 5 cm the mean temperature was 40''C This temperature of 40°C is significantly below temperatures that kill plant tissues, and moreover i t only lasted for a short time and did not have time to kill the regenerative organs One can even assume that i t served to activate the eventual germmation of planted seeds. Thus i t is, thanks to the protective action of the top centimeters of soil that keeps seeds and underground organs ahve, and also the great capacity of most of the species to resprout from stumps, that we have both germination under glass and the fillmg of bare and free areas on the site by the origmal species. Conclusions The survival of seeds and the special ability of a large number of species to put forth shoots are the two prmcipal factors that have played a role in the reoccupa- tion of the burnt areas that we have studied I t is especially the growth of the origmal species that has permitted a rapid regeneration of vegetation m the burnt zones Tlus fillmg of the area is determmed by the plant dynamics, and i t must be emphasized that there is a natural tendency to create rapidly an environment identical to the origmal one by impedmg the growth of foreign species One can consider that the two plant formations studied are m fact relatively stable commumties and not transition stages The groups as they exist now are m equihbrium with the environment, so much so that one can assume that fire acts as a natural factor Other authors have on the other hand described stages, one can assume when this is the case that the species present before the fire do not have a great capacity for reoccupymg the devastated area and that these plants do not vigorously put forth new shoots Hence there is a possibihty that new and transitional species will grow for a certain time For kmds of vegetation strongly affected in the past by the recurrence of fires, which we appear to have m general m our region, one can assume that the present