- Email: [email protected]
Litter production and decomposition from shrubs of Protea repens growing in sand plain lowland and mountain fynbos, south-western Cape
S. Afr. J. Bot., 1987, 53(1)
25
Litter production and decomposition from shrubs of Protea repens growing in sand plain lowland and mountain fynbos, south-western Cape D.T. Mitchell* and P.G.F. Coley Department of Botany, University of Cape Town, Rondebosch, 7700 Republic of South Africa *Present address: Department of Botany, University College, Dublin 4, Eire
Accepted 1 August 1986 Litter production and decomposition of mature shrubs of Protea repens (L.) L. growing in coastal and mountain fynbos , south-western Cape, were studied during 1981 -1983. Mean annual litter production under the canopies ranged from 200 to 450 g m - 2 . Leaf fall accounted for 50-87% of the total with floral bracts being the next dominant category. November- January was the main period of litter release. Winter (June- August) leaf litter contained the lowest soluble carbohydrates and highest lignin levels. The phosphorus and nitrogen concentrations of freshly fallen litter were less than that of attached leaves and the decline ranged from 33 to 66% and 3 to 25% respectively. Leaf litter decomposition was not seasonal and no differences were encountered in the decay of dry mass, lignin, cellulose and soluble carbohydrates in the decomposing leaf litter from the two study areas. Leaf litter decomposition turnover times of P. repens are longer than those of sclerophyllous shrubs of other mediterranean-type ecosystems. Produksie en ontbinding van plant-afval van volwasse Protea repens (L.) L.-struike van kusfynbos en bergfynbos, suidwes Kaap, was bestudeer gedurende 1981- 1983. Die gemiddelde jaarlikse plantafval-produksie onder die plantkrone was tussen 200-450 g m - 2. Blaarval het 50-87% tot die totale plantafval bygedra en blom-skutblare was die volgende dominante kategorie. November- Januarie was die belangrikste peri ode vir plantafval-vrystelling . Gedurende die winter (Junie- Augustus) het die blaarafval die laagste oplosbare koolhidrate bevat en die hoogste lignien-inhoud gehad. Die fosfor- en stikstof konsentrasies van pasgevalle plantafval was minder as die van aangehegte blare en die verval het van 33 tot 66% en 3 tot 25 % onderskeidelik gewissel. Blaarafval-ontbinding was nie seisoenaal nie en geen verskille in die afbraak van die droe massa, lignien, sellulose en oplosbare koolhidrate was in die ontbindende blaarafval van die twee studie-gebiede ondervind nie. Die omsettye van blaarafval ontbinding vir P. repens is Ianger as die van sklerofiele struike van ander mediterreense-tipe-ekosisteme. Keywords: Fynbos , leaf decomposition, litter production, Protea repens , sclerophyllous shrubs •To whom correspondence should be addressed
Introduction Mature stands of sand plain lowland and mountain fynbos (Moll et a/. 1984) are often dominated by members of the Proteaceae, such as Protea repens (L.) L. which grow in acidic sandy soils of a low nutrient status (Mitchell, et a/. 1984; Stock & Lewis 1986). These slow-growing evergreen species have adapted to the low nutrient conditions by absorbing and assimilating small quantities of inorganic nitrogen (Stock & Lewis 1984). Another adaptation is to conserve nutrients and thus produce a poor quality of litter (Read & Mitchell 1983) but there is no evidence that this occurs in shrubs of P. repens. In this study on P. repens, the objectives were to: (a) quantify litter production and decomposition rates, (b) quantify litter quality and determine whether nutrients are withdrawn during leaf abscission and (c) test whether seasonal variations in litter production and decomposition differ between individuals of the same species growing in different areas. Litter quality was assessed by analyzing organic and inorganic constituents, and the decay rates of specific organic components and phosphorus in decomposing leaf litter were estimated. Study Area The two study areas were sand plain lowland fynbos at Pella (33 °31'S, 18°32'E; 160-220 m altitude; 269 ha) 62 km north of Cape Town and mountain fynbos at Swartboschkloof (33 °59'S, 18°57'E; 300-940 m altitude; 375 ha) forming part of the catchment of the Eerste River 48 km east of Cape Town. Both are in a winter rainfall region (Fuggle 1981) but Pella has a mean annual rainfall of 571 mm compared with 1624 mm at Swartboschkloof. Pella has a mean maximum and minimum temperature of 31 o and 8°C respectively whereas at Biesievlei (a station near to Swartboschkloof;
33°58'S, 18°57'E, 282m altitude), the means are 22° and l0°C respectively. At Pella, the shrubs of P. repens were aged at approximately 22 years old during 1981 and are situated in a 1-ha area along the eastern boundary in Longlands soil classified according to MacVicar eta/. (1977). This soil is a medium/fme sand approximately 2 m deep and is usually waterlogged during Winter. At Swartboschkloof, the P. repens shrubs were growing in an area burnt on 2 February 1958 (Van Wilgen 1982; Van der Merwe 1966) and were selected from a 10-ha area at an altitude of 305- 366 m. The soil was the Clovelly form , Mossdale Series classified according to Mac Vicar eta/. (1977), and the B horizon has a 07o particle size distribution of 9,6 (coarse sand), 47,3 (medium sand), 32,7 (fine sand), 3,3 (silt) and 7,0 (clay) (Lambrechts eta/. 1986). Shrub height at both sites was about 2,5 m.
Methods Collection of litter Litter was collected in traps made of polyethylene shade netting attached to a circular frame of galvanized hoop iron (0,25 m 2) suspended 0,5 m above ground on a three-legged steel rod frame and the bottom of the net was drawn closed by polyester cord. The traps were positioned halfway between stem and perimeter of the canopy of 10 randomly selected shrubs at each site (one trap per shrub). Litter was collected at monthly intervals from March 1981 to August 1983, divided into leaves, floral bracts, seed and twigs and oven-dried at 80°C. Litter decomposition Recently fallen leaf litter was collected and 5 g of air-dried mass was sealed into bags (150 x !50 mm) of vinyl-coated
S.-Afr. Tydskr. Plantk., 1987, 53(1)
26 fibre-glass (2,25 mm 2 mesh size). The litter bags were placed on the ground under the canopies of 14 shrubs at each site with 10 bags per shrub on 1 May 1981 at Pella, and 1 March 1981 at Swartboschkloof. Every 2 months, five bags (one per shrub) were removed and oven-dried at 80°C.
Chemical analyses During November 1983, l-year-old fully expanded leaves were removed from the shrubs and these and the leaf litter were
ground through a Wiley mill (60 mesh). The ground samples were then chemically analyzed using the methods of Allen et al. (1974), Schlesinger & Hasey (1981) and Mitchell et al. (1986). For inorganic analyses of Ca, Mg, K and Fe, 1 g of oven-dried, ground material was ashed in a muffle furnace at 550°C for 3 h, dissolved in HF/HClIHN0 3 (1:8:4) and concentrations of each element were determined on a Varian No 6 atomic absorption spectrophotometer. For the phosphorus detenninations, 0,1 g oven-dried material was digested
110 100 90 80 70 60 50 40 30 20 10 N
I
E
.9 U) U)
co E ~
0
110 100 90 80
50 40 30 20 10
AMJ
J
ASONDJ
FMAMJ
1981
J
ASONDJ
1982
FMAMJ
J
AS
1983
Months
Figure 1 Seasonal variations in total litter production from shrubs of Pro tea repens growing in lowland fynbos at Pella (e--e) and mountain fynbos at Swartboschkloof (-------) during 1981 - 1983. Vertical bars represent twice SE.
S. Afr. J . Bot., 1987,53(1)
27
by the method of Jackson (1958) and then assayed using the method of Murphy & Riley (1962). Total nitrogen was determined on 0,1 g oven-dried material using standard Kjeldahl procedures and selenium catalyst and sodium thiosulphate. The N~ + -N was then determined colorimetrically using the method of Smith (1980). The sample sizes for the organic constituents were 0,2 g oven-dried material. Petroleum-ether soluble fats and waxes were determined gravimetrically. Methanol-soluble (50070) carbohydrates were analyzed colorimetrically by the method of Dubois et at. (1956) with D-glucose as the standard. Cellulose was extracted by a delignification procedure with the addition of sodium chlorite and 10% acetic acid at 75°C (Allen et at. 1974) and the carbohydrate content was analyzed by the Dubois et at. method (1956). The lignin fraction was extracted and determined by methods similar to King & Heath (1967) and Schlesinger & Hasey (1981).
Statistical analyses The litter fall data were grouped into four seasons: autumn (March - May), winter (June - August), spring (SeptemberNovember) and summer (December - February), and analyzed by one-way analysis of variance. The results between the two study areas were compared by the Student's t test. In the litter decomposition studies, the analyses of the various constituents were replicated five times and the mean concentration was multiplied by the fraction of the original litter mass remaining. This product was expressed as a percent of the original content of each constituent in the fresh litter. Exponential decay relationships were calculated for each constituent. Overall fractional loss rates (k) were calculated from the formula: natural log (X/Xo) = - kt where Xl and Xo are the dry masses remaining at time t and time respectively (Olson 1963). Instantaneous fractional loss rates (kl) were also calculated from the formula: kl = 1 - e - kt which
°
Table 1 Annual litter production (g m- 2) and seasonal variations (g m- 2 month- 1 ± SE) in the major components of litter fall under the canopies of Protea repens at Pella and Swartboschkloof Pella Floral bracts
Seed
Leaves
Floral bracts
Seed
28,5 75,1
12,7 16,1
388,1 284,2
9,0 53,0
9,2 26,8
5,9 ± 2,2
1,1 ± 0,2
16,4 ± 4,5
1,5 ± 1,1
0,9 ± 0,1
3,2 ± 1,6
0,2 ± 0,1
15,1 ± 2,0
4,0 ± 1,6
2,4 ± 1,2
5,2 ± 1,3
2,3 ± 0,6
28,0 ± 5,8
3,4 ± 1,5
1,2 ± 0,4
3,2 ± 1,4 0,67 3;25 N.S.
1,3 ± 0,3 8,87 3;25 0,001
55,5 ± 6,6 15,77 3;26 0,001
1,1 ± 0,5 1,10 3;26 N.S.
1,1 ± 0,1 1,00 3;26 N.S.
Leaves Annual litter production 1981 - 1982 207,4 1982-1983 205,4 Seasonal litter production Autumn (Mar - May) 10,0 ± 2,9 Winter (lune - Aug) 4,8 ± 0,5 Spring (Sept- Nov) 20,4 ± 9,5 Summer 31,9 ± 6,6 (Dec -Feb) F 5,54 d.f 3;25 p:;::; 0,01
Swartboschkloof
N.S. denotes no significant seasonal variation
Table 2 The organic constituents of attached mature leaves collected during November 1983 and seasonal variations of the organic constituents of the leaf litter of Protea repens at Pella and Swartboschkloof Pella Soluble fats & waxes
Swartboschkloof
Soluble carbohydrates
Cellulose
Attached Leaves 19,4 ± 0,1 116 ± 12 Leaf Litter Autumn (Mar - May) 10,8 ± 1,2 130 Winter (lun - Aug) 11,6 ± 0,8 101 Spring (Sep - Nov) 7,1 ± 0,6 152 Summer (Dec - Feb) 16,0 ± 2,1 149 F
6,7 d.f 3;69 P:;::; 0,001
Lignin
Soluble fats & waxes
Soluble carbohydrates
Cellulose
Lignin
74 ± 1
186 ± 6
19,3 ± 0,2
68 ± 3
84 ± 1
214 ± 4
± 6
95 ± 4
219 ± 6
10,0 ± 0,8
116 ± 5
106 ± 4
239 ±4
± 5
97 ± 5
407 ± 8
11,5 ± 1,0
95 ± 6
104 ± 7
407 ± 11
± 5
72 ± 6
221 ± 20
6,7 ± 0,5
116 ± 6
113 ± 7
281 ± 18
± 8
80 ± 5
164 ± 5
9,9 ± 1,0
118 ± 6
99 ± 8
216 ± 7
5,30 3;66 0,01
91,37 3;61 0,001
15,84 3;66 0,001
Results are expressed as mg g -
I
6,15 3;86 0,001
3,87 3;85 0,05
0,83 3;86 N.S.
dry mass ± SE and N.S. denotes no significant seasonal variation
51,11 3;71 0,001
S.-Afr. Tydskr. Plantk ., 1987, 53(1)
28
and the second dominant category was floral bracts (11 26%). Seeds (5 - 7% of the total annual litter) were released from the flowerheads. Twigs (0,0-1,4%) formed the smallest category of litter. The timing of litter fall at Pella and Swartboschkloof produced a similar major peak between November and January (Figure 1). Leaf fall showed significant seasonal variations, whereas of the other litter categories, only seed release from shrubs at Pella varied significantly (Table 1). A significantly greater release of leaf litter from shrubs at Swartboschkloof compared with those at Pella occurred during winter (d.f = 18; P < 0,001) and summer (d.! = 10: P < 0,05). The soluble carbohydrates of attached mature, l-year-old leaves on the shrubs at Pella were higher than those at Swartboschkloof (d.! = 8; P < 0,001), whereas the concentrations of cellulose (d.! = 8; P < 0,001) and lignin (d.! = 8; P < 0,01) were lower (Table 2). The leaf litter released during the winter months contained the lowest soluble carbohydrates and had the highest lignin concentrations (Table 2). Significant differences in the specific organic components of freshly fallen litter between the two study areas occurred during summer (d.! = 25; P < 0,05) for fats and waxes, spring (d.! = 32; P < 0,001) and summer (d.! = 28; P < 0,01) for soluble carbohydrates, spring (d.! = 33; P <: 0,001) for cellulose, and autumn (d.f = 33; P < 0,01), spring (d.! = 33; P < 0,05) and summer (d! = 33; P < 0,001) for lignin concentrations. Of the inorganic constituents, only nitrogen and phosphorus concentrations of the attached mature leaves and the freshly fallen litter were compared. There was a significant difference in nitrogen concentration of attached leaves between the two study areas and its decline prior to leaf fall at Pella (25%) was greater than that at Swartboschkloof (3%) (Table 3). The phosphorus concentration dropped by 33 - 66% in the freshly fallen litter compared with concentrations in the attached mature leaves (Table 3). The concentrations of phosphorus, nitrogen, potassium, calcium, magnesium and iron in the abscissed leaves at Pella and Swartboschkloof were similar (Table 3).
is useful when comparing decomposition rates for labile constituents in long-term studies (Olson 1963). Turnover times for decomposition (years) were calculated from the reciprocal of k and half times were estimated from 0,693/k (Olson 1963). Results Litter production Shrubs of P. repens at Pella had the lower annual litter production releasing 256 and 297 g m - 2 during 1981 - 1982 and 1982 - 1983 respectively, compared with 448 and 364 g m - 2 from shrubs at Swartboschkloof. Leaf litter was the major category consisting of 69 - 87070 of the total annual litter
Table 3 Inorganic constituents (mg g-1 dry mass ± SE with number of analyses in parentheses) of attached and abscissed leaves of Protea repens at Pella and Swartboschkloof Swartboschkloof
Pella Phosphorus Attached leaves Abscissed leaves Autumn (Mar - May) Winter (lun - Aug) Spring (Sept - Nov) Summer (Dec-Feb)
0,1 0,2 0,1 0, I
Nitrogen Attached leaves Abscissed leaves
6,8 ± 0, I (5) 5,1 ± 0,5 (15)
6,1 ± 0,1 (5) 5,9 ± 0,3 (14)
Potassium Abscissed leaves
1,0 ± 0,2 (4)
1,1 ± 0,3 (5)
Calcium Abscissed leaves
7,0 ± 2,7 (4)
7,3 ± 2,2 (5)
Magnesium Abscissed leaves
1,8 ± 0,6 (4)
1,6 ± 0,4 (5)
Iron Abscissed leaves
0,9 ± 0,3 (4)
0,6 ± 0,2 (5)
0,3 ± 0,01 (5)
± ± ± ±
0,01 0,01 0,01 0,01
(20) (15) (15) (14)
0,3 ± 0,02 (5) 0,1 0,2 0,1 0,1
± ± ± ±
0,01 0,01 0,01 0,01
(20) (15) (20) (18)
Table 4 Decomposition constants for leaf litter from Protea repens at Pella and Swartboschkloof during a 2,5-year litterbag study
Pella Dry mass Soluble fats & waxes Soluble carbohydrates Cellulose Lignin Phosphorus Swartboschkloof Dry mass Soluble fats & waxes Soluble carbohydrates Cellulose Lignin Phosphorus
-k
_kl
Halftime (yr)
Turnover (yr)
n
r
P
-0,05
-0,05
14,14
20,40
15
-0,50
0,05
-0,08
-0,08
8,26
11,90
9
-0,27
N.S.
-0,49 -0,46 +0,10 +0,25
-0,39 -0,37
1,41 1,51
2,04 2,18
12 10 10 14
-0,82 -0,62 +0,81 +0,57
0,001 0,05 0,01 0,05
-0,D7
- 0,07
10,35
14,94
16
-0,83
0,001
-0,44
-0,36
1,58
2,28
10
-0,80
0,01
-0,59 -0,03 +0,07 +0,11
-0,45 -0,03
1,17 20,92
1,68 30,18
12 10 12 15
-0,69 -0,13 +0,58 +0,41
0,01 N.S. 0,05 N.S .
k is calculated from (x/Xo) = e - kt; kl = I - e - kt; Half-time = 0,693/ k turnover of means of five separate samples and r is correlation coefficient
=
Ilk; n is number
S. Afr. J. Bot., 1987 , 53(1)
29
Decomposition There was no apparent seasonal pattern in the decomposition of leaf litter. Turnover times of dry mass, fats and waxes and soluble carbohydrates of decomposing leaf litter from Swartboschkloof were shorter than those at Pella (Table 4). The slopes of exponential decay relationships of dry mass, lignin, cellulose and soluble carbohydrates of decomposing leaf litter at the two study areas were not significantly different. The breakdown and release of cellulose from the decomposing leaf litter at Pella was similar to soluble carbohydrate release whereas the cellulose composition of the decomposing leaves
at Swartboschkloof showed no significant decline (Figure 2). Lignin composition of the litter at the two sites increased during decomposition (Figure 2; Table 4). The phosphorus concentration of the decomposing leaf litter at Pella increased whereas no significant changes occurred in the litter at Swartboschkloof (Table 4).
Discussion The mean annual litter production from mature shrubs of Protea repens growing in sand plain lowland and mountain fynbos fell within the range of 200- 450 g m - 2 • This is similar
U) U)
"'E
90
80
..:---·---.---.---- ---------
A
...
.A.
I
70
• .A..----.•t- ---~--·---~---· ---~---!! I • f·-t---·--·----
.A.
c c Ol
:.:;
•
•
•
. ------- _, ___ ~!"_ ..... -- ·· -· .... -- -~-- ·-·-·. --.- . .11!11_. -- .. . -- . -- .. .............. --.80 OJ
70
U)
c c0 OJ
u
0 :J
a; u
60
•
•
50
(ij
c
40
Ol
8
30
0
c
20
OJ
u
(j;
a.
U)
OJ
'§ '0
>-
.<::: 0
70
£J
60
OJ
50
0
40
:uu
:a:J UJ
30 20 10
... ...
--- -- . • ·-- --• • ------.----------------Duration of incubation (Mon ths)
Figure 2 Changes in dry mass , lignin, cellulose and soluble carbohydrates of decomposing leaf litter of Protea repens growing in lowland fynbos at Pella ( A. ) and mountain fynbos at Swartboschkloof ( • ). Curves are the best fitting exponential decay relationships for decomposing leaf litter at Pella ( - - ) and Swartboschkloof (---).
S.-Afr. Tydskr. Plantk., 1987,53(1)
30 to the results of Leucospermum pari/e (Salisb. ex J. Knight) Sweet, another proteoid shrub at Pella (Mitchell et al. 1986), shrubs of other mediterranean-type ecosystems (Mooney et al. 1977; Lee & Correll 1978; Gray & Schlesinger 1981) and stands of Pinus radiata D. Don adjacent to the mountain fynbos site in this study (Versfeld 1981). The main category of above-ground litter from proteoid shrubs is leaves, but the next important one varies with species. In P. repens in this study, the flowerheads remain on the shrubs but the seed is weakly serotinous (Bond 1985). In L. pari/e, flowerheads are an important category and are released at a different time i.e. November to leaf fall (which is February - March) (Mitchell et al. 1986). Leaf fall and shoot growth of sclerophyllous shrubs, in particular examples belonging to the Proteaceae, occur simultaneously during spring - summer (Specht et al. 1983). The seasonal release of litter from vegetation of mediterranean-type ecosystems may be controlled by high summer temperatures and low summer rainfall (Lee & Correll 1978; Meentemeyer et al. 1982). In the south-western Cape, the south-easterly winds prevailing during the summer may affect above-ground litter fall (Mitchell et al. 1986). In this study, the period of litter fall from P. repens at Pella (a drier site and warmer during summer) and Swartboschkloof was the same occurring during late spring/summer. Litter fall from sclerophyllous shrubs and trees follow a simple harmonic function (Ashton 1975; Maggs & Pearson 1977; Lee & Correll 1978). In P. radiata adjacent to the mountain fynbos site in this study, a harmonic function of needle fall was demonstrated with peak production in March and male cone production occurring during July - September (Versfeld 1981). The factors which influence the different phenophases such as the timing of litter fall are numerous and whether the cues are determined by environmental or endogenous rythms are open to speculation (Pierce 1984). Leaves of P. repens have sclerophyllous characteristics in being low in organic constituents but having high concentrations of polymeric organic compounds such as cellulose and lignin (Swift et al. 1979). Cellulose and lignin would form the main carbon composition of the litter and CN ratios of freshly fallen proteoid litter at Pella can be as high as 90: 1 (Mitchell et al. 1986). At Pella, cellulose in decomposing leaf litter of P. repens showed a similar decay to that of L. parile (Mitchell et al. 1986). A comparison of the exponential decay slopes of dry mass, soluble carbohydrates, cellulose and lignin in decomposing leaf litter of P. repens showed no significant differences between Pella and Swartboschkloof. The increase in lignin during the decomposition of leaf litter at both study sites has been demonstrated in sclerophyllous shrubs of other mediterranean-type ecosystems (Schlesinger & Hasey 1981). Protein-polyphenolic complexes may form in the decomposing leaf litter, and these would be resistant to acid hydrolysis and thus be identified as part of the lignin fraction (Read & Mitchell 1983). Nutrient return from litter of sclerophyllous shrubs may generally be small as the bulk of the nutrients, in particular nitrogen and phosphorus, are withdrawn prior to leaf abscission. The results may not be clearcut as this study has shown that as much as 66070 phosphorus but as little as 3070 nitrogen was withdrawn from abscissing leaves of P. repens. This differs from Banksia ornata F. Muel!. at Dark Island Heath, Australia, in which 90% phosphorus and 30% nitrogen is withdrawn during abscission (Specht 1981). The phosphorus content of decomposing leaves of P. repens did not decline during the 2,5-year litterbag study, agreeing with
another on decomposing leaf litter of L. parile (Mitchell et al. 1986). This may be due to chemical retention as complexed forms, microbiological retention, transfer from the soil by the colonizing microorganisms and absorption from atmospheric precipitation. In conclusion, this study has confirmed that the decomposition turnover times of leaf litter of proteoid elements of the fynbos vegetation are in excess of the 3 - 4 years found in sclerophyllous shrubs from other mediterranean-type ecosystems (Schlesinger & Hasey 1981; Specht 1981). Since the litter is decomposing at a slow rate, the litter layer will be increasing under the canopies of P. repens shrubs. In stands of mountain fynbos, dead material is known to increase with age of vegetation (from 4 to 37 years) and much of the litter in old stands is from senescing Proteas (Van Wilgen 1982). The attainment of a steady state (as defined by Olson 1963) depends upon a constant litter input and constant decomposition rate. Fires usually occur during the mature phase of development of P. repens shrubs, but when they become senescent regeneration is poor (Bond 1980). Shrubs of P. repens appear to be adapted to fires at intervals of less than 30 years (a time when shrubs normally become senescent) and the steady state as defined by Olson (1963) is never attained. Acknowledgements
We acknowledge the support of the C.S.I.R. through its Fynbos Biome Research Programme and thank Mr S. Webb for technical assistance, Mr F. van der Heyden for the Afrikaans translation of the Abstract and Ms N. Allsopp for helpful comments. References ALLEN, S.E., GRIMSHAW, H .M., PARKINSON, l.A. & QUARMBY, C. 1974. Chemical analysis of ecological materials. Blackwell Scientific Publications, Oxford. ASHTON, D.H. 1975. Studies of litter fall in Eucalyptus regnans forests. Ausl. 1. Bot. 23: 413 - 433. BOND, W. 1980. Fire in senescent fynbos in the Swartberg. S. Afr. For. 1. ,114: 68 -71. BOND, W. 1985. Canopy stored seed reserves (serotiny) in Cape Proteaceae. S. Afr. 1. Bot. 51: 181 - 186. DUBOIS, M., GILLES, K.A., HAMILTON, J .K., REBERS, P.A. & SMITH, F. 1956. Colorimetic method for determination of sugars and related substances. Anal. Chern. 28: 350 - 356. FUGGLE, R.F. 1981. Macro-climatic patterns within the Fynbos Biome. Final Report, National Programme for Environmental Sciences (Fynbos Biome Project), e.S.I.R., Pretoria. GRAY, J .T. & SCHLESINGER, W.H. 1981. Biomass, production and litter fall in the coastal sage scrub of southern California. Am. 1. Bot. 68: 24 - 33. JACKSON, M.L. 1958. Soil chemical analysis. Prentice Hall New Jersey. KING, H .G.e. & HEATH, G.W. 1967. The chemical analysis of small samples of leaf material and the relationship between the disappearance and composition of leaves . Pedobiologia 7: 192 - 197. LAMBRECHTS, J.J.N., THERON, AA & FRY, M. 1986. Detail characterization of soils under different fynbos - climate - geology combinations in south and southwestern Cape. Final Report, National Programme for Environmental Sciences (Terrestrial Ecosystems), e.S.I. R., Pretoria. LEE, K.E. & CORRELL, R.L. 1978. Litter fall and its relationship to nutrient cycling in a South Australian dry sclerophyll forest. Aust. 1. Ecol. 3: 243 - 252. MACVICAR, e.N., DE VILLIERS, J.M., LOXTON, R.F., VERSTER, E., LAMBRECHTS, J.J.N., MERRYWEATHER, F.R., LE ROUX, J., VAN ROOYEN, T .H. & HARMSE, H.J. VON M. 1977. Soil Classification: A binomial system for South Africa. Department of Agricultural Technical Services, Pretoria.
s.
Afr. J. Bot., 1987,53(1)
MAGGS, J. & PEARSON, C.J. 1977. Litter fall and litter layer decay in coastal scrub at Sydney, Australia. Oecologia 31 : 239-250. MEENTEMEYER, V., BOX, E.O. & THOMPSON, R. 1982. World patterns and amounts of terrestrial plant litter production. BioScience 32: 125 - 128. MITCHELL, D.T., BROWN, G. & JONGENS-ROBERTS, S.M. 1984. Variation of forms of phosphorus in the sandy soils of coastal fynbos, south-western Cape, South Africa. J. Ecol. 72: 575 - 584. MITCHELL, D.T., COLEY, P.G.F., WEBB, S. & ALLSOPP, N. 1986. Litter fall and decomposition in the coastal fynbos vegetation, south-western Cape, South Africa. J. Ecol. 74 (in press). MOLL, E.J., CAMPBELL, B.M., COWLING, R.M .. BOSSI, L., JARMAN, M.L. & BOUCHER, c. 1984. A description of major vegetation categories in and adjacent to the Fynbos biome. S. Afr. Nat. Sci. Prog. Report 83. C.S.I.R., Pretoria. MOONEY, H .A. , KUMMEROW, J ., JOHNSON, A.W., PARSONS, D.J., KEELEY, S., HOFFMANN, A., HAYS, R.I., GILIBERTO, J., & CHU, C. 1977. The producers their resources and adaptive responses. In: Convergent evolution in Chile and California: mediterranean climate ecosystems. Series 5, United States International Biological Program Synthesis, ed. Mooney, H .A. pp. 85 - 143, Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. MURPHY, J. & RILEY, J.P . 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 7: 31 - 36. OLSON, J .S. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44: 322 - 331. PIERCE, S.M. 1984. A synthesis of plant phenology in the Fynbos Biome. S. Afr. Nat. Sci. Prog. Report 88, C.S.I.R., Pretoria. READ, D.J. & MITCHELL, D.T. 1983. Decomposition and mineralization processes in mediterranean-type ecosystems and in heathlands of similar structure. In: Mediterranean-type
31 ecosystems. The role of nutrients. eds Kruger, F.J., Mitchell, D.T. & Jarvis, J.U.M. Vol. 43, pp. 208-232, Ecological studies, Springer, Berlin. SCHLESINGER, W.H. & HASEY, M.M. 1981. Decomposition of chaparral shrub foliage: Losses of organic and inorganic constituents from deciduous and evergreen leaves. Ecology 62: 762-774. SMITH, V.R. 1980. A phenol-hypochlorite manual aetermination of ammonium-nitrogen in Kjeldahl digests of plant tissue. Comm. Soil Sci. Plant Anal. 11, 7rfJ - 722. SPECHT, R.L. 1981. Nutrient release from decomposing leaf litter of Banksia ornata, Dark Island heathland, South Australia. Aust. J. Ecol. 6: 59 - 63 . SPECHT, R.L., MOLL, E.J., PRESSINGER, F. & SOMMERVILLE, J. 1983. Moisture regime and nutrient control of seasonal growth in mediterranean ecosystems. In: Mediterranean-type ecosystems. The role of nutrients. eds Kruger, F.J., Mitchell, D.T. & Jarvis, J.U .M. Vol. 43, pp. 120 - 132, Ecological Studies. Springer, Berlin. STOCK, W.D. & LEWIS, O.A.M . 1984. Uptake and assimilation of nitrate and ammonium by an evergreen fynbos shrub species Protea repens L. (Proteaceae). New Phytol. 97: 261 - 268. STOCK, W.D. & LEWIS , O.A .M. 1986. Soil nitrogen and the role of fire as a mineralizing agent in a South African coastal fynbos ecosystem. J. Ecol. 74: 317 - 328. SWIFT, M.J., HEAL, O.W. & ANDERSON, J.M . 1979. Decomposition in terrestrial ecosystems. Studies in Ecology Vol. 5, pp. 372, Blackwell Scientific Publications, Oxford. VAN DER MERWE, P. 1966. Die flora van Swartboschkloof, Stellenbosch, en die herstei van die soorte na 'n brand. Annis. Univ. of Stellenbosch 41A: 691 -736. VAN WILGEN, B.W. 1982. Some effects of post-fire age on the above-ground plant biomass of fynbos (macchia) vegetation in South Africa. J. Ecol. 70: 217 - 225 . VERSFELD, D.B. 1981. Litter fall and decomposition in stands of mature Pinus radiata. S. Afr. For. J. 116: 40 - 50.
25
Litter production and decomposition from shrubs of Protea repens growing in sand plain lowland and mountain fynbos, south-western Cape D.T. Mitchell* and P.G.F. Coley Department of Botany, University of Cape Town, Rondebosch, 7700 Republic of South Africa *Present address: Department of Botany, University College, Dublin 4, Eire
Accepted 1 August 1986 Litter production and decomposition of mature shrubs of Protea repens (L.) L. growing in coastal and mountain fynbos , south-western Cape, were studied during 1981 -1983. Mean annual litter production under the canopies ranged from 200 to 450 g m - 2 . Leaf fall accounted for 50-87% of the total with floral bracts being the next dominant category. November- January was the main period of litter release. Winter (June- August) leaf litter contained the lowest soluble carbohydrates and highest lignin levels. The phosphorus and nitrogen concentrations of freshly fallen litter were less than that of attached leaves and the decline ranged from 33 to 66% and 3 to 25% respectively. Leaf litter decomposition was not seasonal and no differences were encountered in the decay of dry mass, lignin, cellulose and soluble carbohydrates in the decomposing leaf litter from the two study areas. Leaf litter decomposition turnover times of P. repens are longer than those of sclerophyllous shrubs of other mediterranean-type ecosystems. Produksie en ontbinding van plant-afval van volwasse Protea repens (L.) L.-struike van kusfynbos en bergfynbos, suidwes Kaap, was bestudeer gedurende 1981- 1983. Die gemiddelde jaarlikse plantafval-produksie onder die plantkrone was tussen 200-450 g m - 2. Blaarval het 50-87% tot die totale plantafval bygedra en blom-skutblare was die volgende dominante kategorie. November- Januarie was die belangrikste peri ode vir plantafval-vrystelling . Gedurende die winter (Junie- Augustus) het die blaarafval die laagste oplosbare koolhidrate bevat en die hoogste lignien-inhoud gehad. Die fosfor- en stikstof konsentrasies van pasgevalle plantafval was minder as die van aangehegte blare en die verval het van 33 tot 66% en 3 tot 25 % onderskeidelik gewissel. Blaarafval-ontbinding was nie seisoenaal nie en geen verskille in die afbraak van die droe massa, lignien, sellulose en oplosbare koolhidrate was in die ontbindende blaarafval van die twee studie-gebiede ondervind nie. Die omsettye van blaarafval ontbinding vir P. repens is Ianger as die van sklerofiele struike van ander mediterreense-tipe-ekosisteme. Keywords: Fynbos , leaf decomposition, litter production, Protea repens , sclerophyllous shrubs •To whom correspondence should be addressed
Introduction Mature stands of sand plain lowland and mountain fynbos (Moll et a/. 1984) are often dominated by members of the Proteaceae, such as Protea repens (L.) L. which grow in acidic sandy soils of a low nutrient status (Mitchell, et a/. 1984; Stock & Lewis 1986). These slow-growing evergreen species have adapted to the low nutrient conditions by absorbing and assimilating small quantities of inorganic nitrogen (Stock & Lewis 1984). Another adaptation is to conserve nutrients and thus produce a poor quality of litter (Read & Mitchell 1983) but there is no evidence that this occurs in shrubs of P. repens. In this study on P. repens, the objectives were to: (a) quantify litter production and decomposition rates, (b) quantify litter quality and determine whether nutrients are withdrawn during leaf abscission and (c) test whether seasonal variations in litter production and decomposition differ between individuals of the same species growing in different areas. Litter quality was assessed by analyzing organic and inorganic constituents, and the decay rates of specific organic components and phosphorus in decomposing leaf litter were estimated. Study Area The two study areas were sand plain lowland fynbos at Pella (33 °31'S, 18°32'E; 160-220 m altitude; 269 ha) 62 km north of Cape Town and mountain fynbos at Swartboschkloof (33 °59'S, 18°57'E; 300-940 m altitude; 375 ha) forming part of the catchment of the Eerste River 48 km east of Cape Town. Both are in a winter rainfall region (Fuggle 1981) but Pella has a mean annual rainfall of 571 mm compared with 1624 mm at Swartboschkloof. Pella has a mean maximum and minimum temperature of 31 o and 8°C respectively whereas at Biesievlei (a station near to Swartboschkloof;
33°58'S, 18°57'E, 282m altitude), the means are 22° and l0°C respectively. At Pella, the shrubs of P. repens were aged at approximately 22 years old during 1981 and are situated in a 1-ha area along the eastern boundary in Longlands soil classified according to MacVicar eta/. (1977). This soil is a medium/fme sand approximately 2 m deep and is usually waterlogged during Winter. At Swartboschkloof, the P. repens shrubs were growing in an area burnt on 2 February 1958 (Van Wilgen 1982; Van der Merwe 1966) and were selected from a 10-ha area at an altitude of 305- 366 m. The soil was the Clovelly form , Mossdale Series classified according to Mac Vicar eta/. (1977), and the B horizon has a 07o particle size distribution of 9,6 (coarse sand), 47,3 (medium sand), 32,7 (fine sand), 3,3 (silt) and 7,0 (clay) (Lambrechts eta/. 1986). Shrub height at both sites was about 2,5 m.
Methods Collection of litter Litter was collected in traps made of polyethylene shade netting attached to a circular frame of galvanized hoop iron (0,25 m 2) suspended 0,5 m above ground on a three-legged steel rod frame and the bottom of the net was drawn closed by polyester cord. The traps were positioned halfway between stem and perimeter of the canopy of 10 randomly selected shrubs at each site (one trap per shrub). Litter was collected at monthly intervals from March 1981 to August 1983, divided into leaves, floral bracts, seed and twigs and oven-dried at 80°C. Litter decomposition Recently fallen leaf litter was collected and 5 g of air-dried mass was sealed into bags (150 x !50 mm) of vinyl-coated
S.-Afr. Tydskr. Plantk., 1987, 53(1)
26 fibre-glass (2,25 mm 2 mesh size). The litter bags were placed on the ground under the canopies of 14 shrubs at each site with 10 bags per shrub on 1 May 1981 at Pella, and 1 March 1981 at Swartboschkloof. Every 2 months, five bags (one per shrub) were removed and oven-dried at 80°C.
Chemical analyses During November 1983, l-year-old fully expanded leaves were removed from the shrubs and these and the leaf litter were
ground through a Wiley mill (60 mesh). The ground samples were then chemically analyzed using the methods of Allen et al. (1974), Schlesinger & Hasey (1981) and Mitchell et al. (1986). For inorganic analyses of Ca, Mg, K and Fe, 1 g of oven-dried, ground material was ashed in a muffle furnace at 550°C for 3 h, dissolved in HF/HClIHN0 3 (1:8:4) and concentrations of each element were determined on a Varian No 6 atomic absorption spectrophotometer. For the phosphorus detenninations, 0,1 g oven-dried material was digested
110 100 90 80 70 60 50 40 30 20 10 N
I
E
.9 U) U)
co E ~
0
110 100 90 80
50 40 30 20 10
AMJ
J
ASONDJ
FMAMJ
1981
J
ASONDJ
1982
FMAMJ
J
AS
1983
Months
Figure 1 Seasonal variations in total litter production from shrubs of Pro tea repens growing in lowland fynbos at Pella (e--e) and mountain fynbos at Swartboschkloof (-------) during 1981 - 1983. Vertical bars represent twice SE.
S. Afr. J . Bot., 1987,53(1)
27
by the method of Jackson (1958) and then assayed using the method of Murphy & Riley (1962). Total nitrogen was determined on 0,1 g oven-dried material using standard Kjeldahl procedures and selenium catalyst and sodium thiosulphate. The N~ + -N was then determined colorimetrically using the method of Smith (1980). The sample sizes for the organic constituents were 0,2 g oven-dried material. Petroleum-ether soluble fats and waxes were determined gravimetrically. Methanol-soluble (50070) carbohydrates were analyzed colorimetrically by the method of Dubois et at. (1956) with D-glucose as the standard. Cellulose was extracted by a delignification procedure with the addition of sodium chlorite and 10% acetic acid at 75°C (Allen et at. 1974) and the carbohydrate content was analyzed by the Dubois et at. method (1956). The lignin fraction was extracted and determined by methods similar to King & Heath (1967) and Schlesinger & Hasey (1981).
Statistical analyses The litter fall data were grouped into four seasons: autumn (March - May), winter (June - August), spring (SeptemberNovember) and summer (December - February), and analyzed by one-way analysis of variance. The results between the two study areas were compared by the Student's t test. In the litter decomposition studies, the analyses of the various constituents were replicated five times and the mean concentration was multiplied by the fraction of the original litter mass remaining. This product was expressed as a percent of the original content of each constituent in the fresh litter. Exponential decay relationships were calculated for each constituent. Overall fractional loss rates (k) were calculated from the formula: natural log (X/Xo) = - kt where Xl and Xo are the dry masses remaining at time t and time respectively (Olson 1963). Instantaneous fractional loss rates (kl) were also calculated from the formula: kl = 1 - e - kt which
°
Table 1 Annual litter production (g m- 2) and seasonal variations (g m- 2 month- 1 ± SE) in the major components of litter fall under the canopies of Protea repens at Pella and Swartboschkloof Pella Floral bracts
Seed
Leaves
Floral bracts
Seed
28,5 75,1
12,7 16,1
388,1 284,2
9,0 53,0
9,2 26,8
5,9 ± 2,2
1,1 ± 0,2
16,4 ± 4,5
1,5 ± 1,1
0,9 ± 0,1
3,2 ± 1,6
0,2 ± 0,1
15,1 ± 2,0
4,0 ± 1,6
2,4 ± 1,2
5,2 ± 1,3
2,3 ± 0,6
28,0 ± 5,8
3,4 ± 1,5
1,2 ± 0,4
3,2 ± 1,4 0,67 3;25 N.S.
1,3 ± 0,3 8,87 3;25 0,001
55,5 ± 6,6 15,77 3;26 0,001
1,1 ± 0,5 1,10 3;26 N.S.
1,1 ± 0,1 1,00 3;26 N.S.
Leaves Annual litter production 1981 - 1982 207,4 1982-1983 205,4 Seasonal litter production Autumn (Mar - May) 10,0 ± 2,9 Winter (lune - Aug) 4,8 ± 0,5 Spring (Sept- Nov) 20,4 ± 9,5 Summer 31,9 ± 6,6 (Dec -Feb) F 5,54 d.f 3;25 p:;::; 0,01
Swartboschkloof
N.S. denotes no significant seasonal variation
Table 2 The organic constituents of attached mature leaves collected during November 1983 and seasonal variations of the organic constituents of the leaf litter of Protea repens at Pella and Swartboschkloof Pella Soluble fats & waxes
Swartboschkloof
Soluble carbohydrates
Cellulose
Attached Leaves 19,4 ± 0,1 116 ± 12 Leaf Litter Autumn (Mar - May) 10,8 ± 1,2 130 Winter (lun - Aug) 11,6 ± 0,8 101 Spring (Sep - Nov) 7,1 ± 0,6 152 Summer (Dec - Feb) 16,0 ± 2,1 149 F
6,7 d.f 3;69 P:;::; 0,001
Lignin
Soluble fats & waxes
Soluble carbohydrates
Cellulose
Lignin
74 ± 1
186 ± 6
19,3 ± 0,2
68 ± 3
84 ± 1
214 ± 4
± 6
95 ± 4
219 ± 6
10,0 ± 0,8
116 ± 5
106 ± 4
239 ±4
± 5
97 ± 5
407 ± 8
11,5 ± 1,0
95 ± 6
104 ± 7
407 ± 11
± 5
72 ± 6
221 ± 20
6,7 ± 0,5
116 ± 6
113 ± 7
281 ± 18
± 8
80 ± 5
164 ± 5
9,9 ± 1,0
118 ± 6
99 ± 8
216 ± 7
5,30 3;66 0,01
91,37 3;61 0,001
15,84 3;66 0,001
Results are expressed as mg g -
I
6,15 3;86 0,001
3,87 3;85 0,05
0,83 3;86 N.S.
dry mass ± SE and N.S. denotes no significant seasonal variation
51,11 3;71 0,001
S.-Afr. Tydskr. Plantk ., 1987, 53(1)
28
and the second dominant category was floral bracts (11 26%). Seeds (5 - 7% of the total annual litter) were released from the flowerheads. Twigs (0,0-1,4%) formed the smallest category of litter. The timing of litter fall at Pella and Swartboschkloof produced a similar major peak between November and January (Figure 1). Leaf fall showed significant seasonal variations, whereas of the other litter categories, only seed release from shrubs at Pella varied significantly (Table 1). A significantly greater release of leaf litter from shrubs at Swartboschkloof compared with those at Pella occurred during winter (d.f = 18; P < 0,001) and summer (d.! = 10: P < 0,05). The soluble carbohydrates of attached mature, l-year-old leaves on the shrubs at Pella were higher than those at Swartboschkloof (d.! = 8; P < 0,001), whereas the concentrations of cellulose (d.! = 8; P < 0,001) and lignin (d.! = 8; P < 0,01) were lower (Table 2). The leaf litter released during the winter months contained the lowest soluble carbohydrates and had the highest lignin concentrations (Table 2). Significant differences in the specific organic components of freshly fallen litter between the two study areas occurred during summer (d.! = 25; P < 0,05) for fats and waxes, spring (d.! = 32; P < 0,001) and summer (d.! = 28; P < 0,01) for soluble carbohydrates, spring (d.! = 33; P <: 0,001) for cellulose, and autumn (d.f = 33; P < 0,01), spring (d.! = 33; P < 0,05) and summer (d! = 33; P < 0,001) for lignin concentrations. Of the inorganic constituents, only nitrogen and phosphorus concentrations of the attached mature leaves and the freshly fallen litter were compared. There was a significant difference in nitrogen concentration of attached leaves between the two study areas and its decline prior to leaf fall at Pella (25%) was greater than that at Swartboschkloof (3%) (Table 3). The phosphorus concentration dropped by 33 - 66% in the freshly fallen litter compared with concentrations in the attached mature leaves (Table 3). The concentrations of phosphorus, nitrogen, potassium, calcium, magnesium and iron in the abscissed leaves at Pella and Swartboschkloof were similar (Table 3).
is useful when comparing decomposition rates for labile constituents in long-term studies (Olson 1963). Turnover times for decomposition (years) were calculated from the reciprocal of k and half times were estimated from 0,693/k (Olson 1963). Results Litter production Shrubs of P. repens at Pella had the lower annual litter production releasing 256 and 297 g m - 2 during 1981 - 1982 and 1982 - 1983 respectively, compared with 448 and 364 g m - 2 from shrubs at Swartboschkloof. Leaf litter was the major category consisting of 69 - 87070 of the total annual litter
Table 3 Inorganic constituents (mg g-1 dry mass ± SE with number of analyses in parentheses) of attached and abscissed leaves of Protea repens at Pella and Swartboschkloof Swartboschkloof
Pella Phosphorus Attached leaves Abscissed leaves Autumn (Mar - May) Winter (lun - Aug) Spring (Sept - Nov) Summer (Dec-Feb)
0,1 0,2 0,1 0, I
Nitrogen Attached leaves Abscissed leaves
6,8 ± 0, I (5) 5,1 ± 0,5 (15)
6,1 ± 0,1 (5) 5,9 ± 0,3 (14)
Potassium Abscissed leaves
1,0 ± 0,2 (4)
1,1 ± 0,3 (5)
Calcium Abscissed leaves
7,0 ± 2,7 (4)
7,3 ± 2,2 (5)
Magnesium Abscissed leaves
1,8 ± 0,6 (4)
1,6 ± 0,4 (5)
Iron Abscissed leaves
0,9 ± 0,3 (4)
0,6 ± 0,2 (5)
0,3 ± 0,01 (5)
± ± ± ±
0,01 0,01 0,01 0,01
(20) (15) (15) (14)
0,3 ± 0,02 (5) 0,1 0,2 0,1 0,1
± ± ± ±
0,01 0,01 0,01 0,01
(20) (15) (20) (18)
Table 4 Decomposition constants for leaf litter from Protea repens at Pella and Swartboschkloof during a 2,5-year litterbag study
Pella Dry mass Soluble fats & waxes Soluble carbohydrates Cellulose Lignin Phosphorus Swartboschkloof Dry mass Soluble fats & waxes Soluble carbohydrates Cellulose Lignin Phosphorus
-k
_kl
Halftime (yr)
Turnover (yr)
n
r
P
-0,05
-0,05
14,14
20,40
15
-0,50
0,05
-0,08
-0,08
8,26
11,90
9
-0,27
N.S.
-0,49 -0,46 +0,10 +0,25
-0,39 -0,37
1,41 1,51
2,04 2,18
12 10 10 14
-0,82 -0,62 +0,81 +0,57
0,001 0,05 0,01 0,05
-0,D7
- 0,07
10,35
14,94
16
-0,83
0,001
-0,44
-0,36
1,58
2,28
10
-0,80
0,01
-0,59 -0,03 +0,07 +0,11
-0,45 -0,03
1,17 20,92
1,68 30,18
12 10 12 15
-0,69 -0,13 +0,58 +0,41
0,01 N.S. 0,05 N.S .
k is calculated from (x/Xo) = e - kt; kl = I - e - kt; Half-time = 0,693/ k turnover of means of five separate samples and r is correlation coefficient
=
Ilk; n is number
S. Afr. J. Bot., 1987 , 53(1)
29
Decomposition There was no apparent seasonal pattern in the decomposition of leaf litter. Turnover times of dry mass, fats and waxes and soluble carbohydrates of decomposing leaf litter from Swartboschkloof were shorter than those at Pella (Table 4). The slopes of exponential decay relationships of dry mass, lignin, cellulose and soluble carbohydrates of decomposing leaf litter at the two study areas were not significantly different. The breakdown and release of cellulose from the decomposing leaf litter at Pella was similar to soluble carbohydrate release whereas the cellulose composition of the decomposing leaves
at Swartboschkloof showed no significant decline (Figure 2). Lignin composition of the litter at the two sites increased during decomposition (Figure 2; Table 4). The phosphorus concentration of the decomposing leaf litter at Pella increased whereas no significant changes occurred in the litter at Swartboschkloof (Table 4).
Discussion The mean annual litter production from mature shrubs of Protea repens growing in sand plain lowland and mountain fynbos fell within the range of 200- 450 g m - 2 • This is similar
U) U)
"'E
90
80
..:---·---.---.---- ---------
A
...
.A.
I
70
• .A..----.•t- ---~--·---~---· ---~---!! I • f·-t---·--·----
.A.
c c Ol
:.:;
•
•
•
. ------- _, ___ ~!"_ ..... -- ·· -· .... -- -~-- ·-·-·. --.- . .11!11_. -- .. . -- . -- .. .............. --.80 OJ
70
U)
c c0 OJ
u
0 :J
a; u
60
•
•
50
(ij
c
40
Ol
8
30
0
c
20
OJ
u
(j;
a.
U)
OJ
'§ '0
>-
.<::: 0
70
£J
60
OJ
50
0
40
:uu
:a:J UJ
30 20 10
... ...
--- -- . • ·-- --• • ------.----------------Duration of incubation (Mon ths)
Figure 2 Changes in dry mass , lignin, cellulose and soluble carbohydrates of decomposing leaf litter of Protea repens growing in lowland fynbos at Pella ( A. ) and mountain fynbos at Swartboschkloof ( • ). Curves are the best fitting exponential decay relationships for decomposing leaf litter at Pella ( - - ) and Swartboschkloof (---).
S.-Afr. Tydskr. Plantk., 1987,53(1)
30 to the results of Leucospermum pari/e (Salisb. ex J. Knight) Sweet, another proteoid shrub at Pella (Mitchell et al. 1986), shrubs of other mediterranean-type ecosystems (Mooney et al. 1977; Lee & Correll 1978; Gray & Schlesinger 1981) and stands of Pinus radiata D. Don adjacent to the mountain fynbos site in this study (Versfeld 1981). The main category of above-ground litter from proteoid shrubs is leaves, but the next important one varies with species. In P. repens in this study, the flowerheads remain on the shrubs but the seed is weakly serotinous (Bond 1985). In L. pari/e, flowerheads are an important category and are released at a different time i.e. November to leaf fall (which is February - March) (Mitchell et al. 1986). Leaf fall and shoot growth of sclerophyllous shrubs, in particular examples belonging to the Proteaceae, occur simultaneously during spring - summer (Specht et al. 1983). The seasonal release of litter from vegetation of mediterranean-type ecosystems may be controlled by high summer temperatures and low summer rainfall (Lee & Correll 1978; Meentemeyer et al. 1982). In the south-western Cape, the south-easterly winds prevailing during the summer may affect above-ground litter fall (Mitchell et al. 1986). In this study, the period of litter fall from P. repens at Pella (a drier site and warmer during summer) and Swartboschkloof was the same occurring during late spring/summer. Litter fall from sclerophyllous shrubs and trees follow a simple harmonic function (Ashton 1975; Maggs & Pearson 1977; Lee & Correll 1978). In P. radiata adjacent to the mountain fynbos site in this study, a harmonic function of needle fall was demonstrated with peak production in March and male cone production occurring during July - September (Versfeld 1981). The factors which influence the different phenophases such as the timing of litter fall are numerous and whether the cues are determined by environmental or endogenous rythms are open to speculation (Pierce 1984). Leaves of P. repens have sclerophyllous characteristics in being low in organic constituents but having high concentrations of polymeric organic compounds such as cellulose and lignin (Swift et al. 1979). Cellulose and lignin would form the main carbon composition of the litter and CN ratios of freshly fallen proteoid litter at Pella can be as high as 90: 1 (Mitchell et al. 1986). At Pella, cellulose in decomposing leaf litter of P. repens showed a similar decay to that of L. parile (Mitchell et al. 1986). A comparison of the exponential decay slopes of dry mass, soluble carbohydrates, cellulose and lignin in decomposing leaf litter of P. repens showed no significant differences between Pella and Swartboschkloof. The increase in lignin during the decomposition of leaf litter at both study sites has been demonstrated in sclerophyllous shrubs of other mediterranean-type ecosystems (Schlesinger & Hasey 1981). Protein-polyphenolic complexes may form in the decomposing leaf litter, and these would be resistant to acid hydrolysis and thus be identified as part of the lignin fraction (Read & Mitchell 1983). Nutrient return from litter of sclerophyllous shrubs may generally be small as the bulk of the nutrients, in particular nitrogen and phosphorus, are withdrawn prior to leaf abscission. The results may not be clearcut as this study has shown that as much as 66070 phosphorus but as little as 3070 nitrogen was withdrawn from abscissing leaves of P. repens. This differs from Banksia ornata F. Muel!. at Dark Island Heath, Australia, in which 90% phosphorus and 30% nitrogen is withdrawn during abscission (Specht 1981). The phosphorus content of decomposing leaves of P. repens did not decline during the 2,5-year litterbag study, agreeing with
another on decomposing leaf litter of L. parile (Mitchell et al. 1986). This may be due to chemical retention as complexed forms, microbiological retention, transfer from the soil by the colonizing microorganisms and absorption from atmospheric precipitation. In conclusion, this study has confirmed that the decomposition turnover times of leaf litter of proteoid elements of the fynbos vegetation are in excess of the 3 - 4 years found in sclerophyllous shrubs from other mediterranean-type ecosystems (Schlesinger & Hasey 1981; Specht 1981). Since the litter is decomposing at a slow rate, the litter layer will be increasing under the canopies of P. repens shrubs. In stands of mountain fynbos, dead material is known to increase with age of vegetation (from 4 to 37 years) and much of the litter in old stands is from senescing Proteas (Van Wilgen 1982). The attainment of a steady state (as defined by Olson 1963) depends upon a constant litter input and constant decomposition rate. Fires usually occur during the mature phase of development of P. repens shrubs, but when they become senescent regeneration is poor (Bond 1980). Shrubs of P. repens appear to be adapted to fires at intervals of less than 30 years (a time when shrubs normally become senescent) and the steady state as defined by Olson (1963) is never attained. Acknowledgements
We acknowledge the support of the C.S.I.R. through its Fynbos Biome Research Programme and thank Mr S. Webb for technical assistance, Mr F. van der Heyden for the Afrikaans translation of the Abstract and Ms N. Allsopp for helpful comments. References ALLEN, S.E., GRIMSHAW, H .M., PARKINSON, l.A. & QUARMBY, C. 1974. Chemical analysis of ecological materials. Blackwell Scientific Publications, Oxford. ASHTON, D.H. 1975. Studies of litter fall in Eucalyptus regnans forests. Ausl. 1. Bot. 23: 413 - 433. BOND, W. 1980. Fire in senescent fynbos in the Swartberg. S. Afr. For. 1. ,114: 68 -71. BOND, W. 1985. Canopy stored seed reserves (serotiny) in Cape Proteaceae. S. Afr. 1. Bot. 51: 181 - 186. DUBOIS, M., GILLES, K.A., HAMILTON, J .K., REBERS, P.A. & SMITH, F. 1956. Colorimetic method for determination of sugars and related substances. Anal. Chern. 28: 350 - 356. FUGGLE, R.F. 1981. Macro-climatic patterns within the Fynbos Biome. Final Report, National Programme for Environmental Sciences (Fynbos Biome Project), e.S.I.R., Pretoria. GRAY, J .T. & SCHLESINGER, W.H. 1981. Biomass, production and litter fall in the coastal sage scrub of southern California. Am. 1. Bot. 68: 24 - 33. JACKSON, M.L. 1958. Soil chemical analysis. Prentice Hall New Jersey. KING, H .G.e. & HEATH, G.W. 1967. The chemical analysis of small samples of leaf material and the relationship between the disappearance and composition of leaves . Pedobiologia 7: 192 - 197. LAMBRECHTS, J.J.N., THERON, AA & FRY, M. 1986. Detail characterization of soils under different fynbos - climate - geology combinations in south and southwestern Cape. Final Report, National Programme for Environmental Sciences (Terrestrial Ecosystems), e.S.I. R., Pretoria. LEE, K.E. & CORRELL, R.L. 1978. Litter fall and its relationship to nutrient cycling in a South Australian dry sclerophyll forest. Aust. 1. Ecol. 3: 243 - 252. MACVICAR, e.N., DE VILLIERS, J.M., LOXTON, R.F., VERSTER, E., LAMBRECHTS, J.J.N., MERRYWEATHER, F.R., LE ROUX, J., VAN ROOYEN, T .H. & HARMSE, H.J. VON M. 1977. Soil Classification: A binomial system for South Africa. Department of Agricultural Technical Services, Pretoria.
s.
Afr. J. Bot., 1987,53(1)
MAGGS, J. & PEARSON, C.J. 1977. Litter fall and litter layer decay in coastal scrub at Sydney, Australia. Oecologia 31 : 239-250. MEENTEMEYER, V., BOX, E.O. & THOMPSON, R. 1982. World patterns and amounts of terrestrial plant litter production. BioScience 32: 125 - 128. MITCHELL, D.T., BROWN, G. & JONGENS-ROBERTS, S.M. 1984. Variation of forms of phosphorus in the sandy soils of coastal fynbos, south-western Cape, South Africa. J. Ecol. 72: 575 - 584. MITCHELL, D.T., COLEY, P.G.F., WEBB, S. & ALLSOPP, N. 1986. Litter fall and decomposition in the coastal fynbos vegetation, south-western Cape, South Africa. J. Ecol. 74 (in press). MOLL, E.J., CAMPBELL, B.M., COWLING, R.M .. BOSSI, L., JARMAN, M.L. & BOUCHER, c. 1984. A description of major vegetation categories in and adjacent to the Fynbos biome. S. Afr. Nat. Sci. Prog. Report 83. C.S.I.R., Pretoria. MOONEY, H .A. , KUMMEROW, J ., JOHNSON, A.W., PARSONS, D.J., KEELEY, S., HOFFMANN, A., HAYS, R.I., GILIBERTO, J., & CHU, C. 1977. The producers their resources and adaptive responses. In: Convergent evolution in Chile and California: mediterranean climate ecosystems. Series 5, United States International Biological Program Synthesis, ed. Mooney, H .A. pp. 85 - 143, Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. MURPHY, J. & RILEY, J.P . 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 7: 31 - 36. OLSON, J .S. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44: 322 - 331. PIERCE, S.M. 1984. A synthesis of plant phenology in the Fynbos Biome. S. Afr. Nat. Sci. Prog. Report 88, C.S.I.R., Pretoria. READ, D.J. & MITCHELL, D.T. 1983. Decomposition and mineralization processes in mediterranean-type ecosystems and in heathlands of similar structure. In: Mediterranean-type
31 ecosystems. The role of nutrients. eds Kruger, F.J., Mitchell, D.T. & Jarvis, J.U.M. Vol. 43, pp. 208-232, Ecological studies, Springer, Berlin. SCHLESINGER, W.H. & HASEY, M.M. 1981. Decomposition of chaparral shrub foliage: Losses of organic and inorganic constituents from deciduous and evergreen leaves. Ecology 62: 762-774. SMITH, V.R. 1980. A phenol-hypochlorite manual aetermination of ammonium-nitrogen in Kjeldahl digests of plant tissue. Comm. Soil Sci. Plant Anal. 11, 7rfJ - 722. SPECHT, R.L. 1981. Nutrient release from decomposing leaf litter of Banksia ornata, Dark Island heathland, South Australia. Aust. J. Ecol. 6: 59 - 63 . SPECHT, R.L., MOLL, E.J., PRESSINGER, F. & SOMMERVILLE, J. 1983. Moisture regime and nutrient control of seasonal growth in mediterranean ecosystems. In: Mediterranean-type ecosystems. The role of nutrients. eds Kruger, F.J., Mitchell, D.T. & Jarvis, J.U .M. Vol. 43, pp. 120 - 132, Ecological Studies. Springer, Berlin. STOCK, W.D. & LEWIS, O.A.M . 1984. Uptake and assimilation of nitrate and ammonium by an evergreen fynbos shrub species Protea repens L. (Proteaceae). New Phytol. 97: 261 - 268. STOCK, W.D. & LEWIS , O.A .M. 1986. Soil nitrogen and the role of fire as a mineralizing agent in a South African coastal fynbos ecosystem. J. Ecol. 74: 317 - 328. SWIFT, M.J., HEAL, O.W. & ANDERSON, J.M . 1979. Decomposition in terrestrial ecosystems. Studies in Ecology Vol. 5, pp. 372, Blackwell Scientific Publications, Oxford. VAN DER MERWE, P. 1966. Die flora van Swartboschkloof, Stellenbosch, en die herstei van die soorte na 'n brand. Annis. Univ. of Stellenbosch 41A: 691 -736. VAN WILGEN, B.W. 1982. Some effects of post-fire age on the above-ground plant biomass of fynbos (macchia) vegetation in South Africa. J. Ecol. 70: 217 - 225 . VERSFELD, D.B. 1981. Litter fall and decomposition in stands of mature Pinus radiata. S. Afr. For. J. 116: 40 - 50.