0493-B2

Humus forms of beech (Fagus orientalis) forests in north of Iran

T.Sajedi

Abstract

Due to the various humus forms' role in forest ecosystem and its intimate relationships among vegetation and soil characteristics, humus forms are one of the principal components of terrestrial ecosystems. In this research we have studied humus forms and their variation in pure and mixed Beech( Fagus orientalis) stands, one of the most important components of Hyrcanian forests in Iran.

12 tree types were identified and 6 of them, which had covered the most area, were chosen for humus form study. For scattering the humus form samples we used landform units. To identify and describe the humus forms we used the methods described in Green at al. (1993). In addition all samples were analyzed in 11 chemical properties (pH, total C, total N, C/N ratio, (Ca + K), K/Ca, CaCO3, moisture, silt, sand and clay) and then all of these properties were estimated for each humus form group.

To examine correlation between ectorganic layer nutrient properties and humus forms and also with tree types we used a multivariate analysis (PCA). We used regression analysis to identify possible linear correlations between chemical properties, too.

According to this study two humus form orders, Mulls and Moders, and six groups were identified. 87 percent of the samples were Moder and the others were Mull. The result of chemical properties analysis confirmed all morphological similarities and differences between all humus forms. These results indicate that morphological properties of the horizons and humus forms are related to their chemical properties. Furthermore, some of these properties represented a good variation in relation to the tree species such as: total N, C/N ratio, exchangeable (K & Ca) and (Ca + K) but the others like pH and total C didn't represent patterns which were expected. According to these results, humus form is an important factor for site classification

1. Introduction

Humus forms provide the habitat for decomposer organisms and also have an important role in carbon and nutrient sink and resource. They are one of the determinants of rooting zone tempreture, moisture, aeration and nutrient. Due to its role and it's initimate relationships among vegetation and soil charactristics, humus forms are one of the principal components of terrestrial ecosystems.

In this research, which is the first research on humus forms in Iran, we have studied humus forms and their variation in pure and mixed Beech (Fagus orientalis) stands, which are one of the most important components of forest ecosystems in north of Iran, in a part of Chelir forest with about 700 ^ha area.

It is an unmaneged mountain forest with about 1200 - 1400 ^mm annual precipitation. The range of altitude is about 800 - 1700 ^m .

2. Materials & methods

There wasn't any vegetation map in this area, so at first the tree typology has been done through the whole area on the basis of a physionomic method. To increase the relative accuracy the typology was done in the lines with 120 ^m widths and 50 ^m lengths. To delimitate the tree types we used physiographic factors and 50 ^m distances.

After all, 12 types, which beeches were dominant, were identified and 6 of them, which had covered the most area, were chosen for humus form study. For scattering the humus form samples better and to have the most natural conditions in our sampling we used landform units so we identified 76 point for humus form study.

To identify and describe the humus forms we used the methods described in Green at al. (1993), which is based on morphological charactristics. We studied five master horizons: L, F, H, Ah and A1, a mineral layer before B horizon. All 313 samples were analyzed in 11 chemical properties: pH, total C, total N, C/N ratio, exchangable K and Ca, sum of exchangable Ca and K (K+Ca), K/Ca ratio in all layers and CaCO3, moisture, silt, sand and clay just in mineral layer. Then all of these properties were estimated for each humus form group and each tree type.

To examine correlation between ectorganic layer nutrient properties and humus forms and also with tree types we used a multivariate analysis (PCA). We used regression analysis to identify possible linear correlations between chemical properties, too.

3. Results

3.1. Morphological properties

Based on similarityin humus form profiles, we stratified 74 samples into 6 groups: group 1: Lv, Fa, Fz, Hr or Hz( n = 22); group 2: Lv, Fz, Hr or Hz( n = 19); group 3: Lv, Fz or Fa, (Hz ), Ah( n = 10); group 4: Lv, Fa, Hr( n = 12); group 5: Lv, Fa, Fz, Hz or Hh( n = 9); group 6: Lv, Fz, Hr, Ah( n = 4).

The samples of group 1 had A1 horizon with moderate blocky structure & firm consistance, loose to friable F horizon, which consist of Fa and a thin (1 cm) Fz layers and contains more than 50% of F+ H. The fauna activity was few and the mycelium was absent.

Group 4 samples were similar in all morphological charactristics except for absence of Fz horizon. The group 2 samples had loose and non-compact matted Fz, which contains more than 50% of F+ H and Hr horizon with common roots and fauna activity.

In group 3 the samples had A1 horizon with granular to blocky structure and loose to friable consistance and Fz horizon contains a considerable volume of the profile. The most important character in this group is the presence of Ah horizon with ≥2 cm in thikness (table 1) and all of the samples had abundant root and fauna activity.

In group 5 the organic layers thikness was more than other groups (6 -12 ^cm ) and also the thikness of Hz layer was considerable (3 - 5 ^cm ). Root and fauna activities and dropping were common in this group.

The samples of group 6 had A1 horizon with recumbent and resilient structure with dark gray coloure. F horizon is more than 50% of F+ H thikness.the Hh horizon with greasy character and dark gray coloure was very thin (0.5 - 1.5 ^cm ) and overlayed on a thin A1 horizon. Alnus subcordata, which shows the high soil moisture (Habibi H., 1992), is observed in all samples of this group.

Table 1 Range, mean and standard deviation (SD) of thikness (cm) of organic horizons in the six groups of study humus forms

On the basis of our observations group 5 had the most well developed root system. Group 3 represented the most fauna activity and an opposite trend was for group 1 and 4, which had the least activity. Furthermore, most of the morphological charactristics such as horizons sequence, their thikness, structure, consistance and also root and fauna activities were similar in both group 1 and 4, except for the presence of a thin (1 - 1.5 cm) Fz horizon in group 1. So we can put them in one group.

The L horizon in most of the samples was thick as a result of slow decomposition of beech litter. F horizon is not very thick, but in 78% of the samples it contains more than 50% of F+H.

According to the taxonomic classification of Green et al. (1993) the samples of group 1 and 4 were identified as Mormoder. In this group the mean thickness of F+H>2 cm, the dominant Fa layer contains more than 50% of F horizon and fauna activity is few. Group 3 samples were identified as Vermimulls. The total thickness of F and H horizons was ≤ 2 cm. They had an Ah horizon with more than 2cm thickness, which had a granular structure and common to abundant root and fauna activity. In group 5 samples the total thickness of F+H was more than 2cm and common fauna activity were observed.samples of group 6 had different conditions. They were observed under the influence of excessive moisture so the Hh layer had a greasy character. The combined thickness of F and H horizons was ≤ 2 cm and Ah>2 cm. Group 6 was identified as Hydromoder. We had some problems with identifing group 2. They had some charactristics between Mulls and Mullmoders. The combined thickness of F and H horizons was less than 3 cm and H horizon thickness had been less than 1 cm if it was present. In the other hand they hadn't Ah horizon. The roots were few in organic layer but common in A1 horizon. The fauna activity were common althogh the A1 horizon had a dry and firm structure.

Anyhow, 13% of the 74 samples were Mull and 87% were Moder. Thus in beech stands, which beeches are dominant (more than 50% of canopy) most of the humus forms are Moder and decomposition is slow.

3.2.nutrient properties

Some of the properties which have been studied, represented a good variation between humus form groups such as: total N, C/N ratio, exchangable K and Ca and K+Ca but the otherslike pH and total C didn't represent patterns which were expected (table 2).

To study the correlation between nutrient properties and humus form groups we used PCA (principal component analysis) analysis. The first three axis of the PCA ordination accounted for 77% of the total variance. The first axis, which accounted for about 36% of the varianceis described with C/N, tC, tN,Ca in L layer, tN, K in F layer, C/N, tC, K, K+Ca in H layer and tN, Ca, K/Ca, K+Ca and in A1 layer. The second axis accounted for about 27% of the total variance and was related to the lower layers properties: C/N, K+Ca, in F layer, Ca in H layer and C/N, tC, tN, sand and clay in A1 layer. The third axis, which accounted for about 16% of the total variance, was related to the K, K+Ca in L layer and pH in A1 layer.

Table2 - Means of selected chemical properties in the six groups of studied humus forms

Location of groups in two-dimentional ordination in Fig1 clearly shows the similarities between group 1 and 4. These two groups have a lower tN and higher C/N, tC, exchangable K in organic layers, lower Ca, K+Ca, and higher K and K/Ca in mineral (A1) layer. All of these conditions make decomposition and mineralization very difficult in these two groups. An opposite trend would be expected for group 3.

The results clearly shows the higher tN, lower tC, C/N ratio and exchangable K in organic layers, higher Ca and K+Ca and lower K and K/Cain in mineral (A1) layerof this group. So decomposition and mineralization should be very rapid and this group must be an active Mull humus form order.

Table 3. - Correlations of nutrient properties (L, F, H, A1) with the third axes of PCA

Group 5 has a high correlation with second axis (r = 0.7). It has lower C/N, higher tN in F and H horizons, lower K, Ca, K+Ca, pH, clay and higher C/N, CaCO3 and silt in mineral (A1) horizon. According to these results in mineral layer of this group exchangable Ca is lower, although the CaCO3 is more than other groups. Group 6 is almost the opposite location of group 5. It has higher pH, tC, Ca in organic horizons and lower C/N, tC, silt, sand and higher Ca and clay in mineral (A1) layer. The location of group 2 shows that it hasn't any correlation with the first two axes of the analysis. But according to the table 3 it has a good correlation with the third axis. This group has relatively higher pH in L, F, H horizons and K in L, F, A1 horizons. On the basis of these conditions, rapid decomposition and mineralization could be expected for this group.

Fig. 1. Ordination of humus form groups along the First two axes of PCA on 34 humus form properties (table 3).

4. Conclusions

Study of humus forms in mountain beech forests, which beeches are dominant, in Iran showed a low decomposition because Mor order contains 87% and Mormoders 43% of the samples, although we used landform units for sampling and so more variation would be expected in humus form samples.

Meanwhile, the range of pH of mineral layer in this broadleave and temparate forest was low (range: 4.1 - 5.7, mean = 4.9, SD: 0.3). It maybe because of the considerable volume of blowdown trees in this area.

The result of nutrient properties analysis confirmed all morphological similarities and differences between humus form groups. These results indicate that morphological properties of the horizons and humus forms are related to their chemical properties. So Green et al. (1993) method is a useful method for describing and identifying of humus forms specialy in such a countries which the analysis costs are very high. Furthermore, some of these properties represented a good variation in relation to the beech ecosystem conditions such as: tN, C/N ratio, exchangable K and Ca and sume of K and Ca.

References

1. Fisher, R.F. and Binkly D., 1999, Ecoloy and management of forest soils, Vol 489,sd 390, F56

2. Fons, J., Klinka K., Kabzems, R.D., 1998, Humus forms of trembling aspen ecosystems in northeastern British Columbia, Forest ecology and management, 105: 241-250.

3. Green, R.N., Trowbridge, R.L. and Klinka, K., 1993, Towards a taxonomic classification of humus forms, Forest science, Mon.29. Vol.39, N.1.

4. Klinka, K., Wang, Q. and Carter, R.E., 1990, Relationships among humus forms, forest floor nutrient properties and understory vegetation, Forest science, vol. 36, No. 3: 564 - 581.

5. Klinka, K., Fons, J. and Krestor, P., 1997, Towards a taxonomic classification of humus forms: Third approximation, Scientia silvica, extention series. No. 9

6. Sioen, G., Neirynuck, J., Maddelein, D. and Muys, B., 1994, Site classification in a hardwood forest (Hallerbos, Belgium) with a homogeneus ground vegetation dominated Hyacinthoides Non-Scripta, Silva gandavensis, N. 59, p. 15 - 28

7. Zahedi, Amiri, Gh., 1998, Relations between ground vegetation and soil charactristics in a mixed hardwood stand, Ph.D.thesis, Academic press, University of Gent, Belgium, 319 pp.

8. Habibi Kasseb, H., 1982, Fundamentals of forest soil science, Tehran university publication, N. 2118, vol. 424.

9. Zarrinkafsh, M., 1988, Methods of analysis for soil, plants and waters, Tehran university publication, N. 1955, vol. 248.