Nutrient Dynamics in Litterfall and Decomposing Leaf Litter

at the Kwangnung LTER Site, Korea


Choonsig Kim, Jeong-Soo Oh, Jong-Hwan Lim and Kyung Choi

Department of Forest Environment

Korea Forest Research Institute, 130-012 Korea





Litterfall and litter decomposition represent a major contribution to the nutrient and carbon inputs in forest ecosystem. We measured litterfall quantity and nutrient dynamics in decomposing litter for two years at the Kwangnung Long-Term Ecological Research (LTER) site in Korea. Litterfall was collected in circular littertraps (collecting area : 0.25m2) and mass loss rates and nutrient release in decomposing litter were estimated using the litterbag technique employing 30cm30cm nylon bags with 1.5mm mesh size. Total annual litterfall was 5,627kg/ha/yr and leaf litter account for 61% of the litterfall. The leaf litter quantity was highest in Quercus serrata, followed by Carpinus laxiflora and C. cordata, etc., which are dominant tree species in the site. Mass loss rates from decomposing litter were more rapid in C. laxiflora and C. cordata than in Q. serrata litter. About 77% and 84% of C. laxiflora and C. cordata litter disappeared, while about 48% in the Q. serrata litter lost for two year. Lower mass loss rates of Q. serrata litter may be attributed to the difference of substrate quality such as lower nutrient concentrations compared with the other litter types. Nutrient concentrations (N, P, Mg) of three litter types except for potassium (K) increased compared with initial nutrient concentrations of litter over the study period. The results suggest that litter mass loss and nutrient dynamic processes among tree species vary considerably on same site condition.





Litterfall inputs and litter decomposition represent a large and dynamic portion of the nutrient cycling in forest ecosystem. In addition, the turnover of litter is a major pathway of the nutrient and carbon inputs to forest soils. Significant amounts of organic matter and nutrients in the soils can be transferred during litter decomposition processes.

Natural hardwood stands in the temperate forest zone of Korea are mixed with various kinds of deciduous tree species. Although several studies have reported litterfall inputs and litter decomposition in hardwood forest ecosystem in Korea, little is known about the direction and rates of change associated with mixed-hardwood forest ecosystem. The objectives of this study were to measure litterfall and nutrient quantity; 2) to examine decomposition rates in Quercus serrata, Carpinus laxiflora and C. cordata litter; 3) to determine patterns of nutrient release from decomposing litters at the LTER site of Kwangnung, a mixed-hardwood forest ecosystem in Korea.



Material and Methods


This study was conducted in the National Arboretum in Kwangnung, Kyunggi-do, Korea. This area has been designated as the LTER site in Korea since 1998. The study site was located in the northern temperate forest zone (374516N , 1271020E) in Korea and the soils were classified brown forest soils (mostly Inceptisols) developed on Granite gneiss. Annual precipitation in the site averages 1,365mm and is higher than the average of the country (1,274mm). Annual mean temperature is 11.3. Tree density of the site was 1,473 trees/ha and basal area was 28m2/ha. Dominant tree species in the site were Q. serrata, which occupies 51% of the basal area, and followed by 23% in C. laxiflora, and 7.8% in C. cordata etc..

Litterfall was collected in circular traps devised by Hughes et al. (1987) using 1.5 mm nylon net. The collecting area was 0.25m2. The twelve traps  in three plots (2010m2) were installed 50cm above ground.  Litter was collected at approximately monthly intervals from October 1998 to October 2000. Litter collected from each trap was transported to the laboratory and oven-dried at 60 for 48 hours All dried samples were separated into leaf, bark, flowers, acorn, woody and miscellaneous components and each portion was weighed.

Mass loss and nutrient release in decomposing litter were estimated using the litterbag techniques employing 3030cm nylon bags with 1.5mm mesh size. Fresh leaf litter from the site was collected during heavy litterfall season (late November) in 1998. Collected litter samples were dried to constant mass at room temperature for 14 days and sorted into representative deciduous foliage in the stands. Ten grams litter of air-dried three dominant tree species (Q. serrata, C. laxiflora, and C. cordata) was weighed to nearest 0.01 g and placed in numbered litter bags. Subsamples from each litter type were also taken to determine oven-dried mass at 65 for 48 hours. The litterbags were randomly placed on the forest floor on 4 December 1998. The twenty-seven bags (3 plots3 species3 replications) in each sampling time were collected on five occasions over the study period. Collected bags were oven-dried at 65 for 48 hours. Litter in the bag were cleaned by gentle brushing with a soft paintbrush to remove mineral soil and weighed to determine litter mass loss rates.

Litterfall and litter in the bag were ground in a Wiley mill to pass a 40-mesh stainless steel sieve. All nutrients (N, P, K, Ca, Mg) were analyzed by the standard method of National Institute of Agriculture Science and Technology (1988)



Results and Discussion


The total annual litterfall at the Kwangnung LTER site was 5,627 kg/ha/yr (Table 1). These values fall within the range for temperate deciduous forest (Bray and Gorham 1964). Leaf litter was the major component of total litterfall in the stands. Leaf litter accounted for 61% of the total annual litterfall, followed by branch (16%)>miscellaneous(10%)>acorns(6%)>bark(4%). Heavy litterfall season in the site was November (Fig 1). Litterfall during this period involved 53% of annual litterfall.


 Table. 1. Annual litterfall inputs at the Kwangnung LTER site.


Leaf litter(kg/ha)







Q. serrata

C. laxiflora

C. cordata

Other tree



























   Figure 1.  Monthly patterns of leaf litter inputs at the Kwangnung LTER site.

Nutrient inputs (kg/ha/yr) by leaf litter were highest in Ca (34.0), followed by  N (20.5) > K (11.8) > Mg (6.0) > P (1.0) (Fig. 2). A Q. accutisima stand in the same area (Kwangnung) showed a similar nutrient distribution pattern that is highest in Ca and lowest in P (Kim et al. 1998). Q. serrata (67.1kg/ha/yr) among dominant tree species in the site was a major contributor of the nutrient inputs in the site, followed by C. laxiflora (20kg/ha/yr) and C. cordata (2.7kg/ha/yr).



    Figure 2.  Nutrient inputs by annual litterfall at the Kwangnung LTER site.


Mass loss rates for two year from decomposing litter were Q. serrata < C. laxiflora < C. cordata. Mass loss rates were lowest in Q. serrata litter among three litter types (Fig. 3). About 48% of the original mass in the Q. serrata litter disappeared, while about 77% in C. laxiflora and 84% in C. cordata. Lower mass loss in Q. serrata litter may be attributed to the difference of substrate quality such as lower nitrogen concentration compared with the other litter types (Fig. 4). Also, nitrogen concentration in decomposing litter during the study period showed lower in Q. serrata than in C. laxiflora and C. cordata litter (Fig. 4).



   Figure 3.  Mass loss rates of leaf litter for two year at the Kwangnung LTER site.


Nitrogen concentration from decomposing litter increased over the study period in three litter types. Many studies have noted increased N concentration in litter during decomposition process (Berg 1988, Van Vuuren and Van der Eerden 1992). This increase could be due to microbial or non-microbial N immobilization and additions by atmospheric N decomposition during decomposition. In addition, fungal activity has been reported to be a major source of increased N in decomposing litter. Fungal mycelia contain 35% N on a dry mass basis and have the capacity to translocate N from organic and mineral soil layers during litter decomposition. Also, microorganisms decomposing litter can take up 15N applied in artificial rain water (Van Vuuren and Van der Eerden 1992). If a portion of this N could be absorbed by decomposing litter, it could influence the gains of N in decomposing litter.

It is difficult to explain the variation in P concentrations of three litter types during the study period. Phosphorus concentration decreased during the intial stage of decomposition, but increased after one year. Similar patterns were observed in the litter the other hardwoods, such as flowering dogwood, red maple, and chestnut oak (Blair 1988). Phosphorus concentrations also were higher in C. laxiflora and C. cordata than in Q. serrata litter. Higher P concentration in C. laxiflora and C. cordata litter may be due to rapid loss of dry matter throughout decomposition process.

Potassium was the most readily released element compared with other nutrients because potassium is present in litter as a water soluble nutrient. Potassium concentration dropped rapidly during the first 5 months of litter incubation and then stabilized. Rapid release of K early during litter decomposition process is a commonly observed phenomenon in other litter decomposition studies (Lisanework and Michelsen 1994). 

   Calcium showed similar concentrations during the litter decomposition process. Calcium is present in plant tissues in the form of calcium ions or insoluble salts in the vacuoles. It is firmly bound as calcium pectates in the cell walls. This result suggests that calcium may have less leaching characteristics compared with other nutrients. 

   Magnesium generally tended to increase. Magnesium is generally mobile in litter and exposed to leaching. The increase of Mg concentration in three litter types may be due to rapid loss of dry matter throughout decomposition processes.









   Figure 4.  Nutrient concentration change of decomposing litter at the Kwangnung LTER site



Berg, B. 1988. Dynamics of nitrogen (15N) in decomposing Scots pine (Pinus sylvestris) needle litter. Long-term decomposition in a Scots pine forest VI. Can J. Bot. 66 : 1539-1546.

Blair, J.M. 1988. Nitrogen, sulfur, and phosphorus dynamics in decomposing deciduous leaf litter in the southern Appalachians. Soil Biol. Biochem. 20: 693-701.

Bray, J.R. and E. Gorham. 1964. Litter production in forests of the world. Adv. Ecol. Res. 2 : 101-157.

Hughes, J.W., T.J. Fahey and B. Browne. 1987. A better seed and litter trap. Can. J. For. Res. 17 : 1623-1624.

Kim, C., K-S Koo, Y-K Kim, W-K Lee, J-H Jeong and H-S Seo. 1997. Dynamics of litterfall and nutrient inputs in Quercus acutissima and Pinus koraiensis stands. FRI. J. For. Sci. 55 : 1318.

Lisanework, N. and A. Michelsen 1994. Litterfall and nutrient release by decomposition in three plantations compared with a natural forest in the Ethiopian highland. For. Ecol. and Manage. 65 : 1489-164.

National Institute of Agricultural Science and Technology. 1988. Methods of Soil Chemical Analysis. Sammi Press, Suwon 450 pp.

van Vuuren, M.M.I. and van der Eerden, L.J. 1992. Effects of three rates of atmospheric nitrogen deposition enriched with 15N on litter decomposition in a heathland. Soil Biol. Biochem. 24: 527-532.