Soil carbon balance in a tropical grassland: Estimation of soil respiration and its partitioning using a semi-empirical model
In savannah and tropical grasslands, which account for 60% of grasslands worldwide, a large share of ecosystem carbon is located below ground due to high root:shoot ratios. Temporal variations in soil CO2 efflux (RS) were investigated in a grassland of coastal Congo over two years. The objectives were (1) to identify the main factors controlling seasonal variations in RS and (2) to develop a semi-empirical model describing RS and including a heterotrophic component (RH) and an autotrophic component (RA). Plant above-ground activity was found to exert strong control over soil respiration since 71% of seasonal RS variability was explained by the quantity of photosynthetically active radiation absorbed (APAR) by the grass canopy. We tested an additive model including a parameter enabling RS partitioning into RA and RH. Assumptions underlying this model were that RA mainly depended on the amount of photosynthates allocated below ground and that microbial and root activity was mostly controlled by soil temperature and soil moisture. The model provided a reasonably good prediction of seasonal variations in RS (R2 = 0.85) which varied between 5.4 _mol m?2 s?1 in the wet season and 0.9 _mol m?2 s?1 at the end of the dry season. The model was subsequently used to obtain annual estimates of RS, RA and RH. In accordance with results reported for other tropical grasslands, we estimated that RH accounted for 44% of RS, which represented a flux similar to the amount of carbon brought annually to the soil from below-ground litter production. Overall, this study opens up prospects for simulating the carbon budget of tropical grasslands on a large scale using remotely sensed data.
| Main Authors: | , , , , , , |
|---|---|
| Format: | article biblioteca |
| Language: | eng |
| Published: |
Elsevier
|
| Subjects: | U10 - Informatique, mathématiques et statistiques, P33 - Chimie et physique du sol, modèle de simulation, cycle du carbone, dioxyde de carbone, photosynthèse, respiration du sol, indice de surface foliaire, prairie, variation saisonnière, zone tropicale, http://aims.fao.org/aos/agrovoc/c_24242, http://aims.fao.org/aos/agrovoc/c_17299, http://aims.fao.org/aos/agrovoc/c_1302, http://aims.fao.org/aos/agrovoc/c_5812, http://aims.fao.org/aos/agrovoc/c_33553, http://aims.fao.org/aos/agrovoc/c_35196, http://aims.fao.org/aos/agrovoc/c_6154, http://aims.fao.org/aos/agrovoc/c_24894, http://aims.fao.org/aos/agrovoc/c_7979, http://aims.fao.org/aos/agrovoc/c_1811, |
| Online Access: | http://agritrop.cirad.fr/563825/ http://agritrop.cirad.fr/563825/1/document_563825.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In savannah and tropical grasslands, which account for 60% of grasslands worldwide, a large share of ecosystem carbon is located below ground due to high root:shoot ratios. Temporal variations in soil CO2 efflux (RS) were investigated in a grassland of coastal Congo over two years. The objectives were (1) to identify the main factors controlling seasonal variations in RS and (2) to develop a semi-empirical model describing RS and including a heterotrophic component (RH) and an autotrophic component (RA). Plant above-ground activity was found to exert strong control over soil respiration since 71% of seasonal RS variability was explained by the quantity of photosynthetically active radiation absorbed (APAR) by the grass canopy. We tested an additive model including a parameter enabling RS partitioning into RA and RH. Assumptions underlying this model were that RA mainly depended on the amount of photosynthates allocated below ground and that microbial and root activity was mostly controlled by soil temperature and soil moisture. The model provided a reasonably good prediction of seasonal variations in RS (R2 = 0.85) which varied between 5.4 _mol m?2 s?1 in the wet season and 0.9 _mol m?2 s?1 at the end of the dry season. The model was subsequently used to obtain annual estimates of RS, RA and RH. In accordance with results reported for other tropical grasslands, we estimated that RH accounted for 44% of RS, which represented a flux similar to the amount of carbon brought annually to the soil from below-ground litter production. Overall, this study opens up prospects for simulating the carbon budget of tropical grasslands on a large scale using remotely sensed data. |
|---|