Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Litter decomposition of six tree species on indigenous agroforestry farms in south-eastern Ethiopia in relation to litterfall carbon inputs and modelled soil respiration

Litter decomposition of six tree species on indigenous agroforestry farms in south-eastern... Agroforest Syst (2021) 95:755–766 https://doi.org/10.1007/s10457-021-00630-w(0123456789().,-volV)(0123456789().,-volV) Litter decomposition of six tree species on indigenous agroforestry farms in south-eastern Ethiopia in relation to litterfall carbon inputs and modelled soil respiration Mesele Negash Mike Starr Received: 17 November 2020 / Accepted: 3 April 2021 / Published online: 13 April 2021 The Author(s) 2021 Abstract The indigenous agroforestry systems prac- (Rh)to Rs ratio of 0.5, modelled Rh C losses were 89 to tised by smallholders in south-eastern Ethiopia have 238% of litterfall decomposition C losses estimated high biodiversity and productivity. However, little is using k values. However, using an Rh/Rs ratio of 0.27, known about their carbon (C) inputs and outputs. We which is appropriate for tropical humid forests, Rh C carried out a 1-year litterbag study to determine leaf losses were 11 to 138% of estimated litterfall decom- litter decomposition k constants for six woody species position C losses. Our decomposition and soil respi- common to these agroforestry systems. The k values ration estimates indicate that litterfall is sufficient to were then used to calculate the decomposition C losses maintain soil organic C contents and thereby the soil from measured litterfall C fluxes and the results fertility of these unique agroforestry systems. compared to modelled soil respiration (Rs) C losses. Litterbag weight loss at the end of the year was 100% Keywords Litterbags  Litterfall  Partial least -1 or nearly so, k values 2.582–6.108 (yr ) and half-life squares regression  Soil respiration  Tropics 41–112 days. k values were significantly (p = 0.023) correlated with litter N contents, nearly so with C/N ratios (p = 0.053), but not with other nutrients (Ca, Mg and K), and negatively correlated with temperature Introduction (p = 0.080). Using species, farm elevation, tempera- ture and litter quality as predictors, partial least Interest in agroforestry has greatly increased recently squares regression explained 48% of the variation in because of food security issues and because of its k. Depending on species, estimated decomposition C potential for carbon (C) sequestration and hence losses from litterfall were 18 to 58% lower than annual climate change mitigation (Thangata and Hildebrand litterfall C inputs. Using a heterotrophic respiration 2012; Mbow et al. 2014a, b). Although agroforestry is widely practised by smallholder farmers throughout Africa, there have been relatively few studies inves- M. Negash tigating the C cycle in African agroforestry systems in Wondo Genet College of Forestry and Natural Resources, terms of litterfall production, litter decomposition and Hawassa University, P.O. Box 128, Shashemene, Ethiopia e-mail: meselenegash72@gmail.com soil respiration. Litterfall production and quality in agroforestry M. Starr (&) systems, as in general, varies with many factors Department of Forest Sciences, University of Helsinki, including tree species, size and age, climate and P.O Box 27, 00014 Helsinki, Finland e-mail: mike.starr@helsinki.fi 123 756 Agroforest Syst (2021) 95:755–766 season, and soil and management practice (Isaac et al. climate (mean annual temperature). Using the k values 2005; Isaac and Nair 2006; Hairiah et al. 2006; Das and measured C contents we also calculated litterbag and Das 2010; Dawoe et al. 2010; Murovhi et al. decomposition losses of C and compared them with 2012). Studies on litter decomposition in tropical measured litterfall and modelled soil respiration C agroforestry systems in Asia and South America have fluxes, hypothesizing that litterbag C loss would show reported high but highly variable decomposition rates correspondence with both. This study is part of a larger that are related to litter quality and climatic factors study that has dealt with the C budget of indigenous (Hairiah et al. 2006; Hossain et al. 2011; Jairo et al. agroforestry systems in south-eastern Ethiopia (Ne- 2017; Petit-Aldana et al. 2019). The rates are compa- gash et al. 2012, 2013a,b; Negash and Starr rable to those reported for tropical forests (Powers 2013, 2015; Negash and Kanninen 2015). et al. 2009; Waring 2012). Agroforestry related litter decomposition studies carried out in eastern Africa have shown that leaf litter decomposition also greatly Material and methods differs among species, whether the trees are legumi- nous or not, the leaves are green or senesced, and on Study sites the season (Gindaba et al. 2004; Teklay 2004, 2007; Teklay and Malmer 2004; Mahari 2014; Abay 2018). The study was carried out in indigenous agroforestry Soil respiration CO effluxes are directly correlated systems practised on Rift Valley escarpment in the to litterfall fluxes and litter decomposition rates and Gedeo zone in south-eastern Ethiopia (68 10 N, 388 are driven by the same climatic factors (Raich and 20 E). The elevation of the 14 study sites (smallholder Schlesinger 1992; Raich and Tufekcioglu 2000; Bond- farms) ranged from 1524 to 2190 m a.s.l. According to Lamberty and Thomson 2010). Field measurements of the Ko¨ppen-Geiger classification system, the climate soil respiration rates from tropical agroforestry sys- is classified as tropical savanna with dry-winter tems are few (Bae et al 2013; Costa et al. 2018), but characteristics (As). The mean monthly temperature 0 00 reported rates are high and similar to those reported for recorded at Dilla weather station (6 22 49.6 N, 38 0 00 tropical humid forests (Oertel et al. 2016). Hetero- 18 24.9 E, 1515 m a.s.l.) is 18 C varying little trophic respiration, which measures the decomposi- during the year and the annual precipitation is about tion of soil organic matter, could be expected to show 1470 mm with a drier season from November to the most correspondence to litterfall and litter decom- March (monthly rainfall \ 100 mm) and two rainfall position C fluxes. peaks, generally in May and October (Fig. 1). The Agroforestry has a long history in Ethiopia (Asfaw soils in the study area are mainly Nitisols, i.e. and Nigatu 1995). The indigenous systems practised relatively fertile, deep, well-drained red tropical soils today by small-holders on the escarpment of the Rift with high clay and organic matter contents. For the Valley in south-eastern Ethiopia are unique and have developed from natural forest through gradual inten- sification of land use centuries ago (Negash and Achalu 2008). They are characterised by a high diversity of fruit tree, shrub species and herbaceous food crops, and are highly productive (Negash et al. 2012). The main objectives of this study were to determine the litter decomposition rates (k decay constants) of six woody species that are common in the indigenous agroforestry systems of south-eastern Ethiopia and to determine the dependence of k on litter quality factors and climate. We hypothesized that the k decay con- Fig. 1 Monthly mean (2005–2011; shaded) and 2011 (un- stants would differ among the six tree species because shaded) temperatures (lines) and rainfall (bars) recorded at Dilla of differences in the initial chemical composition of 0 00 0 00 weather station (6 22 49.6 N; 3818 24.9 E; 1515 m a.s.l.) the litter and that the k values would be correlated to (Ethiopian National Meteorology Agency) 123 Agroforest Syst (2021) 95:755–766 757 study farms, the soil texture of the 0–30 cm layer determined using the Kjeldahl method. Calcium (Ca), varied from loam to clay loam, clay contents from 17 magnesium (Mg) and potassium (K) contents were to 58%, pH (in water) from 4.93 to 7.83, organic determined after dry ashing and HCl acid digestion. carbon content from 2.2 to 7.7%, and nitrogen content Mg and Ca contents were determined using atomic from 0.12 to 0.43%. Agroforestry dominates the land- absorption spectrophotometry and K by flame pho- use (94.5%) in the area with the remainder comprising tometry. The analyses were performed in triplicate for grassland (1.4%), wetland (0.8%), natural forest each element and the mean values used in further (0.5%), plantations (0.1%) and others (2.7%). calculations. Leaf litterfall and decomposition (litterbag dry The monthly litterfall fluxes of C for each farm and weight loss) for four native tree species (Cordia species (LF ) were calculated as follows: Ci,j africana Lam., Croton macrostachyus Del., Erythrina LF ¼ LF C =100 ð1Þ Ci;j DWi;j j brucei Schweinf., Millettia ferruginea (Hochst.) Bak) and two non-native species (Mangifera indica L. and where LF is the oven-dry weight litterfall flux (g DWij -2 th Persea americana Mill.) were determined in the field. m ) for the i month and jth species and farm, and C These six species are the dominant species used in the is the annual C content (%) of litterfall for the jth agroforestry practised throughout the study area, species averaged across the farms. typically accounting for [ 70% of basal area and stem density and [ 80% of the crown area (Negash Litterbag incubation and k decay constants et al. 2012). They provide fruits, fodder, fuel, shade for crops, and improve the microclimate for the growth of Litterbags containing the combined monthly leaf coffee (Coffea arabica L.) and enset (Ensete vetrico- litterfall collected in 2010 were installed at 14 of the sum (Welw.) Cheesman). E. brucei and M. ferruginea farms on 1 January 2011. In the case of C. are also N-fixing species. macrostachyus, E. brucei, M. ferruginea and P. americana there were two farms and in the case of Litterfall C. africana and M. indica three farms (Table 1). The litterbags were prepared by inserting 200 g (± 0.01 g) Monthly litterfall was collected at seventeen small- of the oven-dried leaf litterfall material into holder farms during 2010 (Negash et al. 2013a, b). For 20 9 20 cm bags made from nylon netting having a each of the six tree species, two or three farms were mesh size of 1 mm. A labelled litterbag was installed selected (farms where the trees were healthy, mature, at each corner of each litterfall trap using a small spade and of good form). A litterfall trap was placed under to make an incision at an angle of 45 and carefully the canopy of two trees of the selected species at each inserting the bag to a depth of 15 cm. In total, 112 farm. The traps consisted of nylon netting (1 mm mesh litterbags were placed out. A bag at each trap was diameter) draped over four 1.5 m tall wooden poles retrieved on 31st March, 30th June, 30th September forming a 1 9 1 m collection area. To ensure com- and 31st December, i.e. after 89, 180, 272 and plete collection, a stone was placed in the centre of the 364 days respectively, and transported to the netting to weigh it down. At the end of each month the laboratory. litterfall, which was mainly foliage, was collected. In the laboratory, the bags were carefully cut open Litterfall other than leaves (e.g. small branches, and any ingrown roots and soil aggregates removed by flower, seeds) were removed before the remaining hand. The remaining litter was gently rinsed with tap leaves were air-dried for a day and then, oven-dried water to remove adhering soil particles, oven-dried (24 h at 65 C), weighed and stored. Monthly litterfall (48 h at 55 C) and weighed (± 0.01 g). The litter was -2 fluxes (g m ) for each trap were calculated from the then composited by species, ground into a fine powder oven-dry mass and trap catch area (1 m ). and three sub-samples analysed for organic C and N At the end of the year, the stored leaf litterfall contents using the same methods as described for samples were composited by species and a subsample litterfall. Species mean contents were used in subse- of each species taken and milled into a fine powder for quent calculations. chemical analysis. C contents were determined using the Walkley–Black method and nitrogen (N) contents 123 758 Agroforest Syst (2021) 95:755–766 Table 1 Litterfall (initial litterbag material) carbon and nutrient contents and ratios by species. Species having the same lower case letter are not significantly different (ANOVA followed by Tukey’s multiple comparison tests, a = 0.05) Species C, % N, % C/N Ca, % Mg, % K, % b ab Cordia africana 46.0 2.12 21.7 1.352 0.529 0.691 a a a a Croton macrostachyus 48.0 2.97 16.2 3.524 0.572 0.852 a ab a bc Erythrina brucei 46.8 2.98 15.7 3.817 0.920 0.604 b cd Mangifera indica 45.9 1.67 27.5 1.276 0.679 0.401 b b bc Millettia ferruginea 49.8 2.72 18.3 1.247 0.783 0.605 ab d Persea americana 50.4 1.40 36.0 1.914 0.731 0.330 -2 Decomposition constants were derived by fitting remaining litterfall C. The monthly C loss (g C m ) the following exponential decay equation to the from the litterfall through decomposition was then litterbag weight loss data for each trap: calculated as the difference between the monthly litterfall C flux and the estimated total amount of C kt W =W 100 ¼ 100e ð2Þ t 0 remaining at the end of the month: where W = dry weight of remaining litter at time t n t (year fraction), W = dry weight of litter at time 0, LF ¼ LF  ðLF Þð3Þ 0 Clossi;j Ci;j Cremaini;j ði¼iÞ and k = litter decomposition decay rate constant -1 (yr ). Equation 2 was fitted to the untransformed where LF = amount of litterfall C lost through Clossi,j litterbag percentage weight remaining values using th th decomposition at the end of the i month for the j nonlinear regression analysis and the Y-intercepts species and farm, LF = the amount of litterfall Cremaini,j forced through 100% (i.e. weight remaining at t =0) th th C remaining at the end of the i month for the j (Adair et al. 2010). th species and farm, and n = months up to the i month. Calculation of litterfall decomposition C loss Modelled soil respiration fluxes fluxes In the absence of measured soil respiration fluxes, we We used the litter decay constants to estimate how estimated soil respiration CO –C fluxes for each farm much of each month’s litterfall flux (mean of the two using the linear regression models presented by Fung traps) that was measured in 2010 would be remaining et al. (1987) and Raich and Schlesinger (1992). These at the end of each subsequent month until the end of simple models were used as they only require the year. These values were converted into monthly temperature data. The model by Fung et al. (1987)is -2 amounts of C remaining (g C m ) for each farm and for tropical/subtropical woody vegetation sites (Fung species using Eq. 1. C contents (%) of the litterfall et al. 1987): remaining at the end of each month were assumed to follow the mean C contents (%) of the litterbags Rs =Rs ¼ 0:78 T =T 0:14 r ¼ 0:64 ð4Þ i max i max measured during 2011. The C content of the litter -2 where Rs = mean monthly soil respiration (g C m remaining for the months between the measured at 0, per day) for the ith month, and Rs and T = max- max max 3, 6, 9 and 12 months was estimated using linear imum monthly soil respiration and temperature (C), interpolation. Each month’s litterfall was treated as a respectively. For Rs we used a value of 3.0, which max cohort such that the total amount of C in litterfall is based on values from Raich and Schlesinger (1992), -2 remaining (g C m ) at the end of each month was the Oertel et al. (2016) and Wood et al. (2013) for moist sum of the current and previous month’s amount of C (humid) tropical forests. remaining. For example, the total amount of C in the litterfall remaining in March was the sum of the amount of January’s, February’s and March’s 123 Agroforest Syst (2021) 95:755–766 759 The model by Raich and Schlesinger (1992)is we used JMP software (JMP, PRO 14.3.0. SAS global but data from tropical sites were included in the Institute Inc., Cary, NC, USA). derivation of the model (Raich and Schlesinger, 1992): Rs ¼ 25:6T þ 300 r ¼ 0:42 ð5Þ Results -2 where Rs = annual soil respiration (g C m ) and T = mean annual temperature ( C). Farm-wise 2011 Other than C and Mg contents, significant differences in the chemical characteristics of the litterfall (initial mean temperatures were calculated from Dilla 0 00 0 00 litter material) among the species were found weather station (6 22 49.6 N, 38 18 24.9 E, (Table 1). C. macrostachyus, E. brucei and M. 1515 m a.s.l., Fig. 1) data using an adiabatic lapse rate ferruginea had higher N contents and lower C/N correction value of 0.65 C per 100 m difference in ratios than the other species. C. macrostachyus and E. elevation. brucei also had the highest Ca contents, although the difference from other species was not significant. M. Statistical analysis indica and P. americana had the lowest K contents, the difference from other species being significant. Mg Significance testing for differences in the chemical contents did not significantly differ among the species. characteristics of the litterfall samples among the six tree species was carried out using ANOVA and Tukey The exponential decay model (Fig. 2) explained 91.4–99.8% of species litterbag weight loss (Table 2). multiple comparison tests (a B 0.05). Pearson corre- While there were too few data to allow for statistical lation coefficients were employed to describe the testing of differences among the species, k values degree of association between k and the individual varied from 2.582 (P. americana, farm 44) to 6.108 variables (litter quality, and farm elevation and 2011 (M. ferruginea, farm 13) (Table 2). The half-life (time mean annual temperature). Partial least squares to 50% weight loss) was the highest for M. indica regression, PLS-R, was used to examine the combined (95–112 days), and the lowest for M. ferruginea effect and relative importance of all explanatory (41–76 days) and E. brucei (52–67 days) (Table 2). variables on k. Species was entered as a nominal variable. The SIMPLS method and leave-one-out The measured amount of litterbag weight remaining at the end of the year varied from 0 to 12% (M. cross-validation method were implemented. PLS-R ferruginea, farm 10) and the modelled amounts from extracts a set of orthogonal latent factors (vectors) 0.2 to 11% (M. indica, farm 51) (Fig. 2). from the predictor variables regardless of any multi- The k values significantly increased with the N collinearity among them to predict the dependent content of the initial litter and correlations with litter variable. The lowest root mean predicted residual error initial C/N ratios (negative), farm elevation (positive) sum of squares (PRESS) value was used to identify the and temperature (negative) were nearly significant optimal number of factors to use. Variable importance (Table 3). Interestingly, the correlations between the in projection (VIP) coefficients for the predictor litterbag half-life values and the litter quality and variables reflect their importance to each of the latent environmental variables were stronger than those with factors and values [ 0.8 indicate an important vari- able. The size of the model regression coefficients the k values. The PLS-R analysis using species and the ten other (using centred and scaled data) indicates the relative importance of each predictor and its sign ( ±) predictor variables indicated a 1-factor model (PRESS = 1.0182) that could explain 48% of the indicates the direction of the relationship between it variation in the k values. The fit between the predicted and the latent factor. The goodness-of-fit of the actual and actual k values had an R value of 48% (Fig. 3a). and PLS-R predicted k values was examined by linear Of the six species only E. brucei, M. indica and M. regression and the coefficient of determination (R ). ferruginea had VIP values [ 0.8, and of those M. For the nonlinear regression, ANOVA and corre- indica and M. ferruginea had the greatest standardized lation analyses we used GraphPad PRISM software model coefficients (Fig. 3b). Elevation, temperature, (version 8.3.0 for Windows, GraphPad Software, San initial litter N, K and Mg contents, and C/N ratio had Diego, California, USA) and for the PLS-R analysis, 123 760 Agroforest Syst (2021) 95:755–766 Fig. 2 Weight of litterbag remaining (as percentage of initial weight) for each species and farm plotted against time (year fraction), and -kt the fitted exponential decay curves (y = 100 e ) calculated using nonlinear regression (k values are given in Table 2) Table 2 Farm elevation, mean temperature (2011), litterbag k decay constants, half-life (number of days until litterbag weight loss = 50%) and coefficient of determination (R ) by farm and species -1 2 Species farm no Elevation (m a.s.l.) Temp. (C) k (yr ) Half-life (days) R (%) Cordia africana 9 2263 15.8 4.294 59 99.8 Cordia africana 17 2241 16.0 2.780 91 95.6 Cordia africana 20 2248 15.9 2.908 87 91.4 Croton macrostachyus 8 2169 16.4 3.753 67 99.3 Croton macrostachyus 12 2115 16.8 3.245 78 98.4 Erythrina brucei 35 1904 18.2 3.775 67 98.6 Erythrina brucei 40 1912 18.1 4.881 52 98.5 Mangifera indica 45 1673 19.7 2.662 95 98.5 Mangifera indica 51 1571 20.3 2.248 112 96.0 Mangifera indica 52 1521 20.6 2.347 108 98.7 Millettia ferruginea 10 2262 15.8 3.315 76 97.8 Millettia ferruginea 13 2207 16.2 6.108 41 98.7 Persea americana 44 1695 19.5 2.582 98 95.7 Persea americana 50 1697 19.5 3.334 76 98.1 VIP values [ 0.8, indicating these were the most and 58% (M. indica) (Fig. 4). The modelled soil important predictors of k. respiration C losses at the end of the year were Litterfall C inputs were greater than the litterbag considerably greater than both litterfall C inputs and decomposition C losses for all six species, the litterbag decomposition C losses. In comparison to difference varying between 14% (E. brucei) litterbag C losses at the end of the year, Fung et al. 123 Agroforest Syst (2021) 95:755–766 761 Table 3 Pearson correlation coefficients (r) and two-tailed farm elevation and mean temperature (2011) and chemical p values describing the association between exponential decay characteristics of litter. Significant (a \ 0.05) coefficients are model parameters (k and half-life) for litterbag weight loss and in bold Decomposition parameter Elevation Temp C N C/N Ca Mg K R 0.485 - 0.483 0.287 0.602 - 0.526 0.267 0.361 0.366 p 0.079 0.080 0.320 0.023 0.053 0.356 0.205 0.199 Half-life R - 0.597 0.594 - 0.308 - 0.687 0.589 - 0.403 - 0.312 - 0.492 p 0.024 0.025 0.285 0.007 0.027 0.154 0.278 0.074 (Eq. 4) modelled Rs values were 123 to 510% greater for its deviation from the PLS-R predicted value was than litterbag C losses; and 14 to 29% greater than Rs the result of the litterbag weight loss recorded at values modelled using the Raich and Schlesinger 3 months. Excluding the 3-month weight loss data (1982) model (Eq. 5) (Fig. 4). Assuming an Rh/Rs resulted in a k value of 4.394, which compares well ratio of 0.5, Rh C losses using the Raich and with the PLS-R predicted value of 4.303. This Schlesinger (1982) model (Eq. 5) were similar to suggests that the k value for M. ferruginea recorded litterbag decomposition C losses for C. macrosta- at farm 13 as given in Table 2 is erroneous. Never- chyus, C. africana and E. brucei but greater for the theless, the overall high k values we found undoubt- other species. However, using an Rh/Rs ratio of 0.27 edly reflect the effect of the higher temperatures on microbial activity and decomposition in the tropics resulted in Rh values less than or similar to litterbag decomposition C losses. (Zhang et al. 2008). At the global (Zhang et al. 2008), European (Portillo-Estrada et al. 2016) and tropical (Salinas Discussion et al. 2010; Waring 2012; Bothwell et al. 2014) scales, k values are positively correlated to temperature. The rapid decomposition of the leaf litter and resulting Therefore, the nearly significant negative correlation high k values we found are typical for tropical forests between k and annual temperature (positive for (Bernhard-Reversat 1982; Powers et al. 2009) and elevation) and the significant correlation between agroforestry systems (Hossain et al. 2011; Petit- half-life and annual temperature (negative for half- Aldana et al. 2019; Jairo et al. 2017). The high k value life) were the opposite of that expected. However, the for M. ferruginea recorded at farm 13, and the reason direct correlation between soil organic matter Fig. 3 a Actual vs PLS-R predicted species k values (error bars are SEM; dotted line is the 1:1 line), and b PLS-R variable importance in projection (VIP) values vs model coefficients (using centred and scaled data; horizontal dotted line = VIP 0.8, values [ 0.8 indicate an important predictor) 123 762 Agroforest Syst (2021) 95:755–766 Fig. 4 Averaged cumulative monthly litterfall, litterbag litter (Eq. 3); soil respiration (Rs) C fluxes calculated using the decomposition loss and modelled soil respiration C fluxes for temperature-based models presented by Fung et al. (1987) for each of the six tree species studied. Litterfall C input values were subtropical/subtropical woody vegetation biome (Eq. 4) and for calculated from measured litterfall fluxes (Negash and Starr, global sites by Raich and Schlesinger (1982) (Eq. 5), and soil 2013) and measured C contents (Eq. 1); litterfall decomposition heterotrophic respiration (Rh) calculated from Raich and C losses were calculated from the litterfall fluxes, the litterbag Schlesinger (1982) Rs values assuming Rh/Rs ratios of 50% k decay constants and measured C contents of the remaining and 27% decomposition and temperature has been questioned. annual rainfall in the area is only some 1500 mm, it For example, the global decomposition rates of forest would suggest that the decomposition of our buried soil organic matter were shown to vary little with mean bags is faster than if they had been placed on the annual temperature (Giardina and Ryan 2000). surface. We do not have rainfall data at the farm scale, Besides temperature, precipitation and evapotranspi- but the variation among the farms is probably small ration have also been found to be important climatic and therefore also the variation in soil moisture. We, factors controlling decomposition at continental and therefore, consider it unlikely that moisture conditions global scales (Meentemeyer et al.1982; Aerts 1997; have had a controlling effect on our decomposition Zhang et al. 2008). In a pan-tropical decomposition rates. However, a possible explanation for the weak study, k values were found to be unrelated to mean negative dependence of k on temperature found in our annual temperature (MAT) but were strongly related study may be substrate availability and quality. to mean annual precipitation (MAP) (Powers et al. After a rapid loss of the more easily decomposable 2009). However, there was only a modest range in N-rich substances, decomposition becomes limited by MAT but a large range in MAP in their study. the lignin content (Giardina and Ryan 2000). Litter Whether the litterbags are placed on the surface or decomposition rates are known to decrease with lignin buried affects decomposition because of differences in contents (Zhang et al. 2008), and tropical forest litters temperature and moisture conditions. Powers et al. tend to have higher lignin contents than temperate and (2009) found that the decomposition of buried lit- subtropical litters (Bernhard-Reversat and Schwar terbags was faster than that of surface litterbags in 1997). Lignin contents were not determined in our drier (MAP \ 3000 mm) tropical forests. As the study and therefore we cannot assess its importance in 123 Agroforest Syst (2021) 95:755–766 763 controlling the observed decomposition rates. How- Lamberty and Thomson 2010). However, the Rh/Rs ever, the significant correlation between k and initial ratio may be much lower in the tropics than elsewhere. litter N content, the nearly significant correlation For instance, an average Rh/Rs ratio of 0.27 has been between k and C/N ratio (correlation with half-life reported for tropical humid evergreen forests (Oertel values were significant), and the results from the PLS- et al. 2016). Using this value to partition our Raich and R analysis all indicate that litter N contents played an Schlesinger (1982) modelled Rs values resulted in C important role in controlling decomposition rates. The loss values that were more similar to our litterfall dependence of litter decomposition on N contents, at decomposition C losses, suggesting that an Rh/Rs ratio least during the early stages, is widely known (Berg lower than 0.5 is more appropriate for the tropics. and Laskowski 2006). The litter N contents of the six Furthermore, the sum of the mean autotrophic and species are similar to values reported by others heterotrophic soil respiration values reported by (Schaffer and Gaye 1989; Thorp et al. 1997; Gindaba Oertel et al. (2016) for tropical humid evergreen -2 -1 et al. 2004; Tekay 2004; Tekay and Malmer 2004; forest (873 g C m yr ) is similar to our estimated -2 -1 Urban and Lechaudel 2005; Kotur and Keshava 2010, mean Rs value of 917 g C m yr based on litterfall Mahari 2014; Abay 2018). Interestingly, the highest production, litterbag C contents and decay k constants. k values were associated with E. brucei and M. The contribution of roots and belowground exudation ferruginea, species that are nitrogen fixers. of organic compounds to soil organic carbon contents C cycling in tropical forests has been considered to in addition to the incorporation of aboveground be more limited by litter contents of Ca, Mg and K than litterfall, and therefore to Rh, is unknown. However, by N (Cleveland et al. 2011; Waring, 2012). However, estimates of belowground biomass in the study farms the correlations between the litter contents of these are less than 25% of total biomass (Negash and Starr elements and k were not significant in our study. C. 2015). macrostachyus, E. brucei, and M. ferruginea are deciduous species while C. africana, M. indica and P. americana are evergreen species. The generally Conclusions greater k values for the former group of species compared to the latter group may thus be related to The exponential decay constants (k) describing the differences in leaf longevity and phenology (Cornwell weight loss of leaf litter for six woody species widely et al. 2008; Salinas et al. 2010). grown in the indigenous agroforestry systems of Our litterfall C inputs to the soil surface were south-eastern Ethiopia are high compared to global greater than the litterbag decomposition C losses for values. The losses of litterfall carbon through decom- all six species and modelled soil respiration C losses position calculated using the k values were less than substantially greater than both. The greater Rs values litterfall production inputs for each species, indicating calculated with the Fung et al. (1987) model compared a net input of carbon to the soil. However, actual litter to the Raich and Schlesinger (1982) model is at least decomposition losses are likely to be less than the partly related to the maximum monthly soil respiration calculated values because litter is removed by the (Rs , Eq. 4) value we used to covert the normalized farmers for fuel and by grazing animals. Comparing max values into absolute values for the farms. We used an soil respiration C effluxes, modelled using a simple -2 -1 Rs value of 3 g m day , which was based on temperature-based model, supported the notion that max soil respiration values presented by Raich and Sch- heterotrophic respiration accounts for considerably lesinger (1992), Oertel et al. (2016) and Wood et al. less of total soil respiration than the generally accepted (2013) for tropical forests. However, daily mean soil value of 50%. To better evaluate the C dynamics and respiration values across these studies varied between sequestration of these unique and diverse indigenous -2 -1 2.4 and 6.0 g C m day . Given the small variation agroforestry systems there is a clear need for longer- in monthly temperature at our sites, we consider the term studies and field-based measurements of soil -2 -1 Rs value of 3 g m day to be a reasonable value respiration. Nevertheless, efforts should be made to max for our purposes. prevent their conversion to cash crop production with Heterotrophic respiration is generally considered to the resulting loss of woody species diversity, soil account for 50–60% of soil total respiration (Bond- fertility and food security for the smallholder farmers. 123 764 Agroforest Syst (2021) 95:755–766 Acknowledgements We acknowledge the financial support Asfaw Z, Nigatu A (1995) Home-gardens in Ethiopia: Charac- for the first author from International Foundation for Science teristics and plant diversity. SINET Ethiopian J Sci (IFS Grt. No. D/4836-2), the Finnish Cultural Foundation, and 18(2):235–266 the Finnish Society of Forest Science. The Viikki Tropical Bae K, Lee DK, Fahey TJ et al (2013) Seasonal variation of soil Resource Institute (VITRI), Department of Forest Sciences, respiration rates in a secondary forest and agroforestry University of Helsinki, and the Wondo Genet College of systems. Agrofor Syst 87:131–139. https://doi.org/10. Forestry and Natural Resources, Hawassa University, Ethiopia 1007/s10457-012-9530-8 are also acknowledged for arranging logistics and laboratory Berg B, Laskowski R (2006) Litter decomposition: a guide to facilities for the study. We are also indebted to the farmers in carbon and nutrient turnover. Adv Ecol Res 38:1–421 Gedeo for allowing us to visit and carry out this study on their Bernhard-Reversat F (1982) Measuring litter decomposition in a farms and for providing marvellous experiences. tropical forest ecosystem: comparison of some methods. Int J Ecol Environ Sci 8:63–71 Bernhard-Reversat F, Schwar D (1997) Change in lignin content Authors’ contributions Negash—Field work and data during litter decomposition in tropical forest soils (Congo): collection, drafting of manuscript; Starr—statistical analysis comparison of exotic plantations and native stands. C R and drafting of manuscript. Acad Sci Ser IIA Earth Planet Sci 325:427–432 Bond-Lamberty B, Thomson A (2010) A global database of soil Funding Open access funding provided by University of respiration data. Biogeosciences 7:1915–1926. https://doi. Helsinki. Financial support for the first author (Negash) came org/10.5194/bg-7-1915-2010 from the International Foundation for Science (IFS Grt. No. Bothwell LD, Selmants PC, Giardina CP, Litton CM (2014) D/4836–2), the Finnish Cultural Foundation, and the Finnish Leaf litter decomposition rates increase with rising mean Society of Forest Science. annual temperature in Hawaiian tropical montane wet forests. PeerJ 2:e685. https://doi.org/10.7717/peerj.685 Availability of data and materials Available upon request Cleveland CC, Townsend AR, Taylor P, Alvarez-Clare S, from authors. Bustamante MMC, Chuyong G et al (2011) Relationships among net primary productivity, nutrients and climate in Code availability Not applicable. tropical rain forest: a pan-tropical analysis. Ecol Lett 14:939–947. https://doi.org/10.1111/j.1461-0248.2011. Declarations 01658.x Cornwell WK, Cornelissen JHC, Amatangelo K, Dorrepaal E, Conflicts of interest None. Eviner VT, Godoy O et al (2008) Plant species traits are the predominant control on litter decomposition rates within Open Access This article is licensed under a Creative Com- biomes worldwide. Ecol Lett 11:1065–1071. https://doi. mons Attribution 4.0 International License, which permits use, org/10.1111/j.1461-0248.2008.01219 sharing, adaptation, distribution and reproduction in any med- Costa ENDd, Landim de Souza MF, Lima Marrocos PC, Loba˜o ium or format, as long as you give appropriate credit to the D, Lopes da Silva DM (2018) Soil organic matter and CO original author(s) and the source, provide a link to the Creative fluxes in small tropical watersheds under forest and cacao Commons licence, and indicate if changes were made. The agroforestry. PLoS ONE 13(7):e0200550. https://doi.org/ images or other third party material in this article are included in 10.1371/journal.pone.0200550 the article’s Creative Commons licence, unless indicated Das T, Das AK (2010) Litter production and decomposition in otherwise in a credit line to the material. If material is not the forested areas of traditional home gardens: a case study included in the article’s Creative Commons licence and your from Barak Valley, Assam, northeast India. Agrofor Syst intended use is not permitted by statutory regulation or exceeds 79:157–170. https://doi.org/10.1007/s10457-010-9284-0 the permitted use, you will need to obtain permission directly Dawoe EK, Isaac ME, Quashie-Sam J (2010) Litterfall and litter from the copyright holder. To view a copy of this licence, visit nutrient dynamics under cocoa ecosystems in lowland http://creativecommons.org/licenses/by/4.0/. humid Ghana. Plant Soil 330:55–64. https://doi.org/10. 1007/s11104-009-0173-0 Fung IY, Tucker CJ, Prentice KC (1987) Application of advanced very high resolution radiometer vegetation index References to study atmosphere-biosphere exchange of CO2. J Geo- phys Res 92(D3):2999–3015. https://doi.org/10.1029/ Abay A (2018) Nitrogen release dynamics of Erythrina abys- JD092iD03p02999 sinica and Erythrina brucei litters as influenced by their Giardina CP, Ryan MG (2000) Evidence that decomposition biochemical composition. Afr J Plant Sci 2(12):331–340. rates of organic carbon in mineral soil do not vary with https://doi.org/10.5897/AJPS2018.1717 temperature. Nature 404:858–861 Adair EC, Hobbie SE, Hobbie RK (2010) Single-pool expo- Gindaba J, Olsson M, Itanna F (2004) Nutrient composition and nential decomposition models: potential pitfalls in their use short-term release from Croton macrostachyus Del. and in ecological studies. Ecology 91(4):1225–1236 Millettia ferruginea (Hochst.) Baker leaves. Biol Fertil Aerts R (1997) Climate, leaf litter chemistry and leaf litter Soils 40:393–397. https://doi.org/10.1007/s00374-004- decomposition in terrestrial ecosystems: a triangular rela- 0767-x tionship. Oikos 79:439–449 123 Agroforest Syst (2021) 95:755–766 765 Hairiah K, Sulistyani H, Suprayogo D, Widianto PP, Widodo 97(1–3):29–34. https://doi.org/10.1007/s10705-013-9590- RH, Van Noordwijk M (2006) Litter layer residence time in 9 forest and coffee agroforestry systems in Sumberjaya. Negash M, Starr M (2015) Biomass and soil carbon stocks of West Lampung For Ecol Manag 224(1–2):45–57. https:// indigenous agroforestry systems on the south-eastern Rift doi.org/10.1016/j.foreco.2005.12.007 Valley escarpment, Ethiopia. Plant Soil 393:95–107. Hossain M, Siddique MRH, Rahman S, Hossain Z, Hasan M https://doi.org/10.1007/s11104-015-2469-6 (2011) Nutrient dynamics associated with leaf litter Negash M, Starr M, Kanninen M (2013) Allometric equations decomposition of three agroforestry tree species (Azadir- for biomass estimation of Enset (Ensete ventricosum) achta indica, Dalbergia sissoo, and Melia azedarach)of grown in indigenous agroforestry systems in the Rift Val- Bangladesh. J For Res 22(4):577–582. https://doi.org/10. ley escarpment of southern-eastern Ethiopia. Agrofor Syst 1007/s11676-011-0175-7 87:571–581. https://doi.org/10.1007/s10457-012-9577-6 Isaac SR, Nair MA (2006) Litter dynamics of six multipurpose Negash M, Starr M, Kanninen M, Berhe L (2013) Allometric trees in a home garden in Southern Kerala, India. Agrofor equations for aboveground biomass of Coffea arabica L. Syst 67:203–213. https://doi.org/10.1007/s10457-005- grown in indigenous agroforestry systems in the Rift Val- 1107-3 ley escarpment of southern-eastern Ethiopia. Agrofor Syst Isaac ME, Gordon AM, Thevathasan N, Oppong SK, Quashie- 87:953–966. https://doi.org/10.1007/s10457-013-9611-3 Sam J (2005) Temporal changes in soil carbon and nitrogen Negash M, Yirdaw E, Luukkanen O (2012) Potential of in west African multistrata agroforestry systems: a indigenous multistrata agroforests for maintaining native chronosequence of pools and fluxes. Agrofor Syst floristic diversity in the south-eastern Rift Valley escarp- 65:23–31. https://doi.org/10.1007/s10457-004-4187-6 ment, Ethiopia. Agrofor Syst 85:9–28. https://doi.org/10. Jairo RM, Caicedo V, Jaimes Y (2017) Biomass decomposition 1007/s10457-011-9408-1 dynamic in agroforestry systems with Theobroma cacao L Oertel C, Matschullat J, Zurba K, Zimmermann F, Erasmi S in Rionegro Santander (Colombia). Agronomıa Colom- (2016) Greenhouse gas emissions from soils—a review. biana 35(2):182–189. https://doi.org/10.15446/agron. Chem Erde 76:327–352. https://doi.org/10.1016/j.chemer. colomb.v35n2.60981 2016.04.002 Kotur SC, Keshava Murthy SV (2010) Nutrient dynamics of Petit-Aldana J, Rahman MM, Parraguirre-Lezama C, Infante- annual growth-flush in mango (Mangifera indica L.). Cruz A, Romero-Arenas O (2019) Litter decomposition J Hortic Sci 5(1):75–77 process in coffee agroforestry systems. J Environ Sci Mahari A (2014) Leaf litter decomposition and nutrient release 35(2):121–139. https://doi.org/10.7747/JFES.2019.35.2. from Cordia africana Lam. and Croton Macrostachyus Del. 121 tree species. J Environ Earth Sci 4(15):1–7 Portillo-Estrada M, Pihlatie M, Korhonen JFJ, Levula J, Frumau Mbow C, Smith P, Skole D, Duguma L, Bustamante M (2014a) AKF, Ibrom A, Lembrechts JJ, Morillas L, Horva´th L, Achieving mitigation and adaptation to climate change Jones SK, Niinemets U (2016) Climatic controls on leaf through sustainable agroforestry practices in Africa. Curr litter decomposition across European forests and grass- Opin Environ Sustain 6:8–14. https://doi.org/10.1016/j. lands revealed by reciprocal litter transplantation experi- cosust.2013.09.002 ments. Biogeosciences 13:1621–1633. https://doi.org/10. Mbow C, Van Noordwijk M, Luedeling E, Neufeldt H, Minang 5194/bg-13-1621-2016 PA, Kowero G (2014) Agroforestry solutions to address Powers JS, Montgomery RA, Adair EC et al (2009) Decompo- food security and climate change challenges in Africa. Curr sition in tropical forests: a pan-tropical study of the effects Opin Environ Sustain 6:61–67. https://doi.org/10.1016/j. of litter type, litter placement and mesofaunal exclusion cosust.2013.10.014 across a precipitation gradient. J Ecol 97:801–811. https:// Meentemeyer V, Box EO, Thompson R (1982) World patterns doi.org/10.1111/j.1365-2745.2009.01515.x and amounts of terrestrial plant litter production. Bio- Raich JW, Schlesinger WH (1992) The global carbon dioxide sciences 32(2):125–128 flux in soil respiration and its relationship to vegetation and Murovhi NR, Materechera SA, Mulugeta SD (2012) Seasonal climate. Tellus 44B:81–99 changes in litter fall and its quality from three sub-tropical Raich JW, Tufekcioglu A (2000) Vegetation and soil respira- fruit tree species at Nelspruit South Africa. Agrofor Syst tion: correlations and controls. Biogeochemistry 86:61–71. https://doi.org/10.1007/s10457-012-9508-6 48(1):71–90 Negash M, Achalu N (2008) History of indigenous agro-forestry Salinas N, Malhi Y, Meir P, Silman M, Cuesta RR, Huaman J, in Gedeo, Ethiopia based on local community interviews: Salinas D, Huaman V, Gibaja A, Mamani M, Farfan F Vegetation diversity and structure in these land use sys- (2010) The sensitivity of tropical leaf litter decomposition tems. Ethiopian J Nat Resour 10(1):31–52 to temperature: results from a large-scale leaf translocation Negash M, Kanninen M (2015) Modeling biomass and soil experiment along an elevation gradient in Peruvian forests. carbon pools change of indigenous agroforestry system New Phytologist 189(4):967–977. https://doi.org/10.1111/ using CO2FIX approach. Agr Ecosyst Environ j.1469-8137.2010.03521.x 203:147–155. https://doi.org/10.1016/j.agee.2015.02. Schaffer B, Gaye GO (1989) Gas exchange, chlorophyll and 0040167-8809/a˜2 nitrogen content of mango leaves as influenced by light Negash M, Starr M (2013) Litterfall production and associated environment. HortScience 24(3):507–509 carbon and nitrogen fluxes of seven woody species grown Teklay T (2004) Seasonal dynamics in the concentrations of in indigenous agroforestry systems in the Rift Valley macronutrients and organic constituents in green and escarpment of Ethiopia. Nutr Cycl Agroecosyst 123 766 Agroforest Syst (2021) 95:755–766 senesced leaves of three agroforestry species in southern Urban L, Lechaudel M (2005) Effect of leaf-to-fruit ratio on leaf Ethiopia. Plant Soil 267:297–307 nitrogen content and net photosynthesis in girdled branches Teklay T (2007) Decomposition and nutrient release from of Mangifera indica L. Trees 19:564–571. https://doi.org/ pruning residues of two indigenous agroforestry species 10.1007/s00468-005-0415-6 during the wet and dry seasons. Nutrient Cycl Agroecosyst Waring BG (2012) A meta-analysis of climatic and chemical 77:115–126. https://doi.org/10.1007/s10705-006-9048-4 controls on leaf litter decay rates in tropical forests. Teklay T, Malmer A (2004) Decomposition of leaves from two Ecosystems 15:999–1009. https://doi.org/10.1007/s10021- indigenous trees of contrasting qualities under shaded 012-9561-z coffee and agricultural land-uses during the dry season at Wood TE, Detto M, Silver WL (2013) Sensitivity of soil res- Wondo Genet, Ethiopia. Soil Biol Biochem 36:777–786. piration to variability in soil moisture and temperature in a https://doi.org/10.1016/j.soilbio.2003.12.013 humid tropical forest. PLoS ONE 8(12):e80965. https:// Thangata PH, Hildebrand PE (2012) Carbon stock and seques- doi.org/10.1371/journal.pone.0080965 tration potential of agroforestry systems in smallholder Zhang D, Hui D, Luo Y, Zhou G (2008) Rates of litter decom- agroecosystems of sub-Saharan Africa: Mechanisms for position in terrestrial ecosystems: global patterns and ‘reducing emissions from deforestation and forest degra- controlling factors. J Plant Ecol 1(1):85–93. https://doi.org/ dation’ (REDD?). Agric Ecosyst Environ 158:172–183. 10.1093/jpe/rtn002 https://doi.org/10.1016/j.agee.2012.06.007 Thorp TG, Hutching D, Lowe T, Marsh KB (1997) Survey of Publisher’s Note Springer Nature remains neutral with fruit mineral concentrations and postharvest quality of regard to jurisdictional claims in published maps and New Zealand-grown ‘‘Hass’’ avocado (Persea americana institutional affiliations. Mill.). New Zealand J Crop Hortic Sci 25:251–260 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Agroforestry Systems Springer Journals

Litter decomposition of six tree species on indigenous agroforestry farms in south-eastern Ethiopia in relation to litterfall carbon inputs and modelled soil respiration

Negash, Mesele; Starr, Mike
Agroforestry Systems , Volume 95 (4) – Apr 13, 2021
Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/litter-decomposition-of-six-tree-species-on-indigenous-agroforestry-0tWPRBwzZa

References (47)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
0167-4366
eISSN
1572-9680
DOI
10.1007/s10457-021-00630-w
Publisher site
See Article on Publisher Site

Abstract

Agroforest Syst (2021) 95:755–766 https://doi.org/10.1007/s10457-021-00630-w(0123456789().,-volV)(0123456789().,-volV) Litter decomposition of six tree species on indigenous agroforestry farms in south-eastern Ethiopia in relation to litterfall carbon inputs and modelled soil respiration Mesele Negash Mike Starr Received: 17 November 2020 / Accepted: 3 April 2021 / Published online: 13 April 2021 The Author(s) 2021 Abstract The indigenous agroforestry systems prac- (Rh)to Rs ratio of 0.5, modelled Rh C losses were 89 to tised by smallholders in south-eastern Ethiopia have 238% of litterfall decomposition C losses estimated high biodiversity and productivity. However, little is using k values. However, using an Rh/Rs ratio of 0.27, known about their carbon (C) inputs and outputs. We which is appropriate for tropical humid forests, Rh C carried out a 1-year litterbag study to determine leaf losses were 11 to 138% of estimated litterfall decom- litter decomposition k constants for six woody species position C losses. Our decomposition and soil respi- common to these agroforestry systems. The k values ration estimates indicate that litterfall is sufficient to were then used to calculate the decomposition C losses maintain soil organic C contents and thereby the soil from measured litterfall C fluxes and the results fertility of these unique agroforestry systems. compared to modelled soil respiration (Rs) C losses. Litterbag weight loss at the end of the year was 100% Keywords Litterbags  Litterfall  Partial least -1 or nearly so, k values 2.582–6.108 (yr ) and half-life squares regression  Soil respiration  Tropics 41–112 days. k values were significantly (p = 0.023) correlated with litter N contents, nearly so with C/N ratios (p = 0.053), but not with other nutrients (Ca, Mg and K), and negatively correlated with temperature Introduction (p = 0.080). Using species, farm elevation, tempera- ture and litter quality as predictors, partial least Interest in agroforestry has greatly increased recently squares regression explained 48% of the variation in because of food security issues and because of its k. Depending on species, estimated decomposition C potential for carbon (C) sequestration and hence losses from litterfall were 18 to 58% lower than annual climate change mitigation (Thangata and Hildebrand litterfall C inputs. Using a heterotrophic respiration 2012; Mbow et al. 2014a, b). Although agroforestry is widely practised by smallholder farmers throughout Africa, there have been relatively few studies inves- M. Negash tigating the C cycle in African agroforestry systems in Wondo Genet College of Forestry and Natural Resources, terms of litterfall production, litter decomposition and Hawassa University, P.O. Box 128, Shashemene, Ethiopia e-mail: meselenegash72@gmail.com soil respiration. Litterfall production and quality in agroforestry M. Starr (&) systems, as in general, varies with many factors Department of Forest Sciences, University of Helsinki, including tree species, size and age, climate and P.O Box 27, 00014 Helsinki, Finland e-mail: mike.starr@helsinki.fi 123 756 Agroforest Syst (2021) 95:755–766 season, and soil and management practice (Isaac et al. climate (mean annual temperature). Using the k values 2005; Isaac and Nair 2006; Hairiah et al. 2006; Das and measured C contents we also calculated litterbag and Das 2010; Dawoe et al. 2010; Murovhi et al. decomposition losses of C and compared them with 2012). Studies on litter decomposition in tropical measured litterfall and modelled soil respiration C agroforestry systems in Asia and South America have fluxes, hypothesizing that litterbag C loss would show reported high but highly variable decomposition rates correspondence with both. This study is part of a larger that are related to litter quality and climatic factors study that has dealt with the C budget of indigenous (Hairiah et al. 2006; Hossain et al. 2011; Jairo et al. agroforestry systems in south-eastern Ethiopia (Ne- 2017; Petit-Aldana et al. 2019). The rates are compa- gash et al. 2012, 2013a,b; Negash and Starr rable to those reported for tropical forests (Powers 2013, 2015; Negash and Kanninen 2015). et al. 2009; Waring 2012). Agroforestry related litter decomposition studies carried out in eastern Africa have shown that leaf litter decomposition also greatly Material and methods differs among species, whether the trees are legumi- nous or not, the leaves are green or senesced, and on Study sites the season (Gindaba et al. 2004; Teklay 2004, 2007; Teklay and Malmer 2004; Mahari 2014; Abay 2018). The study was carried out in indigenous agroforestry Soil respiration CO effluxes are directly correlated systems practised on Rift Valley escarpment in the to litterfall fluxes and litter decomposition rates and Gedeo zone in south-eastern Ethiopia (68 10 N, 388 are driven by the same climatic factors (Raich and 20 E). The elevation of the 14 study sites (smallholder Schlesinger 1992; Raich and Tufekcioglu 2000; Bond- farms) ranged from 1524 to 2190 m a.s.l. According to Lamberty and Thomson 2010). Field measurements of the Ko¨ppen-Geiger classification system, the climate soil respiration rates from tropical agroforestry sys- is classified as tropical savanna with dry-winter tems are few (Bae et al 2013; Costa et al. 2018), but characteristics (As). The mean monthly temperature 0 00 reported rates are high and similar to those reported for recorded at Dilla weather station (6 22 49.6 N, 38 0 00 tropical humid forests (Oertel et al. 2016). Hetero- 18 24.9 E, 1515 m a.s.l.) is 18 C varying little trophic respiration, which measures the decomposi- during the year and the annual precipitation is about tion of soil organic matter, could be expected to show 1470 mm with a drier season from November to the most correspondence to litterfall and litter decom- March (monthly rainfall \ 100 mm) and two rainfall position C fluxes. peaks, generally in May and October (Fig. 1). The Agroforestry has a long history in Ethiopia (Asfaw soils in the study area are mainly Nitisols, i.e. and Nigatu 1995). The indigenous systems practised relatively fertile, deep, well-drained red tropical soils today by small-holders on the escarpment of the Rift with high clay and organic matter contents. For the Valley in south-eastern Ethiopia are unique and have developed from natural forest through gradual inten- sification of land use centuries ago (Negash and Achalu 2008). They are characterised by a high diversity of fruit tree, shrub species and herbaceous food crops, and are highly productive (Negash et al. 2012). The main objectives of this study were to determine the litter decomposition rates (k decay constants) of six woody species that are common in the indigenous agroforestry systems of south-eastern Ethiopia and to determine the dependence of k on litter quality factors and climate. We hypothesized that the k decay con- Fig. 1 Monthly mean (2005–2011; shaded) and 2011 (un- stants would differ among the six tree species because shaded) temperatures (lines) and rainfall (bars) recorded at Dilla of differences in the initial chemical composition of 0 00 0 00 weather station (6 22 49.6 N; 3818 24.9 E; 1515 m a.s.l.) the litter and that the k values would be correlated to (Ethiopian National Meteorology Agency) 123 Agroforest Syst (2021) 95:755–766 757 study farms, the soil texture of the 0–30 cm layer determined using the Kjeldahl method. Calcium (Ca), varied from loam to clay loam, clay contents from 17 magnesium (Mg) and potassium (K) contents were to 58%, pH (in water) from 4.93 to 7.83, organic determined after dry ashing and HCl acid digestion. carbon content from 2.2 to 7.7%, and nitrogen content Mg and Ca contents were determined using atomic from 0.12 to 0.43%. Agroforestry dominates the land- absorption spectrophotometry and K by flame pho- use (94.5%) in the area with the remainder comprising tometry. The analyses were performed in triplicate for grassland (1.4%), wetland (0.8%), natural forest each element and the mean values used in further (0.5%), plantations (0.1%) and others (2.7%). calculations. Leaf litterfall and decomposition (litterbag dry The monthly litterfall fluxes of C for each farm and weight loss) for four native tree species (Cordia species (LF ) were calculated as follows: Ci,j africana Lam., Croton macrostachyus Del., Erythrina LF ¼ LF C =100 ð1Þ Ci;j DWi;j j brucei Schweinf., Millettia ferruginea (Hochst.) Bak) and two non-native species (Mangifera indica L. and where LF is the oven-dry weight litterfall flux (g DWij -2 th Persea americana Mill.) were determined in the field. m ) for the i month and jth species and farm, and C These six species are the dominant species used in the is the annual C content (%) of litterfall for the jth agroforestry practised throughout the study area, species averaged across the farms. typically accounting for [ 70% of basal area and stem density and [ 80% of the crown area (Negash Litterbag incubation and k decay constants et al. 2012). They provide fruits, fodder, fuel, shade for crops, and improve the microclimate for the growth of Litterbags containing the combined monthly leaf coffee (Coffea arabica L.) and enset (Ensete vetrico- litterfall collected in 2010 were installed at 14 of the sum (Welw.) Cheesman). E. brucei and M. ferruginea farms on 1 January 2011. In the case of C. are also N-fixing species. macrostachyus, E. brucei, M. ferruginea and P. americana there were two farms and in the case of Litterfall C. africana and M. indica three farms (Table 1). The litterbags were prepared by inserting 200 g (± 0.01 g) Monthly litterfall was collected at seventeen small- of the oven-dried leaf litterfall material into holder farms during 2010 (Negash et al. 2013a, b). For 20 9 20 cm bags made from nylon netting having a each of the six tree species, two or three farms were mesh size of 1 mm. A labelled litterbag was installed selected (farms where the trees were healthy, mature, at each corner of each litterfall trap using a small spade and of good form). A litterfall trap was placed under to make an incision at an angle of 45 and carefully the canopy of two trees of the selected species at each inserting the bag to a depth of 15 cm. In total, 112 farm. The traps consisted of nylon netting (1 mm mesh litterbags were placed out. A bag at each trap was diameter) draped over four 1.5 m tall wooden poles retrieved on 31st March, 30th June, 30th September forming a 1 9 1 m collection area. To ensure com- and 31st December, i.e. after 89, 180, 272 and plete collection, a stone was placed in the centre of the 364 days respectively, and transported to the netting to weigh it down. At the end of each month the laboratory. litterfall, which was mainly foliage, was collected. In the laboratory, the bags were carefully cut open Litterfall other than leaves (e.g. small branches, and any ingrown roots and soil aggregates removed by flower, seeds) were removed before the remaining hand. The remaining litter was gently rinsed with tap leaves were air-dried for a day and then, oven-dried water to remove adhering soil particles, oven-dried (24 h at 65 C), weighed and stored. Monthly litterfall (48 h at 55 C) and weighed (± 0.01 g). The litter was -2 fluxes (g m ) for each trap were calculated from the then composited by species, ground into a fine powder oven-dry mass and trap catch area (1 m ). and three sub-samples analysed for organic C and N At the end of the year, the stored leaf litterfall contents using the same methods as described for samples were composited by species and a subsample litterfall. Species mean contents were used in subse- of each species taken and milled into a fine powder for quent calculations. chemical analysis. C contents were determined using the Walkley–Black method and nitrogen (N) contents 123 758 Agroforest Syst (2021) 95:755–766 Table 1 Litterfall (initial litterbag material) carbon and nutrient contents and ratios by species. Species having the same lower case letter are not significantly different (ANOVA followed by Tukey’s multiple comparison tests, a = 0.05) Species C, % N, % C/N Ca, % Mg, % K, % b ab Cordia africana 46.0 2.12 21.7 1.352 0.529 0.691 a a a a Croton macrostachyus 48.0 2.97 16.2 3.524 0.572 0.852 a ab a bc Erythrina brucei 46.8 2.98 15.7 3.817 0.920 0.604 b cd Mangifera indica 45.9 1.67 27.5 1.276 0.679 0.401 b b bc Millettia ferruginea 49.8 2.72 18.3 1.247 0.783 0.605 ab d Persea americana 50.4 1.40 36.0 1.914 0.731 0.330 -2 Decomposition constants were derived by fitting remaining litterfall C. The monthly C loss (g C m ) the following exponential decay equation to the from the litterfall through decomposition was then litterbag weight loss data for each trap: calculated as the difference between the monthly litterfall C flux and the estimated total amount of C kt W =W 100 ¼ 100e ð2Þ t 0 remaining at the end of the month: where W = dry weight of remaining litter at time t n t (year fraction), W = dry weight of litter at time 0, LF ¼ LF  ðLF Þð3Þ 0 Clossi;j Ci;j Cremaini;j ði¼iÞ and k = litter decomposition decay rate constant -1 (yr ). Equation 2 was fitted to the untransformed where LF = amount of litterfall C lost through Clossi,j litterbag percentage weight remaining values using th th decomposition at the end of the i month for the j nonlinear regression analysis and the Y-intercepts species and farm, LF = the amount of litterfall Cremaini,j forced through 100% (i.e. weight remaining at t =0) th th C remaining at the end of the i month for the j (Adair et al. 2010). th species and farm, and n = months up to the i month. Calculation of litterfall decomposition C loss Modelled soil respiration fluxes fluxes In the absence of measured soil respiration fluxes, we We used the litter decay constants to estimate how estimated soil respiration CO –C fluxes for each farm much of each month’s litterfall flux (mean of the two using the linear regression models presented by Fung traps) that was measured in 2010 would be remaining et al. (1987) and Raich and Schlesinger (1992). These at the end of each subsequent month until the end of simple models were used as they only require the year. These values were converted into monthly temperature data. The model by Fung et al. (1987)is -2 amounts of C remaining (g C m ) for each farm and for tropical/subtropical woody vegetation sites (Fung species using Eq. 1. C contents (%) of the litterfall et al. 1987): remaining at the end of each month were assumed to follow the mean C contents (%) of the litterbags Rs =Rs ¼ 0:78 T =T 0:14 r ¼ 0:64 ð4Þ i max i max measured during 2011. The C content of the litter -2 where Rs = mean monthly soil respiration (g C m remaining for the months between the measured at 0, per day) for the ith month, and Rs and T = max- max max 3, 6, 9 and 12 months was estimated using linear imum monthly soil respiration and temperature (C), interpolation. Each month’s litterfall was treated as a respectively. For Rs we used a value of 3.0, which max cohort such that the total amount of C in litterfall is based on values from Raich and Schlesinger (1992), -2 remaining (g C m ) at the end of each month was the Oertel et al. (2016) and Wood et al. (2013) for moist sum of the current and previous month’s amount of C (humid) tropical forests. remaining. For example, the total amount of C in the litterfall remaining in March was the sum of the amount of January’s, February’s and March’s 123 Agroforest Syst (2021) 95:755–766 759 The model by Raich and Schlesinger (1992)is we used JMP software (JMP, PRO 14.3.0. SAS global but data from tropical sites were included in the Institute Inc., Cary, NC, USA). derivation of the model (Raich and Schlesinger, 1992): Rs ¼ 25:6T þ 300 r ¼ 0:42 ð5Þ Results -2 where Rs = annual soil respiration (g C m ) and T = mean annual temperature ( C). Farm-wise 2011 Other than C and Mg contents, significant differences in the chemical characteristics of the litterfall (initial mean temperatures were calculated from Dilla 0 00 0 00 litter material) among the species were found weather station (6 22 49.6 N, 38 18 24.9 E, (Table 1). C. macrostachyus, E. brucei and M. 1515 m a.s.l., Fig. 1) data using an adiabatic lapse rate ferruginea had higher N contents and lower C/N correction value of 0.65 C per 100 m difference in ratios than the other species. C. macrostachyus and E. elevation. brucei also had the highest Ca contents, although the difference from other species was not significant. M. Statistical analysis indica and P. americana had the lowest K contents, the difference from other species being significant. Mg Significance testing for differences in the chemical contents did not significantly differ among the species. characteristics of the litterfall samples among the six tree species was carried out using ANOVA and Tukey The exponential decay model (Fig. 2) explained 91.4–99.8% of species litterbag weight loss (Table 2). multiple comparison tests (a B 0.05). Pearson corre- While there were too few data to allow for statistical lation coefficients were employed to describe the testing of differences among the species, k values degree of association between k and the individual varied from 2.582 (P. americana, farm 44) to 6.108 variables (litter quality, and farm elevation and 2011 (M. ferruginea, farm 13) (Table 2). The half-life (time mean annual temperature). Partial least squares to 50% weight loss) was the highest for M. indica regression, PLS-R, was used to examine the combined (95–112 days), and the lowest for M. ferruginea effect and relative importance of all explanatory (41–76 days) and E. brucei (52–67 days) (Table 2). variables on k. Species was entered as a nominal variable. The SIMPLS method and leave-one-out The measured amount of litterbag weight remaining at the end of the year varied from 0 to 12% (M. cross-validation method were implemented. PLS-R ferruginea, farm 10) and the modelled amounts from extracts a set of orthogonal latent factors (vectors) 0.2 to 11% (M. indica, farm 51) (Fig. 2). from the predictor variables regardless of any multi- The k values significantly increased with the N collinearity among them to predict the dependent content of the initial litter and correlations with litter variable. The lowest root mean predicted residual error initial C/N ratios (negative), farm elevation (positive) sum of squares (PRESS) value was used to identify the and temperature (negative) were nearly significant optimal number of factors to use. Variable importance (Table 3). Interestingly, the correlations between the in projection (VIP) coefficients for the predictor litterbag half-life values and the litter quality and variables reflect their importance to each of the latent environmental variables were stronger than those with factors and values [ 0.8 indicate an important vari- able. The size of the model regression coefficients the k values. The PLS-R analysis using species and the ten other (using centred and scaled data) indicates the relative importance of each predictor and its sign ( ±) predictor variables indicated a 1-factor model (PRESS = 1.0182) that could explain 48% of the indicates the direction of the relationship between it variation in the k values. The fit between the predicted and the latent factor. The goodness-of-fit of the actual and actual k values had an R value of 48% (Fig. 3a). and PLS-R predicted k values was examined by linear Of the six species only E. brucei, M. indica and M. regression and the coefficient of determination (R ). ferruginea had VIP values [ 0.8, and of those M. For the nonlinear regression, ANOVA and corre- indica and M. ferruginea had the greatest standardized lation analyses we used GraphPad PRISM software model coefficients (Fig. 3b). Elevation, temperature, (version 8.3.0 for Windows, GraphPad Software, San initial litter N, K and Mg contents, and C/N ratio had Diego, California, USA) and for the PLS-R analysis, 123 760 Agroforest Syst (2021) 95:755–766 Fig. 2 Weight of litterbag remaining (as percentage of initial weight) for each species and farm plotted against time (year fraction), and -kt the fitted exponential decay curves (y = 100 e ) calculated using nonlinear regression (k values are given in Table 2) Table 2 Farm elevation, mean temperature (2011), litterbag k decay constants, half-life (number of days until litterbag weight loss = 50%) and coefficient of determination (R ) by farm and species -1 2 Species farm no Elevation (m a.s.l.) Temp. (C) k (yr ) Half-life (days) R (%) Cordia africana 9 2263 15.8 4.294 59 99.8 Cordia africana 17 2241 16.0 2.780 91 95.6 Cordia africana 20 2248 15.9 2.908 87 91.4 Croton macrostachyus 8 2169 16.4 3.753 67 99.3 Croton macrostachyus 12 2115 16.8 3.245 78 98.4 Erythrina brucei 35 1904 18.2 3.775 67 98.6 Erythrina brucei 40 1912 18.1 4.881 52 98.5 Mangifera indica 45 1673 19.7 2.662 95 98.5 Mangifera indica 51 1571 20.3 2.248 112 96.0 Mangifera indica 52 1521 20.6 2.347 108 98.7 Millettia ferruginea 10 2262 15.8 3.315 76 97.8 Millettia ferruginea 13 2207 16.2 6.108 41 98.7 Persea americana 44 1695 19.5 2.582 98 95.7 Persea americana 50 1697 19.5 3.334 76 98.1 VIP values [ 0.8, indicating these were the most and 58% (M. indica) (Fig. 4). The modelled soil important predictors of k. respiration C losses at the end of the year were Litterfall C inputs were greater than the litterbag considerably greater than both litterfall C inputs and decomposition C losses for all six species, the litterbag decomposition C losses. In comparison to difference varying between 14% (E. brucei) litterbag C losses at the end of the year, Fung et al. 123 Agroforest Syst (2021) 95:755–766 761 Table 3 Pearson correlation coefficients (r) and two-tailed farm elevation and mean temperature (2011) and chemical p values describing the association between exponential decay characteristics of litter. Significant (a \ 0.05) coefficients are model parameters (k and half-life) for litterbag weight loss and in bold Decomposition parameter Elevation Temp C N C/N Ca Mg K R 0.485 - 0.483 0.287 0.602 - 0.526 0.267 0.361 0.366 p 0.079 0.080 0.320 0.023 0.053 0.356 0.205 0.199 Half-life R - 0.597 0.594 - 0.308 - 0.687 0.589 - 0.403 - 0.312 - 0.492 p 0.024 0.025 0.285 0.007 0.027 0.154 0.278 0.074 (Eq. 4) modelled Rs values were 123 to 510% greater for its deviation from the PLS-R predicted value was than litterbag C losses; and 14 to 29% greater than Rs the result of the litterbag weight loss recorded at values modelled using the Raich and Schlesinger 3 months. Excluding the 3-month weight loss data (1982) model (Eq. 5) (Fig. 4). Assuming an Rh/Rs resulted in a k value of 4.394, which compares well ratio of 0.5, Rh C losses using the Raich and with the PLS-R predicted value of 4.303. This Schlesinger (1982) model (Eq. 5) were similar to suggests that the k value for M. ferruginea recorded litterbag decomposition C losses for C. macrosta- at farm 13 as given in Table 2 is erroneous. Never- chyus, C. africana and E. brucei but greater for the theless, the overall high k values we found undoubt- other species. However, using an Rh/Rs ratio of 0.27 edly reflect the effect of the higher temperatures on microbial activity and decomposition in the tropics resulted in Rh values less than or similar to litterbag decomposition C losses. (Zhang et al. 2008). At the global (Zhang et al. 2008), European (Portillo-Estrada et al. 2016) and tropical (Salinas Discussion et al. 2010; Waring 2012; Bothwell et al. 2014) scales, k values are positively correlated to temperature. The rapid decomposition of the leaf litter and resulting Therefore, the nearly significant negative correlation high k values we found are typical for tropical forests between k and annual temperature (positive for (Bernhard-Reversat 1982; Powers et al. 2009) and elevation) and the significant correlation between agroforestry systems (Hossain et al. 2011; Petit- half-life and annual temperature (negative for half- Aldana et al. 2019; Jairo et al. 2017). The high k value life) were the opposite of that expected. However, the for M. ferruginea recorded at farm 13, and the reason direct correlation between soil organic matter Fig. 3 a Actual vs PLS-R predicted species k values (error bars are SEM; dotted line is the 1:1 line), and b PLS-R variable importance in projection (VIP) values vs model coefficients (using centred and scaled data; horizontal dotted line = VIP 0.8, values [ 0.8 indicate an important predictor) 123 762 Agroforest Syst (2021) 95:755–766 Fig. 4 Averaged cumulative monthly litterfall, litterbag litter (Eq. 3); soil respiration (Rs) C fluxes calculated using the decomposition loss and modelled soil respiration C fluxes for temperature-based models presented by Fung et al. (1987) for each of the six tree species studied. Litterfall C input values were subtropical/subtropical woody vegetation biome (Eq. 4) and for calculated from measured litterfall fluxes (Negash and Starr, global sites by Raich and Schlesinger (1982) (Eq. 5), and soil 2013) and measured C contents (Eq. 1); litterfall decomposition heterotrophic respiration (Rh) calculated from Raich and C losses were calculated from the litterfall fluxes, the litterbag Schlesinger (1982) Rs values assuming Rh/Rs ratios of 50% k decay constants and measured C contents of the remaining and 27% decomposition and temperature has been questioned. annual rainfall in the area is only some 1500 mm, it For example, the global decomposition rates of forest would suggest that the decomposition of our buried soil organic matter were shown to vary little with mean bags is faster than if they had been placed on the annual temperature (Giardina and Ryan 2000). surface. We do not have rainfall data at the farm scale, Besides temperature, precipitation and evapotranspi- but the variation among the farms is probably small ration have also been found to be important climatic and therefore also the variation in soil moisture. We, factors controlling decomposition at continental and therefore, consider it unlikely that moisture conditions global scales (Meentemeyer et al.1982; Aerts 1997; have had a controlling effect on our decomposition Zhang et al. 2008). In a pan-tropical decomposition rates. However, a possible explanation for the weak study, k values were found to be unrelated to mean negative dependence of k on temperature found in our annual temperature (MAT) but were strongly related study may be substrate availability and quality. to mean annual precipitation (MAP) (Powers et al. After a rapid loss of the more easily decomposable 2009). However, there was only a modest range in N-rich substances, decomposition becomes limited by MAT but a large range in MAP in their study. the lignin content (Giardina and Ryan 2000). Litter Whether the litterbags are placed on the surface or decomposition rates are known to decrease with lignin buried affects decomposition because of differences in contents (Zhang et al. 2008), and tropical forest litters temperature and moisture conditions. Powers et al. tend to have higher lignin contents than temperate and (2009) found that the decomposition of buried lit- subtropical litters (Bernhard-Reversat and Schwar terbags was faster than that of surface litterbags in 1997). Lignin contents were not determined in our drier (MAP \ 3000 mm) tropical forests. As the study and therefore we cannot assess its importance in 123 Agroforest Syst (2021) 95:755–766 763 controlling the observed decomposition rates. How- Lamberty and Thomson 2010). However, the Rh/Rs ever, the significant correlation between k and initial ratio may be much lower in the tropics than elsewhere. litter N content, the nearly significant correlation For instance, an average Rh/Rs ratio of 0.27 has been between k and C/N ratio (correlation with half-life reported for tropical humid evergreen forests (Oertel values were significant), and the results from the PLS- et al. 2016). Using this value to partition our Raich and R analysis all indicate that litter N contents played an Schlesinger (1982) modelled Rs values resulted in C important role in controlling decomposition rates. The loss values that were more similar to our litterfall dependence of litter decomposition on N contents, at decomposition C losses, suggesting that an Rh/Rs ratio least during the early stages, is widely known (Berg lower than 0.5 is more appropriate for the tropics. and Laskowski 2006). The litter N contents of the six Furthermore, the sum of the mean autotrophic and species are similar to values reported by others heterotrophic soil respiration values reported by (Schaffer and Gaye 1989; Thorp et al. 1997; Gindaba Oertel et al. (2016) for tropical humid evergreen -2 -1 et al. 2004; Tekay 2004; Tekay and Malmer 2004; forest (873 g C m yr ) is similar to our estimated -2 -1 Urban and Lechaudel 2005; Kotur and Keshava 2010, mean Rs value of 917 g C m yr based on litterfall Mahari 2014; Abay 2018). Interestingly, the highest production, litterbag C contents and decay k constants. k values were associated with E. brucei and M. The contribution of roots and belowground exudation ferruginea, species that are nitrogen fixers. of organic compounds to soil organic carbon contents C cycling in tropical forests has been considered to in addition to the incorporation of aboveground be more limited by litter contents of Ca, Mg and K than litterfall, and therefore to Rh, is unknown. However, by N (Cleveland et al. 2011; Waring, 2012). However, estimates of belowground biomass in the study farms the correlations between the litter contents of these are less than 25% of total biomass (Negash and Starr elements and k were not significant in our study. C. 2015). macrostachyus, E. brucei, and M. ferruginea are deciduous species while C. africana, M. indica and P. americana are evergreen species. The generally Conclusions greater k values for the former group of species compared to the latter group may thus be related to The exponential decay constants (k) describing the differences in leaf longevity and phenology (Cornwell weight loss of leaf litter for six woody species widely et al. 2008; Salinas et al. 2010). grown in the indigenous agroforestry systems of Our litterfall C inputs to the soil surface were south-eastern Ethiopia are high compared to global greater than the litterbag decomposition C losses for values. The losses of litterfall carbon through decom- all six species and modelled soil respiration C losses position calculated using the k values were less than substantially greater than both. The greater Rs values litterfall production inputs for each species, indicating calculated with the Fung et al. (1987) model compared a net input of carbon to the soil. However, actual litter to the Raich and Schlesinger (1982) model is at least decomposition losses are likely to be less than the partly related to the maximum monthly soil respiration calculated values because litter is removed by the (Rs , Eq. 4) value we used to covert the normalized farmers for fuel and by grazing animals. Comparing max values into absolute values for the farms. We used an soil respiration C effluxes, modelled using a simple -2 -1 Rs value of 3 g m day , which was based on temperature-based model, supported the notion that max soil respiration values presented by Raich and Sch- heterotrophic respiration accounts for considerably lesinger (1992), Oertel et al. (2016) and Wood et al. less of total soil respiration than the generally accepted (2013) for tropical forests. However, daily mean soil value of 50%. To better evaluate the C dynamics and respiration values across these studies varied between sequestration of these unique and diverse indigenous -2 -1 2.4 and 6.0 g C m day . Given the small variation agroforestry systems there is a clear need for longer- in monthly temperature at our sites, we consider the term studies and field-based measurements of soil -2 -1 Rs value of 3 g m day to be a reasonable value respiration. Nevertheless, efforts should be made to max for our purposes. prevent their conversion to cash crop production with Heterotrophic respiration is generally considered to the resulting loss of woody species diversity, soil account for 50–60% of soil total respiration (Bond- fertility and food security for the smallholder farmers. 123 764 Agroforest Syst (2021) 95:755–766 Acknowledgements We acknowledge the financial support Asfaw Z, Nigatu A (1995) Home-gardens in Ethiopia: Charac- for the first author from International Foundation for Science teristics and plant diversity. SINET Ethiopian J Sci (IFS Grt. No. D/4836-2), the Finnish Cultural Foundation, and 18(2):235–266 the Finnish Society of Forest Science. The Viikki Tropical Bae K, Lee DK, Fahey TJ et al (2013) Seasonal variation of soil Resource Institute (VITRI), Department of Forest Sciences, respiration rates in a secondary forest and agroforestry University of Helsinki, and the Wondo Genet College of systems. Agrofor Syst 87:131–139. https://doi.org/10. Forestry and Natural Resources, Hawassa University, Ethiopia 1007/s10457-012-9530-8 are also acknowledged for arranging logistics and laboratory Berg B, Laskowski R (2006) Litter decomposition: a guide to facilities for the study. We are also indebted to the farmers in carbon and nutrient turnover. Adv Ecol Res 38:1–421 Gedeo for allowing us to visit and carry out this study on their Bernhard-Reversat F (1982) Measuring litter decomposition in a farms and for providing marvellous experiences. tropical forest ecosystem: comparison of some methods. Int J Ecol Environ Sci 8:63–71 Bernhard-Reversat F, Schwar D (1997) Change in lignin content Authors’ contributions Negash—Field work and data during litter decomposition in tropical forest soils (Congo): collection, drafting of manuscript; Starr—statistical analysis comparison of exotic plantations and native stands. C R and drafting of manuscript. Acad Sci Ser IIA Earth Planet Sci 325:427–432 Bond-Lamberty B, Thomson A (2010) A global database of soil Funding Open access funding provided by University of respiration data. Biogeosciences 7:1915–1926. https://doi. Helsinki. Financial support for the first author (Negash) came org/10.5194/bg-7-1915-2010 from the International Foundation for Science (IFS Grt. No. Bothwell LD, Selmants PC, Giardina CP, Litton CM (2014) D/4836–2), the Finnish Cultural Foundation, and the Finnish Leaf litter decomposition rates increase with rising mean Society of Forest Science. annual temperature in Hawaiian tropical montane wet forests. PeerJ 2:e685. https://doi.org/10.7717/peerj.685 Availability of data and materials Available upon request Cleveland CC, Townsend AR, Taylor P, Alvarez-Clare S, from authors. Bustamante MMC, Chuyong G et al (2011) Relationships among net primary productivity, nutrients and climate in Code availability Not applicable. tropical rain forest: a pan-tropical analysis. Ecol Lett 14:939–947. https://doi.org/10.1111/j.1461-0248.2011. Declarations 01658.x Cornwell WK, Cornelissen JHC, Amatangelo K, Dorrepaal E, Conflicts of interest None. Eviner VT, Godoy O et al (2008) Plant species traits are the predominant control on litter decomposition rates within Open Access This article is licensed under a Creative Com- biomes worldwide. Ecol Lett 11:1065–1071. https://doi. mons Attribution 4.0 International License, which permits use, org/10.1111/j.1461-0248.2008.01219 sharing, adaptation, distribution and reproduction in any med- Costa ENDd, Landim de Souza MF, Lima Marrocos PC, Loba˜o ium or format, as long as you give appropriate credit to the D, Lopes da Silva DM (2018) Soil organic matter and CO original author(s) and the source, provide a link to the Creative fluxes in small tropical watersheds under forest and cacao Commons licence, and indicate if changes were made. The agroforestry. PLoS ONE 13(7):e0200550. https://doi.org/ images or other third party material in this article are included in 10.1371/journal.pone.0200550 the article’s Creative Commons licence, unless indicated Das T, Das AK (2010) Litter production and decomposition in otherwise in a credit line to the material. If material is not the forested areas of traditional home gardens: a case study included in the article’s Creative Commons licence and your from Barak Valley, Assam, northeast India. Agrofor Syst intended use is not permitted by statutory regulation or exceeds 79:157–170. https://doi.org/10.1007/s10457-010-9284-0 the permitted use, you will need to obtain permission directly Dawoe EK, Isaac ME, Quashie-Sam J (2010) Litterfall and litter from the copyright holder. To view a copy of this licence, visit nutrient dynamics under cocoa ecosystems in lowland http://creativecommons.org/licenses/by/4.0/. humid Ghana. Plant Soil 330:55–64. https://doi.org/10. 1007/s11104-009-0173-0 Fung IY, Tucker CJ, Prentice KC (1987) Application of advanced very high resolution radiometer vegetation index References to study atmosphere-biosphere exchange of CO2. J Geo- phys Res 92(D3):2999–3015. https://doi.org/10.1029/ Abay A (2018) Nitrogen release dynamics of Erythrina abys- JD092iD03p02999 sinica and Erythrina brucei litters as influenced by their Giardina CP, Ryan MG (2000) Evidence that decomposition biochemical composition. Afr J Plant Sci 2(12):331–340. rates of organic carbon in mineral soil do not vary with https://doi.org/10.5897/AJPS2018.1717 temperature. Nature 404:858–861 Adair EC, Hobbie SE, Hobbie RK (2010) Single-pool expo- Gindaba J, Olsson M, Itanna F (2004) Nutrient composition and nential decomposition models: potential pitfalls in their use short-term release from Croton macrostachyus Del. and in ecological studies. Ecology 91(4):1225–1236 Millettia ferruginea (Hochst.) Baker leaves. Biol Fertil Aerts R (1997) Climate, leaf litter chemistry and leaf litter Soils 40:393–397. https://doi.org/10.1007/s00374-004- decomposition in terrestrial ecosystems: a triangular rela- 0767-x tionship. Oikos 79:439–449 123 Agroforest Syst (2021) 95:755–766 765 Hairiah K, Sulistyani H, Suprayogo D, Widianto PP, Widodo 97(1–3):29–34. https://doi.org/10.1007/s10705-013-9590- RH, Van Noordwijk M (2006) Litter layer residence time in 9 forest and coffee agroforestry systems in Sumberjaya. Negash M, Starr M (2015) Biomass and soil carbon stocks of West Lampung For Ecol Manag 224(1–2):45–57. https:// indigenous agroforestry systems on the south-eastern Rift doi.org/10.1016/j.foreco.2005.12.007 Valley escarpment, Ethiopia. Plant Soil 393:95–107. Hossain M, Siddique MRH, Rahman S, Hossain Z, Hasan M https://doi.org/10.1007/s11104-015-2469-6 (2011) Nutrient dynamics associated with leaf litter Negash M, Starr M, Kanninen M (2013) Allometric equations decomposition of three agroforestry tree species (Azadir- for biomass estimation of Enset (Ensete ventricosum) achta indica, Dalbergia sissoo, and Melia azedarach)of grown in indigenous agroforestry systems in the Rift Val- Bangladesh. J For Res 22(4):577–582. https://doi.org/10. ley escarpment of southern-eastern Ethiopia. Agrofor Syst 1007/s11676-011-0175-7 87:571–581. https://doi.org/10.1007/s10457-012-9577-6 Isaac SR, Nair MA (2006) Litter dynamics of six multipurpose Negash M, Starr M, Kanninen M, Berhe L (2013) Allometric trees in a home garden in Southern Kerala, India. Agrofor equations for aboveground biomass of Coffea arabica L. Syst 67:203–213. https://doi.org/10.1007/s10457-005- grown in indigenous agroforestry systems in the Rift Val- 1107-3 ley escarpment of southern-eastern Ethiopia. Agrofor Syst Isaac ME, Gordon AM, Thevathasan N, Oppong SK, Quashie- 87:953–966. https://doi.org/10.1007/s10457-013-9611-3 Sam J (2005) Temporal changes in soil carbon and nitrogen Negash M, Yirdaw E, Luukkanen O (2012) Potential of in west African multistrata agroforestry systems: a indigenous multistrata agroforests for maintaining native chronosequence of pools and fluxes. Agrofor Syst floristic diversity in the south-eastern Rift Valley escarp- 65:23–31. https://doi.org/10.1007/s10457-004-4187-6 ment, Ethiopia. Agrofor Syst 85:9–28. https://doi.org/10. Jairo RM, Caicedo V, Jaimes Y (2017) Biomass decomposition 1007/s10457-011-9408-1 dynamic in agroforestry systems with Theobroma cacao L Oertel C, Matschullat J, Zurba K, Zimmermann F, Erasmi S in Rionegro Santander (Colombia). Agronomıa Colom- (2016) Greenhouse gas emissions from soils—a review. biana 35(2):182–189. https://doi.org/10.15446/agron. Chem Erde 76:327–352. https://doi.org/10.1016/j.chemer. colomb.v35n2.60981 2016.04.002 Kotur SC, Keshava Murthy SV (2010) Nutrient dynamics of Petit-Aldana J, Rahman MM, Parraguirre-Lezama C, Infante- annual growth-flush in mango (Mangifera indica L.). Cruz A, Romero-Arenas O (2019) Litter decomposition J Hortic Sci 5(1):75–77 process in coffee agroforestry systems. J Environ Sci Mahari A (2014) Leaf litter decomposition and nutrient release 35(2):121–139. https://doi.org/10.7747/JFES.2019.35.2. from Cordia africana Lam. and Croton Macrostachyus Del. 121 tree species. J Environ Earth Sci 4(15):1–7 Portillo-Estrada M, Pihlatie M, Korhonen JFJ, Levula J, Frumau Mbow C, Smith P, Skole D, Duguma L, Bustamante M (2014a) AKF, Ibrom A, Lembrechts JJ, Morillas L, Horva´th L, Achieving mitigation and adaptation to climate change Jones SK, Niinemets U (2016) Climatic controls on leaf through sustainable agroforestry practices in Africa. Curr litter decomposition across European forests and grass- Opin Environ Sustain 6:8–14. https://doi.org/10.1016/j. lands revealed by reciprocal litter transplantation experi- cosust.2013.09.002 ments. Biogeosciences 13:1621–1633. https://doi.org/10. Mbow C, Van Noordwijk M, Luedeling E, Neufeldt H, Minang 5194/bg-13-1621-2016 PA, Kowero G (2014) Agroforestry solutions to address Powers JS, Montgomery RA, Adair EC et al (2009) Decompo- food security and climate change challenges in Africa. Curr sition in tropical forests: a pan-tropical study of the effects Opin Environ Sustain 6:61–67. https://doi.org/10.1016/j. of litter type, litter placement and mesofaunal exclusion cosust.2013.10.014 across a precipitation gradient. J Ecol 97:801–811. https:// Meentemeyer V, Box EO, Thompson R (1982) World patterns doi.org/10.1111/j.1365-2745.2009.01515.x and amounts of terrestrial plant litter production. Bio- Raich JW, Schlesinger WH (1992) The global carbon dioxide sciences 32(2):125–128 flux in soil respiration and its relationship to vegetation and Murovhi NR, Materechera SA, Mulugeta SD (2012) Seasonal climate. Tellus 44B:81–99 changes in litter fall and its quality from three sub-tropical Raich JW, Tufekcioglu A (2000) Vegetation and soil respira- fruit tree species at Nelspruit South Africa. Agrofor Syst tion: correlations and controls. Biogeochemistry 86:61–71. https://doi.org/10.1007/s10457-012-9508-6 48(1):71–90 Negash M, Achalu N (2008) History of indigenous agro-forestry Salinas N, Malhi Y, Meir P, Silman M, Cuesta RR, Huaman J, in Gedeo, Ethiopia based on local community interviews: Salinas D, Huaman V, Gibaja A, Mamani M, Farfan F Vegetation diversity and structure in these land use sys- (2010) The sensitivity of tropical leaf litter decomposition tems. Ethiopian J Nat Resour 10(1):31–52 to temperature: results from a large-scale leaf translocation Negash M, Kanninen M (2015) Modeling biomass and soil experiment along an elevation gradient in Peruvian forests. carbon pools change of indigenous agroforestry system New Phytologist 189(4):967–977. https://doi.org/10.1111/ using CO2FIX approach. Agr Ecosyst Environ j.1469-8137.2010.03521.x 203:147–155. https://doi.org/10.1016/j.agee.2015.02. Schaffer B, Gaye GO (1989) Gas exchange, chlorophyll and 0040167-8809/a˜2 nitrogen content of mango leaves as influenced by light Negash M, Starr M (2013) Litterfall production and associated environment. HortScience 24(3):507–509 carbon and nitrogen fluxes of seven woody species grown Teklay T (2004) Seasonal dynamics in the concentrations of in indigenous agroforestry systems in the Rift Valley macronutrients and organic constituents in green and escarpment of Ethiopia. Nutr Cycl Agroecosyst 123 766 Agroforest Syst (2021) 95:755–766 senesced leaves of three agroforestry species in southern Urban L, Lechaudel M (2005) Effect of leaf-to-fruit ratio on leaf Ethiopia. Plant Soil 267:297–307 nitrogen content and net photosynthesis in girdled branches Teklay T (2007) Decomposition and nutrient release from of Mangifera indica L. Trees 19:564–571. https://doi.org/ pruning residues of two indigenous agroforestry species 10.1007/s00468-005-0415-6 during the wet and dry seasons. Nutrient Cycl Agroecosyst Waring BG (2012) A meta-analysis of climatic and chemical 77:115–126. https://doi.org/10.1007/s10705-006-9048-4 controls on leaf litter decay rates in tropical forests. Teklay T, Malmer A (2004) Decomposition of leaves from two Ecosystems 15:999–1009. https://doi.org/10.1007/s10021- indigenous trees of contrasting qualities under shaded 012-9561-z coffee and agricultural land-uses during the dry season at Wood TE, Detto M, Silver WL (2013) Sensitivity of soil res- Wondo Genet, Ethiopia. Soil Biol Biochem 36:777–786. piration to variability in soil moisture and temperature in a https://doi.org/10.1016/j.soilbio.2003.12.013 humid tropical forest. PLoS ONE 8(12):e80965. https:// Thangata PH, Hildebrand PE (2012) Carbon stock and seques- doi.org/10.1371/journal.pone.0080965 tration potential of agroforestry systems in smallholder Zhang D, Hui D, Luo Y, Zhou G (2008) Rates of litter decom- agroecosystems of sub-Saharan Africa: Mechanisms for position in terrestrial ecosystems: global patterns and ‘reducing emissions from deforestation and forest degra- controlling factors. J Plant Ecol 1(1):85–93. https://doi.org/ dation’ (REDD?). Agric Ecosyst Environ 158:172–183. 10.1093/jpe/rtn002 https://doi.org/10.1016/j.agee.2012.06.007 Thorp TG, Hutching D, Lowe T, Marsh KB (1997) Survey of Publisher’s Note Springer Nature remains neutral with fruit mineral concentrations and postharvest quality of regard to jurisdictional claims in published maps and New Zealand-grown ‘‘Hass’’ avocado (Persea americana institutional affiliations. Mill.). New Zealand J Crop Hortic Sci 25:251–260

Journal

Agroforestry Systems – Springer Journals

Published: Apr 13, 2021

There are no references for this article.