Soil N2O emission from organic and conventional cotton farming in Northern Tanzania.

The effort to increase the sustainable supply of food and fibre is challenged by the potential for increased greenhouse gas (GHG) emissions from farming systems with intensified production systems. This study aimed at quantifying soil N2O emissions from smallholder organic and conventional cotton production practices in a semi-arid area, Meatu, Northern Tanzania. Field experiments were conducted to quantify N2O emissions under (i) current practices with organic (3 Mg ha-1 farmyard manure (FYM)) and conventional (30 kg mineral N ha-1) cultivation; (ii) a high input practice with organic (5 Mg ha-1 FYM) and conventional (60 kg mineral N ha-1) cultivation; and (iii) an integrated practice with organic (3 Mg FYM + legume intercropping) and conventional (30 kg N + 3 Mg ha-1 FYM) cultivation. In both organic and conventional farming, control treatments with no fertilizer application were included. The study was performed over two growing seasons, where season 1 was rather wet and season 2 was rather dry. Static chambers were used for in-situ measurement of N2O emission from soil. The current organic and conventional cotton farming practices did not differ (P > 0.05) in cumulative area-scaled and yield-scaled N2O emissions. High input conventional cotton showed higher area scaled N2O emissions than organic cotton during the wetter season, but not during the drier season. The inorganic fertilizer + FYM combination did not differ (P > 0.05) in area- and yield-scaled N2O emissions from conventional practice. Intercropping cotton and legumes did not affect (P > 0.05) N2O emission compared to 3 Mg FYM ha-1. The emission factors for both conventional and organic systems were generally above 1% in the dry season 2, but below 1% in the wetter season 1. The use of organic and inorganic fertilizers at rates up to 60 kg N ha-1, FYM-inorganic fertilizer combination, and cotton-legume intercropping increased yields, while N2O emissions stayed low, in particular with use of mineral fertilizers.

Effect of feeding practices and manure quality on CH4 and N2O emissions from uncovered cattle manure heaps in Kenya.

Countries in sub-Saharan Africa (SSA) rely on IPCC emission factors (EF) for GHG emission reporting. However, these were derived for industrialized livestock farms and do not represent conditions of smallholder farms (small, low-producing livestock breeds, poor feed quality, feed scarcity). Here, we present the first measurements of CH4 and N2O emissions from cattle-manure heaps representing feeding practices typical for smallholder farms in the highlands of East Africa: 1) cattle fed below maintenance energy requirements to represent feed scarcity, and 2) cattle fed tropical forage grasses (Napier, Rhodes, Brachiaria). Sub-maintenance feeding reduced cumulative manure N2O emissions compared to cattle receiving sufficient feed but did not change EFN2O. Sub-maintenance feeding did not affect cumulative manure CH4 emissions or EFCH4. When cattle were fed tropical forage grasses, cumulative manure N2O emissions did not differ between diets, but manure EFN2O from Brachiaria and Rhodes diets were lower than the IPCC EFN2O for solid storage (1%, 2019 Refinement of IPCC Guidelines). Manure CH4 emissions were lower in the Rhodes grass diet than when feeding Napier or Brachiaria, and manure EFCH4 from all three grasses were lower than the IPCC default (4.4 gCH4kg-1 VS, 2019 Refinement of IPCC Guidelines). Regression analysis revealed that manure N concentration and C:N were important drivers of N2O emissions, with low N concentrations and high C:N reducing N2O emissions. Our results show that IPCC EFs overestimate excreta GHG emissions, which calls for additional measurements to develop localized EFs for smallholder livestock systems in SSA.

Interactive effects of dung deposited onto urine patches on greenhouse gas fluxes from tropical pastures in Kenya.

Dung and urine patches on grasslands are hotspots of greenhouse gas (GHG) emissions in temperate regions, while its importance remains controversial for tropical regions as emissions seem to be lower. Here we investigated N2O, CH4 and CO2 emissions from urine and dung patches on tropical pastures in Kenya, thereby disentangling interactive and pure water, dung or urine effects. GHG fluxes were monitored with automated chambers for 42-59 days covering three seasons (short rainy season, long rainy season, dry season) for six treatments (Control; +1 L water; +1 kg dung; 1 L urine; 1 L water +1 kg dung; 1 L urine +1 kg dung). Cumulative CO2 emissions did not differ among treatments in any of the seasons. Water or urine addition alone did not affect CH4 fluxes, but these were elevated in all dung-related treatments. Scaled up on the total area covered, dung patches halve the CH4 sink strength of tropical pastures during the dry season, while during the rainy season they may turn tropical pastures into a small CH4 source. For N2O, both dung and urine alone and in combination stimulated emissions. While the N2O emission factor (EFN2O) from dung being constant across seasons, the EFN2O for urine was greater during the short rainy season than during the dry season. Combined application of urine + dung was additive on EFN2O. While the mean dung EFN2O in our study (0.06%) was similar to the IPCC Guidelines for National GHG Inventories EFN2O for dry climate (0.07%), the urine EFN2O we measured (0.03-0.25%) was lower than the IPCC value (0.32%). In addition, the IPCC Guidelines assume a urine-N: dung-N ratio of 0.66:0.34, which is higher than found for SSA (<0.50:0.50). Consequently, IPCC Guidelines still overestimate N2O emissions from excreta patches in SSA.

Pasture enclosures increase soil carbon dioxide flux rate in Semiarid Rangeland, Kenya.

BACKGROUND: Pasture enclosures play an important role in rehabilitating the degraded soils and vegetation, and may also influence the emission of key greenhouse gasses (GHGs) from the soil. However, no study in East Africa and in Kenya has conducted direct measurements of GHG fluxes following the restoration of degraded communal grazing lands through the establishment of pasture enclosures. A field experiment was conducted in northwestern Kenya to measure the emission of CO2, CH4 and N2O from soil under two pasture restoration systems; grazing dominated enclosure (GDE) and contractual grazing enclosure (CGE), and in the adjacent open grazing rangeland (OGR) as control. Herbaceous vegetation cover, biomass production, and surface (0-10 cm) soil organic carbon (SOC) were also assessed to determine their relationship with the GHG flux rate. RESULTS: Vegetation cover was higher enclosure systems and ranged from 20.7% in OGR to 40.2% in GDE while aboveground biomass increased from 72.0 kg DM ha-1 in OGR to 483.1 and 560.4 kg DM ha-1 in CGE and GDE respectively. The SOC concentration in GDE and CGE increased by an average of 27% relative to OGR and ranged between 4.4 g kg-1 and 6.6 g kg-1. The mean emission rates across the grazing systems were 18.6 µg N m-2 h-1, 50.1 µg C m-2 h-1 and 199.7 mg C m-2 h-1 for N2O, CH4, and CO2, respectively. Soil CO2 emission was considerably higher in GDE and CGE systems than in OGR (P < 0.001). However, non-significantly higher CH4 and N2O emissions were observed in GDE and CGE compared to OGR (P = 0.33 and 0.53 for CH4 and N2O, respectively). Soil moisture exhibited a significant positive relationship with CO2, CH4, and N2O, implying that it is the key factor influencing the flux rate of GHGs in the area. CONCLUSIONS: The results demonstrated that the establishment of enclosures in tropical rangelands is a valuable intervention for improving pasture production and restoration of surface soil properties. However, a long-term study is required to evaluate the patterns in annual CO2, N2O, CH4 fluxes from soils and determine the ecosystem carbon balance across the pastoral landscape.

Land use affects total dissolved nitrogen and nitrate concentrations in tropical montane streams in Kenya.

African tropical montane forests are facing fast and dynamic changes in land use. However, the impacts of these changes on stream water quality are understudied. This paper aims at assessing the effect of land use and physical catchment characteristics on stream water concentrations of dissolved organic carbon (DOC), total dissolved nitrogen (TDN), nitrate (NO3-N) and dissolved organic nitrogen (DON) in the Mau Forest, the largest tropical montane forest in Kenya. We conducted five synoptic stream water sampling campaigns at the outlets of 13-16 catchments dominated by either natural forest, smallholder agriculture or commercial tea and tree plantations. Our data show a strong effect of land use on TDN and NO3-N, with highest concentrations in stream water of catchments dominated by tea plantations (1.80±0.50 and 1.62±0.60mgNl-1, respectively), and lowest values in forested catchments (0.55±0.15 and 0.30±0.08mgNl-1, respectively). NO3-N concentration increased with stream temperature and specific discharge, but decreased with increasing catchment area. DOC concentrations increased with catchment area and precipitation and decreased with specific discharge, drainage density and topographic wetness index. Precipitation and specific discharge were also strong predictors for DON concentrations, with an additional small positive effect of tree cover. In summary, land use affects TDN and NO3-N concentrations in stream water in the Mau Forest region in Kenya, while DOC and DON were more related to hydrologic regimes and catchment properties. The importance of land use for NO3-N and TDN concentrations emphasizes the risk of increased nitrogen export along hydrological pathways caused by intensified land use and conversion of land to agricultural uses, which might result in deterioration of drinking water quality and eutrophication in surface water in tropical Africa.

Ammonia volatilization after surface application of laying-hen and broiler-chicken manures.

Ammonia (NH) losses after field application of animal manure are affected by manure characteristics. The objectives of this study were to quantify NH losses from poultry manures obtained from varied handling and storage systems commonly found in eastern Canada and to relate NH emissions to manure characteristics. We measured NH volatilization using wind tunnels for 22 d after soil-surface application of seven solid poultry manures originating from farms varying in production type (laying hens and broiler chickens) and in storage duration and conditions. Cumulative emissions (2.7-7.0 g NH-N m) accounted for 13.6 to 35.0% of the total N applied and 51 to 84% (mean, 70%) of the sum of ammoniacal N, urea N, and uric acid N applied (TAUA). On average, 20% of these losses occurred during the first 4.5 h after application for manures that were not dried in the barn shortly after excretion. Production type and storage durations could not explain differences in NH volatilization between manures. Volatilization losses were linearly related to manure dry matter and to manure-derived NH-N, but sources of N changed with time after application. During the first 7 d, variations in total ammoniacal N applied (TANA) among manures explained most of the variations in cumulative NH losses ( = 0.85 after 26 h and 0.92 after 7 d). After a simulated rainfall (5 mm) on Day 7 that stimulated the decomposition of uric acid in manures, TAUA rather than TANA was related to cumulative emissions ( = 0.77 after 14 and 22 d). Our results indicate that reliable estimates of NH volatilization after land spreading of poultry manures should be based not only on TANA but also on NH-N derived from the decomposition of uric acid, that volatilization losses reported in the literature (including the present study) averaged 50% of TAUA, and that estimates for a given situation also need to account for local environmental conditions.

Nitrous oxide emissions from clayey soils amended with paper sludges and biosolids of separated pig slurry.

Wastes from animal production and from the paper industry are often used as amendments to agricultural soils. Although data exist on the impacts of raw animal wastes on NO production, little is known regarding the effects of paper wastes and biosolids from treated animal waste. We measured NO emissions for two consecutive snow-free seasons (mid-May through the end of October) from poorly drained clayey soils under corn ( L.). Soils were amended with raw pig slurry (PS) or biosolids (four PS-derived and two paper sludges) and compared with soils with mineral N fertilizer (CaNHNO) or without N addition (Control). Area-based NO emissions from the mineral N fertilizer (average, 8.2 kg NO-N ha; 4.2% of applied N) were higher ( < 0.001) than emissions from the organic amendments, which ranged from 1.5 to 6.1 kg NO-N ha (-0.4 to 2.5% of applied N). The NO emissions were positively correlated with mean soil NO availability (calculated as "NO exposure"), which was highest with mineral N fertilizer. In plots treated with organic amendments (i.e., biosolids and raw PS), NO exposure was negatively correlated to the C:N ratio of the amendment. This resulted in lower NO emissions from the higher C:N ratio biosolids, especially compared with the low C:N ratio PS. Application of paper sludge or PS-derived biosolids to these fine-textured soils, therefore, reduced NO emissions compared with raw PS and/or mineral N fertilizers ( < 0.01).

Ammonia volatilization and nitrogen retention: how deep to incorporate urea?

Incorporation of urea decreases ammonia (NH) volatilization, but field measurements are needed to better quantify the impact of placement depth. In this study, we measured the volatilization losses after banding of urea at depths of 0, 2.5, 5, 7.5, and 10 cm in a slightly acidic (pH 6) silt loam soil using wind tunnels. Mineral nitrogen (N) concentration and pH were measured in the top 2 cm of soil to determine the extent of urea N migration and the influence of placement depth on the availability of ammoniacal N for volatilization near the soil surface. Ammonia volatilization losses were 50% of applied N when urea was banded at the surface, and incorporation of the band decreased emissions by an average of 7% cm (14% cm when expressed as a percentage of losses after surface banding). Incorporating urea at depths >7.5 cm therefore resulted in negligible NH emissions and maximum N retention. Cumulative losses increased exponentially with increasing maximum NH-N and pH values measured in the surface soil during the experiment. However, temporal variations in these soil properties were poorly related to the temporal variations in NH emission rates, likely as a result of interactions with other factors (e.g., water content and NH-N adsorption) on, and fixation by, soil particles. Laboratory and field volatilization data from the literature were summarized and used to determine a relationship between NH losses and depth of urea incorporation. When emissions were expressed as a percentage of losses for a surface application, the mean reduction after urea incorporation was approximately 12.5% cm. Although we agree that the efficiency of urea incorporation to reduce NH losses varies depending on several soil properties, management practices, and climatic conditions, we propose that this value represents an estimate of the mean impact of incorporation depth that could be used when site-specific information is unavailable.

Nitrous oxide emissions respond differently to mineral and organic nitrogen sources in contrasting soil types.

The use of various animal manures for nitrogen (N) fertilization is often viewed as a viable replacement for mineral N fertilizers. However, the impacts of amendment type on NO production may vary. In this study, NO emissions were measured for 2 yr on two soil types with contrasting texture and carbon (C) content under a cool, humid climate. Treatments consisted of a no-N control, calcium ammonium nitrate, poultry manure, liquid cattle manure, or liquid swine manure. The N sources were surface applied and immediately incorporated at 90 kg N ha before seeding of spring wheat ( L.). Cumulative NO-N emissions from the silty clay ranged from 2.2 to 8.3 kg ha yr and were slightly lower in the control than in the fertilized plots ( = 0.067). The 2-yr mean NO emission factors ranged from 2.0 to 4.4% of added N, with no difference among N sources. Emissions of NO from the sandy loam soil ranged from 0.3 to 2.2 kg NO-N ha yr, with higher emissions with organic than mineral N sources ( = 0.015) and the greatest emissions with poultry manure ( < 0.001). The NO emission factor from plots amended with poultry manure was 1.8%, more than double that of the other treatments (0.3-0.9%), likely because of its high C content. On the silty clay, the yield-based NO emissions (g NO-N kg grain yield N) were similar between treatments, whereas on the sandy loam, they were greatest when amended with poultry manure. Our findings suggest that, compared with mineral N sources, manure application only increases soil NO flux in soils with low C content.