Quantification of Greenhouse Gas Fluxes and Derivation of Maize Cropping Calendar in Croplands under Rainfall Variability in Kenya
Macharia, Joseph Maina
MetadataShow full item record
Increase in global atmospheric greenhouse gases (GHG) has led to an increase in the radiative forcing resulting in climate variability. The objective of this study was to identify the best management practice which ensures high maize productivity while emitting as little GHG fluxes as possible under the influence of rainfall variability. The study was carried out in three major parts of Kenya, that is, Kenya as a whole, the agricultural potential zones and central highlands of Kenya. The accuracy of the satellite precipitation products was determined based on data obtained between 1983 and 2013 from Climate Hazards Group InfraRed Precipitation with Station (CHIRPS), Global Precipitation Climatology Center (GPCC), National Aeronautics and Space Administration-Prediction Of Worldwide Energy Resource (NASA-POWER) and Tropical Rainfall Measuring Mission - Multi-satellite Precipitation Analysis 3B42 version 7 (TRMM). The satellite data were compared with observed data from Kenya meteorological department. The soil GHG fluxes were quantified from maize production experiment in Mbeere South sub-county for one year using static GHG chambers. The DeNitrification-DeComposition (DNDC) model was parameterised using experimental data and used to simulate GHG fluxes. The length of the cropping season, onset and cessation date in the agricultural zones of Kenya were determined using daily satellite data from NASA-POWER between 1983-2017 using RAIN software. Results show that all satellite products either overestimated or underestimated rainfall amounts on a pixel to pixel comparison. The TRMM product best estimated rainfall in the tropical cool humid (r2=0.64), tropical warm humid (r2=0.58) and tropical cool sub-humid (r2=0.39), GPCC product in the tropical warm semiarid (r2=0.46) and tropical warm sub-humid (r2=0.21), NASA- POWER product in tropical cool arid (r2=0.97) and tropical cool semiarid (r2=0.53) while CHIRPS product best estimated rainfall in the tropical warm arid (r2=0.33). Cumulative annual GHG fluxes ranged from -0.05 to -0.65 kg CH4-C ha-1 yr-1, 1.31 to 3.39 Mg CO2-C ha-1 yr-1 and 0.12 to 1.15 kg N2O-N ha-1 yr-1 for the four different treatments respectively. Animal manure produced the highest amounts of CO2 emissions (P<0.001) and N2O fluxes (P<0.001) and the lowest yield-scaled emission (0.5 g N2O–N kg-1 N). Animal manure + inorganic fertilisers produced the highest amounts of CH4 fluxes (P<0.001) and the highest YSE (2.2 g N2O–N kg-1 N). The DNDC simulated GHGs followed seasonality with peaks recorded immediately after the onset of rains which coincided with fertilisation. The DNDC simulated CO2 was slightly higher than observed while the N2O were slightly lower than observed though not significantly different at P=0.05. Results on cropping calendar demonstrate two key regions in Kenya, one with two seasons namely; the Long rains (LR) and Short rains (SR) and the other one with one season in a year. The LR onset is experienced in March and cessation in July; the SR onset in September and cessation in November while the one season onset in March and cessation in October. The LR and SR length of growing season ranges between 23-90 days while that of one season ranges between 192-259 days of sufficient rainfall. These results demonstrate the promising potential of the satellite data in complementing the unreliable data in Kenya. Animal manure has the ability to increase maize yields while simultaneously reducing yield-scaled GHG emissions. The DNDC model provides an accurate and cheaper alternatives for quantifying GHG for national GHG inventories and reporting to the UNFCCC. The results also derive a cropping calendar crucial for the planning of agricultural farming activities which will ultimately reduce losses and improve rainfed agricultural production.