Individual farm wastes (bean trash, maize trash, banana trash, cattle slurry, goat slurry, and pig slurry) commonly produced by smallholder farmers in two farming systems of Masaka and Lyantonde Districts were composted in containers, each with a capacity of 5000 cc at room temperature. The contents were turned periodically to maintain adequate aeration. This compost was prepared under a shade to minimize nutrient losses through evaporation.

In another experiment, containers each with a capacity of 5000 cc were filled up to 3000 cm³, leaving a 40% headspace with similar wastes, and were located at one open end of the lab and exposed to sunlight all day.

In a control experiment, pure soil was put in the containers instead of farm wastes and either composted or surface deposited under the same laboratory conditions as the test samples. Procedures similar to those followed when collecting GHG emissions from compost and surface-deposited wastes described earlier will be followed in both farming systems.

Using a gas analyzer F-950 (three gas analyzer), we determined Carbon dioxide and oxygen from airtight containers in which individual farm wastes/aggregated farm wastes were contained. Based on the gas laws, we used a formula V 2 = P 1 V 1 P 2 to compute and obtain corresponding methane values. We calibrated the gas analyzer first. We inserted the gas sampling probe into the rubber cup of each container. The gas meter displayed the corresponding quantity for each recordable gas on the screen. We recorded each value in an Excel sheet. We carried out the experiment for 30 days and continued recording every other two days.

Three different types of compost were formulated.

1) C1: Only crop residues that are reported to be commonly used ingredients either in composting or surface deposition by arable smallholder farmers.

2) C2: Only animal wastes that are reported to be commonly used ingredients, either in composting or surface deposition by mixed smallholder farmers.

3) C3: Both crop and animal wastes are reported to be commonly used ingredients, either in composting or surface deposition by mixed smallholder farmers.

The above-named mixtures of wastes were composted in containers, each with a capacity of 5000 cc, filled up to 3000 cm³, leaving a 40% headspace. These were turned periodically to maintain adequate aeration. This compost was prepared under a shade to minimize nutrient losses through evaporation.

In another experiment, containers each with a capacity of 5000 cc were filled up to 3000 cm³, leaving a 40% headspace, with similar wastes and located at one open end of the lab and exposed to sunlight all day. Three waste combinations were labeled as follows:

1) S1: Only crop residues that are reported to be commonly used ingredients, either in composting or surface deposition by arable smallholder farmers.

2) S2: Only animal wastes that are reported to be commonly used ingredients, either in composting or surface deposition, by mixed smallholder farmers.

3) S3: Both crop and animal wastes are reported to be commonly used ingredients either in composting or surface deposition by mixed smallholder farmers. Adopted from[5].

In a control experiment, pure soil was put in the containers instead of farm wastes and either composted or surface deposited under the same laboratory conditions as the test samples. Procedures similar to those followed when collecting GHG emissions from compost and surface-deposited wastes described earlier will be followed in both farming systems.

Using a gas analyzer F-950 (three gas analyzer), we determined carbon dioxide and oxygen from airtight containers in which individual farm wastes/aggregated farm wastes were contained. Based on the gas laws, we used a formula V 2 = P 1 V 1 P 2 to compute and obtained corresponding methane values [6]. We calibrated the gas analyzer first. We inserted the gas sampling probe into the rubber cup of each container. The gas meter displayed the corresponding quantity for each recordable gas on the screen. We recorded each value in an Excel sheet. We carried out the experiment for 30 days and kept recording every other two days.

Masaka District is bordered by Bukomansimbi District to the north-west, Kalungu District to the north, Kalangala District to the east and south, Rakai District to the south-west, and Lwengo District to the west. The coordinates of the district are 00 30S, 31 45E.

Lyantonde is located in the Southwestern region of Uganda, lying between latitude 0˚S and 0.5˚S, and longitude 31.1˚E and 31.3˚E. Lyantonde borders Rakai District in the south, Masaka in the east, Kiruhura in the west, and Sembabule in the northeast.

Masaka District is located in the arable farming system, while Lyantonde District is located in the mixed farming system. Each district therefore represents a corresponding farming system. All samples were collected from these two farming systems.

We used quantitative methods to determine the relative GHG emissions of each system waste component and the aggregated farm wastes from biodegradable wastes of different smallholder farming systems [7]-[9].

Each waste category was weighed and recorded prior to the start of composting and GHG analysis [10]. Both the individual waste category and the aggregated waste category were grouped into two testing groups. One testing group was composted and the second group was surface decomposed.

Contents in the testing group were composted/surface decomposed in containers, each with a capacity of 5000 cc. Farm wastes were filled up to 3000 cm³, leaving a 40% headspace. The contents were turned periodically to maintain adequate aeration.

To collect GHG emissions, a closed chamber technique was used as proposed and used by [11][12], as described by [13] with some modifications. This method was used because it was easy, cheap, and did not require sophisticated instruments. Its versatility, design, and application procedures can be adapted to a wide range of situations.

Total population sampling was used. This is a type of purposive sampling where the whole population of farm wastes was studied. It was the most appropriate procedure for the population of 68 samples, which were manageable in size (48 individual waste samples, 16 aggregated waste samples, and 4 controls were sampled from both farming systems).

All waste materials involved in this study were sampled. The sampling interval was 48 hours or the chamber closure duration. This increased precision and saved on the cost of setting up experiments that would not be sampled in case sampling was randomized.

To collect GHG emissions, a closed chamber technique was used as proposed and used by [10][11][14] as described by [13] with some modifications. This method was appropriate because it was easy, cheap, and did not require sophisticated instruments. Its versatility, design, and application procedures can be adapted to a wide range of situations.

Using a gas analyzer F-950 (three gas analyzer), we determined Carbon dioxide and oxygen from airtight containers in which individual farm wastes/aggregated farm wastes were contained. Based on the gas laws, we used a formula V 2 = P 1 V 1 P 2 to compute and obtain corresponding methane values [6]. We calibrated the gas analyzer first. We inserted the gas sampling probe into the rubber cup of each container. The gas meter displayed the corresponding quantity for each recordable gas on the screen. We recorded each value in an Excel sheet. We carried out the experiment for 30 days, and we continued recording every other two days.

2.6.1. General

In this study, preliminary data were analyzed in Excel using descriptive statistics. Thus, frequency tables and charts were developed to provide simple summaries about the sample and related measurements. The quantitative data were then analyzed in SPSS, starting with composting and crop wastes, followed by surface decomposition. Total volumes of CO₂/CH₄ from the individual/aggregated wastes in each district were pooled together to obtain the overall emission of the village, sub-county, and later district. Volumes were converted to mg. Correlation analysis was run in SPSS to determine any relationships among farm wastes/aggregated farm wastes and farm waste management methods.

2.6.2. Analysis of Carbon Dioxide from Individual Farm Wastes

From the Pearson correlations run, we could see that the correlation coefficients for Beans, Bananas, Maize, Chicken manure, and Cow dung ranged between −0.714 and −0.460, which was significant (P < 0.01 for a two-tailed test), based on 56 observations. Goat slurry and pig slurry had correlation coefficients of 0.467 and 0.460, respectively, which were non-significant (P > 0.01 for a two-tailed test), based on 56 observations. For Goat slurry and pig slurry, when a multiple regression analysis was run, P = 0.012, P > 0.01, which was non-significant (P > 0.01 for a two-tailed test).

2.6.3. Analysis of Carbon Dioxide from Aggregated Farm Wastes

From the correlations run, we could see that the correlation coefficients for Composting and Surface decomposition were 0.944 and 0.81, respectively, which were non-significant (P > 0.01 for a two-tailed test), based on 48 observations. For Composting, when a multiple regression analysis was run, P = 0.012, P > 0.01, which was non-significant (P > 0.01 for a two-tailed test). For Surface decomposition, when a multiple regression analysis was run, P = 0.005, P < 0.01, which was significant (P < 0.01 for a two-tailed test).

Surface decomposition and composting had correlation coefficient values of 0.981 and 0.944, respectively, at the 0.01 significance level (2-tailed) (P > 0.01). At the 0.001 significance level (2-tailed), this indicates a statistically non-significant linear relationship affecting the volumes of CO₂.

2.6.4. Analysis of Methane from Individual Farm Wastes

From the bivariate correlations run, we could see that the correlations for method and farm wastes (Beans, Bananas, Maize, Chicken manure, and Cow dung) ranged between −0.805 and −0.016, which was significant (P < 0.01 for a two-tailed test), based on 56 observations. When Goat slurry and pig slurry were correlated with method, their correlations were 0.884 and 0.878, respectively, which were non-significant (P > 0.01 for a two-tailed test), based on 56 observations. For Goat slurry and pig slurry, when a multiple regression analysis was run, their correlation coefficient was 0.884, P = 0.468, P > 0.01, which was non-significant (P > 0.01 for a two-tailed test).

2.6.5. Analysis of Methane from Aggregated Farm Wastes

We ran the correlation analysis to see how composting and surface decomposition were correlated to emissions in various villages, and we found that the correlation for composting was −0.790, P = 0.20, which was non-significant (P > 0.05 for a two-tailed test). Surface decomposition had a correlation of 0.978, P = 0.000, which was significant (P < 0.01 for a two-tailed test) based on 56 observations. From the Pearson correlations run, we could see that the correlation coefficient for composting and villages was 0.978, P = 0.000, which was significant (P < 0.01 for a two-tailed test), based on 56 observations.

Composting beans from the arable farming system captured the highest volume of CO₂ (784.8 ppm), while maize produced more CO₂ volumes under composting (602 ppm) in the mixed farming system (Table 1).

Aggregated mean of Carbon dioxide (mg) by Farm waste & Waste Management Method for aggregated wastes was higher in Lyantonde up to a tune of 1219.0 ppm compared to 813.3 ppm recorded in Masaka (Table 2).

In Lyantonde District, CO₂ produced under surface decomposition ranged from 982.9 to 1214.9 mg. CO₂ produced under composting ranged from 803.2 to 1219.0 mg. In Masaka District, CO₂ produced under surface decomposition ranged from 538.5 to 813.3 mg. CO₂ produced under composting ranged from 490.1 to 732.9 mg (Table 2).

Table 1.Aggregated mean of carbon dioxide (mg/3000 cm³) by farm waste & waste management method for individual wastes.

3.1.1. Carbon Dioxide from Aggregated Waste (See Table 2)

Table 2. Aggregated mean of carbon dioxide (mg/3000 cm³) by farm waste & waste management method for aggregated wastes.

3.1.2. Methane

Aggregated mean of methane (mg) by farm waste & waste management method for beans was the highest, with a volume of 978.7 ppm in Masaka. Pig slurry in Lyantonde produced the highest volume of Methane (556.0 ppm) (Table 3).

In Lyantonde District, CH₄ produced under surface decomposition ranged from 1196.7 to 1515.0 mg. CH₄ produced under composting ranged from 1071.2 to 1520.6 mg.

In Masaka District, CH₄ produced under surface decomposition ranged from 671.5 to 799.7 mg. CH₄ produced under composting ranged from 611.1 to 792.4 mg (Table 4).

Table 3.Aggregated mean of methane (mg/3000 cm³) by farm waste & waste management method for individual wastes.

3.1.3. Methane from Aggregated Waste (See Table 4)

Table 4.Aggregated mean of methane (mg/3000 cm³) by farm waste & waste management method for aggregated wastes.