Temporal and Spatial Distributions of Waste Facilities and Solid Waste Management Strategies in Rural and Urban Saskatchewan, Canada

: Saskatchewan has the highest number of landﬁlls per capita in Canada. Given the lower population density and the skewed spatial population distribution, comprehensive analysis of municipal solid waste management systems in Saskatchewan is inherently difﬁcult. Most of the published waste studies however focus on city-level waste management, and there is a lack of literature with respect to the rural areas. In this study, landﬁlls and transfer stations are examined temporally and spatially using Geographic Information System. Landﬁlls and transfer stations from 2017 and 2020 were plotted against census division land area, annual budget, and population density to study temporal changes. Saskatchewan witnessed a 54% reduction in the number of landﬁlls and a 55% increase in number of transfer stations between 2017 and 2020. The replacement of landﬁlls with transfer stations are more noticeable in divisions 8, 9, and 16. Regression analysis is conducted, and landﬁll closure operation show no obvious correlation to division land area, annual budget, or population density. Rural division 18, representing Northern Saskatchewan, has approximately 45% of the land area in the province and has the lowest population density. The ﬁndings suggest different waste management strategies are required for urban and rural areas. The results of this study will help policy makers to better implement solid waste management strategies in urban and rural areas.


Introduction
Effective municipal solid waste management (MSWM) demands careful planning, prudent investment, and ongoing monitoring and quality assurance program [1]. In North America, landfills and waste transfer stations play a more dominant role in waste treatment than incineration and other waste-to-energy technologies, probably due to air emissions and residuals such as dioxins, furans, bottom, and fly ashes associated with the municipal waste incineration process [2].
Canada has one of the highest per capita waste generation rates across the globe [3]. As of 2019, the estimated annual waste generation per capita in Canada was 36.1 metric tons, followed by Bulgaria and United States with 26.7 and 25.9 metric tons, respectively [4]. However, literature on MSWM in Canada is limited, especially in Western and Central Canada [5,6]. Data sources are often limited to internal municipal documents, outdated government records, historic company annual reports, or otherwise unpublished internal reports not easily accessible to the general public. The shared managerial responsibilities and roles by all three levels of government (municipal, provincial, and federal) have further complicated the waste data availability issue. In Canada, the federal government is mainly responsible for exportation and importation of hazardous waste, whereas provincial government is mainly responsible for regulation of waste management facilities and inter-provincial waste management issues [7]. Municipalities look after the daily Sustainability 2021, 13, 6887 3 of 14 optimize landfills in a sub-area of the state of Sao Paulo but the potential impacts of transfer stations were ignored [31].
This study addresses the above concerns and analyzes both landfills and transfer stations temporally and spatially in Saskatchewan. The total area of Saskatchewan is about 651,000 km 2 [32], and there are clear divisions of urban and rural areas. In this study, rural area is defined as the area left over after the population centers have been identified using the latest Statistics Canada population data [33]. Rural areas include small towns, villages, developed and undeveloped lands, agricultural lands, and remote and wilderness areas [33]. The objectives of the study are to (i) examine temporal and spatial changes of landfills and transfer stations in Saskatchewan between 2017 and 2020 using GIS tools, (ii) analyze key indicators on MSWM planning and landfill closure using regression analysis, and (iii) investigate the effectiveness of MSWM strategies using a Landfill-Transfer Station ratio (LTR).
Unlike other waste studies focusing on specific urban centers, this study explicitly examines MSWM systems in both rural and urban regions. An original indicator, Landfill Transfer Station Ratio (LTR), is proposed and implemented in this study to examine evolution of MSWM strategies and managerial trend. Canadians rely on permanent land disposal as their primary waste treatment method despite of various limitations of the landfill technology [34][35][36]. The use of the proposed indicator is novel, and the ratio can be implemented in other jurisdictions with high proportion of rural areas. The results of this study will help policy makers to better monitor and implement MSWM strategies in their regions of governance.

Study Area
The study area, the province of Saskatchewan, is shown in red in the map embedded in Table 1. The map was constructed using ArcGIS 10.7.1 [37]. Saskatchewan is one of the Prairie provinces and shares its borders with Alberta to the west, the Northwest Territories to the north, Manitoba to the east and the United States to the south [38]. Table 1 displays key geographical information of the province [32, 39]. neglects the rural areas where waste management lacked the most due to shortage of governance and resources. GIS and Multi-Criteria Decision Analysis (MCDA) were used to site and optimize landfills in a sub-area of the state of Sao Paulo but the potential impacts of transfer stations were ignored [31]. This study addresses the above concerns and analyzes both landfills and transfer stations temporally and spatially in Saskatchewan. The total area of Saskatchewan is about 651,000 km 2 [32], and there are clear divisions of urban and rural areas. In this study, rural area is defined as the area left over after the population centers have been identified using the latest Statistics Canada population data [33]. Rural areas include small towns, villages, developed and undeveloped lands, agricultural lands, and remote and wilderness areas [33]. The objectives of the study are to (i) examine temporal and spatial changes of landfills and transfer stations in Saskatchewan between 2017 and 2020 using GIS tools, (ii) analyze key indicators on MSWM planning and landfill closure using regression analysis, and (iii) investigate the effectiveness of MSWM strategies using a Landfill-Transfer Station ratio (LTR).
Unlike other waste studies focusing on specific urban centers, this study explicitly examines MSWM systems in both rural and urban regions. An original indicator, Landfill Transfer Station Ratio (LTR), is proposed and implemented in this study to examine evolution of MSWM strategies and managerial trend. Canadians rely on permanent land disposal as their primary waste treatment method despite of various limitations of the landfill technology [34][35][36]. The use of the proposed indicator is novel, and the ratio can be implemented in other jurisdictions with high proportion of rural areas. The results of this study will help policy makers to better monitor and implement MSWM strategies in their regions of governance.

Study Area
The study area, the province of Saskatchewan, is shown in red in the map embedded in Table 1. The map was constructed using ArcGIS 10.7.1 [37]. Saskatchewan is one of the Prairie provinces and shares its borders with Alberta to the west, the Northwest Territories to the north, Manitoba to the east and the United States to the south [38]. Table 1 displays key geographical information of the province [32,39]. Saskatchewan is divided into 18 census divisions, with division identification numbers ascending south to north. Out of the total land area of the province, about 45% of the land area (approximately 270,000 km 2 ) is represented by a single division (division 18), covering all of Northern Saskatchewan. The most populated cities, Saskatoon and Regina, are located in Division 15 and Division 6, respectively. In total, about 61 population centers are identified in Saskatchewan. In this study, any area outside the boundary of these population centers is considered as a rural area. The same classification is used to differentiate rural and urban waste facilities.

Data Acquisition
Majority of the data and shapefiles used in this study were obtained from the provincial government's open data portals. Geospatial data of landfills and transfer stations were obtained from Saskatchewan geohub [40]. Annual budget in Canadian dollars (CAD) assigned to rural municipalities was obtained from government municipal revenue sharing portal [41]. Rural municipality is a type of incorporated municipality created by ministerial order, similar to counties in the western United States [42]. A census division contains multiple rural municipalities and thus the assigned budgets of encompassed rural municipalities are combined to obtain the total budget for a given census division. Population density of each census division was calculated using population data [43] and land area of each division [40].

Landfill Transfer Station Ratio (LTR)
A landfill transfer station ratio for each census division was computed to examine the evolution of MSWM during the three-year study period. As shown in Equation (1), the ratio is defined at each individual division and includes only operational waste facilities. Closed or inactive waste facilities are ignored. If there is no operational transfer station, the LTR is not defined.
Since waste transfer stations and disposal sites play two different roles in MSWM in Saskatchewan, the LTR combines these waste facilities into a practical indicator for comparison purposes. The proposed LTR is only applicable when assessing waste facilities spatial design within a pre-determined boundary. Federal census division boundaries were chosen in this study, but the method is equally applicable to different boundaries. The use of LTR is believed to be more appropriate than other "per capita" parameters since it better captures the spatial aspect of MSWM design.

Regression Analysis
Simple linear regression analysis was conducted to investigate the potential relationships of the variables [44]. This linear model is shown below (Equation (2)): where, 'Y' is dependent variable, 'X' is independent variable, 'a' is the intercept, 'b' is the slope and '€' is the residual error. The coefficient of determination (R 2 ) measures the fit of the linear regression line. The regression analysis is used to show the potential relationships between the variables, and no linear relationship between the variable is asserted. As discussed in Section 3, most of the relationships appear non-linear.

Geographical Spread and Numbers of Landfills and Transfer Stations
The spatial distributions of landfills and transfer stations in Saskatchewan in 2017 and 2020 are shown in Figure 1. A comparison between Figure 1a,b indicates a significant temporal change in the geographical spread of waste facilities across the province. The green circles represent operational landfills, and the black triangles represent operational transfer stations. Generally, there are many more waste facilities in Southern Saskatchewan, where major urban centers are located. Despite representing about 45% of the total area of the province, division 18 at the North contains very few waste facilities, predominantly due to its rural nature. It is important to note that there was no transfer station in Division 18 in 2017, as shown in Figure 1a. Compared to Figure 1a, Figure 1b has fewer green circles and more black triangles. Overall, Saskatchewan experienced a 54% reduction in the number of landfills between 2017 and 2020. On the contrary, the province saw a noticeable increase of around 55% in number of transfer stations. This finding suggests there was a major change of MSWM strategies in the province between 2017 and 2020.
The overall percentage changes in number of landfills and transfer stations are not consistent and vary considerably among the census divisions. For example, about 80%, 75%, and 68% of the landfills were shut down in divisions 9, 4, and 8, respectively. In comparison, divisions 7 and 10 witnessed no change (0%) in the number of waste facilities between 2017 and 2020. Interior divisions not sharing their boundaries with any other province/country appear to have experienced less of a change, in comparison to divisions sharing their interprovincial boundaries with either Manitoba or Alberta. For example, division 9, located adjacent to the Saskatchewan-Manitoba border and with the highest number of landfills in 2017, witnessed 80% of its landfills being shut down by 2020. A corresponding increase in number of transfer stations is also observed for the same division. The result suggests a managerial trend of replacing operational landfills with transfer stations. This observation is consistent with the Saskatchewan waste management strategy of development of meta-scale landfills and regionalization of waste management systems [45,46].

Census Division Land Area
The division land areas were plotted with the number of operational landfills and transfer stations, as shown in Figure 2. Division land areas were plotted ascending from Overall, Saskatchewan experienced a 54% reduction in the number of landfills between 2017 and 2020. On the contrary, the province saw a noticeable increase of around 55% in number of transfer stations. This finding suggests there was a major change of MSWM strategies in the province between 2017 and 2020.
The overall percentage changes in number of landfills and transfer stations are not consistent and vary considerably among the census divisions. For example, about 80%, 75%, and 68% of the landfills were shut down in divisions 9, 4, and 8, respectively. In comparison, divisions 7 and 10 witnessed no change (0%) in the number of waste facilities between 2017 and 2020. Interior divisions not sharing their boundaries with any other province/country appear to have experienced less of a change, in comparison to divisions sharing their interprovincial boundaries with either Manitoba or Alberta. For example, division 9, located adjacent to the Saskatchewan-Manitoba border and with the highest number of landfills in 2017, witnessed 80% of its landfills being shut down by 2020. A corresponding increase in number of transfer stations is also observed for the same division. The result suggests a managerial trend of replacing operational landfills with transfer stations. This observation is consistent with the Saskatchewan waste management strategy of development of meta-scale landfills and regionalization of waste management systems [45,46].

Census Division Land Area
The division land areas were plotted with the number of operational landfills and transfer stations, as shown in Figure 2. Division land areas were plotted ascending from Division 18 clearly stands out when comparing to the land areas of the remaining divisions. For comparison purposes, division 18 is larger than 42 out of 50 U.S. states, and only 33% of the countries in the world are larger than division 18 [47,48]. As shown in Figure 2, division 18 is about 10 times larger than other divisions. Division 18 is scarcely populated with only few small population centers located around the south-east corner of the division, and mainly comprised of rural areas. As such, division 18 is removed in some regression analysis to avoid possible skewing of the data set. The finding supports the use of different analytical approach in Northern and Southern Saskatchewan, as further discussed in the following sections.
A comparison between Figure 2a 300,000 Div 10 Div 12 Div 5 Div 1 Div 9 Div 11 Div 2 Div 13 Div 6 Div 3 Div 7 Div 15 Div 4 Div 16 Div 8 Div 17 Div   Division 18 clearly stands out when comparing to the land areas of the remaining divisions. For comparison purposes, division 18 is larger than 42 out of 50 U.S. states, and only 33% of the countries in the world are larger than division 18 [47,48]. As shown in Figure 2, division 18 is about 10 times larger than other divisions. Division 18 is scarcely populated with only few small population centers located around the south-east corner of the division, and mainly comprised of rural areas. As such, division 18 is removed in some regression analysis to avoid possible skewing of the data set. The finding supports the use of different analytical approach in Northern and Southern Saskatchewan, as further discussed in the following sections.
A comparison between Figure 2a (Figure 2b). As discussed in Section 3.1, there was an observable trend of replacing landfills with transfer station in Saskatchewan during the three-year study period.
Simple linear regression was conducted to better illustrate the potential relationships between the parameters. Figure 3 shows the relationship between operational landfills and transfer stations against census division land area. No obvious relationship is observed. In all cases, the linear regression equations have negligible slopes (<0.0003) and R 2 values (<0.06). For a well designed MSWM system, one would generally expect more waste facilities in a larger division, though other factors such as population density should likely be considered.  (Figure 2b). As discussed in Section 3.1, there was an observable trend of replacing landfills with transfer station in Saskatchewan during the three-year study period. Simple linear regression was conducted to better illustrate the potential relationships between the parameters. Figure 3 shows the relationship between operational landfills and transfer stations against census division land area. No obvious relationship is observed. In all cases, the linear regression equations have negligible slopes (<0.0003) and R 2 values (<0.06). For a well designed MSWM system, one would generally expect more waste facilities in a larger division, though other factors such as population density should likely be considered.   and 2020 in Figure 3a,b suggests that the strategical phasing out of the operational landfills in 2020 was not designed in accordance with the division land area. Figure 4 shows the regression analyses between the number of waste facilities and annual budget in 2017 and 2020. Division 18, the largest division, was assigned a budget of CAD 18.2 million in 2017 and increased slightly to CAD 18.4 million in 2020. These budget allocations were about three to five times higher than other divisions. The managerial strategy and financial resources in Northern Saskatchewan were found quite different than the remaining divisions. and 2020 in Figure 3a,b suggests that the strategical phasing out of the operational landfills in 2020 was not designed in accordance with the division land area. Figure 4 shows the regression analyses between the number of waste facilities and annual budget in 2017 and 2020. Division 18, the largest division, was assigned a budget of CAD 18.2 million in 2017 and increased slightly to CAD 18.4 million in 2020. These budget allocations were about three to five times higher than other divisions. The managerial strategy and financial resources in Northern Saskatchewan were found quite different than the remaining divisions.   Figure 4), a significant positive slope with higher R 2 value was observed for landfills and a negative slope and lower R 2 values was observed for transfer stations. The results suggest division 18 can significantly skew the dataset, and its MSWM system is different from other divisions in the province.

Rural Municipalities Annual Budget
More scattering of data points is observed in 2017, with a R 2 of 0.0471 (Figure 4a). It appears that the relationship between the number of landfills and annual budgets was more coherent in 2020, with a slight increase in R 2 (0.0724). The low R 2 values of the regression lines (<0.1) suggest that the relationships are not linear. The slope of the regression line in 2020 was about two times higher than in 2017, suggesting a fairer landfill operation funding mechanism.
As shown in Figure 4b, the funding mechanism with respect to transfer stations was again more coherent in 2020, with a slight increase in R 2 from 0.1412 to 0.2034. Figure 4b suggests a higher annual budget corresponds well to the number of operational transfer stations in both 2017 and 2020. This finding suggests that a higher portion of budgets were spent on the establishments of new transfer stations in 2020.  A positive relationship is observed for landfills (Figure 4a), and for transfer stations (Figure 4b). When considering division 18 (not shown in the Figure 4), a significant positive slope with higher R 2 value was observed for landfills and a negative slope and lower R 2 values was observed for transfer stations. The results suggest division 18 can significantly skew the dataset, and its MSWM system is different from other divisions in the province.

Census Division Population Density
More scattering of data points is observed in 2017, with a R 2 of 0.0471 (Figure 4a). It appears that the relationship between the number of landfills and annual budgets was more coherent in 2020, with a slight increase in R 2 (0.0724). The low R 2 values of the regression lines (<0.1) suggest that the relationships are not linear. The slope of the regression line in 2020 was about two times higher than in 2017, suggesting a fairer landfill operation funding mechanism.
As shown in Figure 4b, the funding mechanism with respect to transfer stations was again more coherent in 2020, with a slight increase in R 2 from 0.1412 to 0.2034. Figure 4b suggests a higher annual budget corresponds well to the number of operational transfer stations in both 2017 and 2020. This finding suggests that a higher portion of budgets were spent on the establishments of new transfer stations in 2020. The landfill curve is more consistent and stable among the divisions in 2020 (blue solid curve, Figure 5b), indicating a more systematic approach in the use landfill technology in the province. Divisions 3, 8, and 9 experienced significant drops in operational landfills in 2020. Specifically, in division 9 (Figure 5b), 29 out of 36 operational landfills (80.6%) were shut down during a three-year period. Geographical location of the divisions appears important, as divisions sharing at least one of their boundaries with other provinces/country generally have a larger reduction in operating landfills. Many Saskatche- Div 18 Div 4 Div 3 Div 14 Div 13 Div 10 Div 2 Div 8 Div 12 Div 16 Div 1 Div 5 Div 17 Div 9 Div 7 Div 15 Div 6 Div 11  The landfill curve is more consistent and stable among the divisions in 2020 (blue solid curve, Figure 5b), indicating a more systematic approach in the use landfill technology in the province. Divisions 3, 8, and 9 experienced significant drops in operational landfills in 2020. Specifically, in division 9 (Figure 5b), 29 out of 36 operational landfills (80.6%) were shut down during a three-year period. Geographical location of the divisions appears important, as divisions sharing at least one of their boundaries with other provinces/country generally have a larger reduction in operating landfills. Many Saskatchewan landfills accept outof-the-province waste, and the closure of landfills in these divisions may have negatively impacted the waste management services at neighbouring towns and cities. Divisions with the highest population densities (15, 6, and 11) also had highest number of transfer stations in both 2017 (Figure 5a) and in 2020 (Figure 5b). This suggests high population density divisions favor the use of transfer stations in MSWM. Unlike the landfill curves, the transfer stations curves (orange broken lines, Figure 5a,b) generally shifted upward in 2020, suggesting the promotion of transfer stations. Figure 6 shows the regression analyses between the number of waste facilities and population density. Divisions 6 and 11 were excluded in the regression analysis due to their high population density values ( Figure 5). The R 2 values for 2017 regression curves are very low, and no apparent relationship is found between the number of landfills and the population density (Figure 6a). This finding is consistent with the trend reported in Figure 5. On the contrary, positive trends are clearly observed between the number of transfer stations and population density (Figure 6b). The slopes of the regression curves are positive (+3.1~+3.3), with moderate R 2 values (0.55~0.59).

Census Division Population Density
sions may have negatively impacted the waste management services at neighbouring towns and cities. Divisions with the highest population densities (15, 6, and 11) also had highest number of transfer stations in both 2017 ( Figure 5a) and in 2020 (Figure 5b). This suggests high population density divisions favor the use of transfer stations in MSWM. Unlike the landfill curves, the transfer stations curves (orange broken lines, Figure 5a,b) generally shifted upward in 2020, suggesting the promotion of transfer stations. Figure 6 shows the regression analyses between the number of waste facilities and population density. Divisions 6 and 11 were excluded in the regression analysis due to their high population density values ( Figure 5). The R 2 values for 2017 regression curves are very low, and no apparent relationship is found between the number of landfills and the population density (Figure 6a). This finding is consistent with the trend reported in Figure 5. On the contrary, positive trends are clearly observed between the number of transfer stations and population density (Figure 6b). The slopes of the regression curves are positive (+3.1~+3.3), with moderate R 2 values (0.55~0.59).
An effective and efficient MSWM system should correspond well to the population distributions. The results suggest that population density was not sufficiently considered in the closure of the landfills. A more systematic approach on strategical siting of landfills is recommended.

Landfill to Transfer Station Ratio (LTR)
The LTRs are plotted with respect to census division land area (Figure 7a) and population distribution (Figure 7b). As shown in Figure 7a, the green broken curve is always below, or equal to, the orange solid line. On average, there was a 72% decrease in LTR An effective and efficient MSWM system should correspond well to the population distributions. The results suggest that population density was not sufficiently considered in the closure of the landfills. A more systematic approach on strategical siting of landfills is recommended.

Landfill to Transfer Station Ratio (LTR)
The LTRs are plotted with respect to census division land area (Figure 7a) and population distribution (Figure 7b). As shown in Figure 7a, the green broken curve is always below, or equal to, the orange solid line. On average, there was a 72% decrease in LTR during the study period across the province. Division 18 is an exception, since it had no operational transfer stations in 2017, and thus LTR was undefined (Equation (1)

Landfill to Transfer Station Ratio (LTR)
The LTRs are plotted with respect to census division land area (Figure 7a) and population distribution (Figure 7b). As shown in Figure 7a, the green broken curve is always below, or equal to, the orange solid line. On average, there was a 72% decrease in LTR during the study period across the province. Division 18 is an exception, since it had no operational transfer stations in 2017, and thus LTR was undefined (Equation (1)) and arbitrarily set to zero in Figure 7a. The LTRs in 2017 are quite erratic, ranging from 0.43 (Division 11) to 24.00 (Division 3). The LTRs in 2020 are more consistent among divisions (0.17-6.00), suggesting a province-wide movement of replacing landfills with transfer stations. The reduction in LTRs is however poorly related to division land area. 300,000 Div 10 Div 12 Div 5 Div 1 Div 9 Div 11 Div 2 Div 13 Div 6 Div 3 Div 7 Div 15 Div 4 Div 16 Div 8 Div 17 Div   In 2017, the average LTR was about 4.2 (Figure 7b), but divisions 3, 8, and 9 had much higher LTRs (24.0, 9.5, and 12.0, respectively). Division population density alone fails to explain these LTR peaks. Between 2017 and 2020, the LTR of most divisions were reduced by half, whereas the LTRs in division 3, 8, and 9 were reduced by around 10 times. The unusual reduction of LTR may be due to the added managerial control of the waste originated outside of the province. The LTRs were much more consistent across the province in 2020 (broken green line, Figure 7b).
A closer look at Figure 7b reveals that population densities among the divisions were Div 18 Div 4 Div 3 Div 14 Div 13 Div 10 Div 2 Div 8 Div 12 Div 16 Div 1 Div 5 Div 17 Div 9 Div 7 Div 15 Div 6 Div 11  In 2017, the average LTR was about 4.2 (Figure 7b), but divisions 3, 8, and 9 had much higher LTRs (24.0, 9.5, and 12.0, respectively). Division population density alone fails to explain these LTR peaks. Between 2017 and 2020, the LTR of most divisions were reduced by half, whereas the LTRs in division 3, 8, and 9 were reduced by around 10 times. The unusual reduction of LTR may be due to the added managerial control of the waste originated outside of the province. The LTRs were much more consistent across the province in 2020 (broken green line, Figure 7b).
A closer look at Figure 7b reveals that population densities among the divisions were mostly constant during the three-year study period, with the exception of divisions 6 and 11. Urbanization in these two divisions were observed. LTRs in these two divisions were relatively constant during the study period. The finding suggests that the closure of landfills was more targeted to low-populated rural areas.

Limitations and Recommendations
A single indicator such as LTR can only approximate a jurisdiction's multidimensional waste management system profile. The proposed LTR should be used in conjunction with other indices and indicators to fully capture the essences of a waste management system. The temporal changes in LTR are likely related to the changes in population and other socio-economic factors such as industrial type, family incomes, etc., and further studies are recommended. In this study, federal census division boundaries were selected. It should be noted that the findings derived from the proposed LTR are sensitive to the selection of the boundaries. Moreover, only active waste facilities were considered in this study. Closed or inactive waste facilities may impact the overall long-term design and operation of MSWM system, and should be consider in a future study.
The use of the proposed LTR is original, and the ratio can be implemented in other jurisdictions with high proportion of rural areas. Due to the rural nature of division 18, special managerial measures are recommended.

Conclusions
The paper examines the temporal and spatial changes of landfills and transfer stations in Saskatchewan. Unlike other waste studies focusing on specific urban centers, this study explicitly examines MSWM systems in both rural and urban regions.
Between 2017 and 2020, the number of landfills decreased by 54% and transfer stations increased by 55% in Saskatchewan. Interior divisions seemed to experience less changes, in comparison to divisions adjacent to interprovincial borders. Division 18, representing the rural Northern Saskatchewan, has a lower population density and larger annual budget than other divisions. Inclusion of division 18 in data analysis may skew the data set, and a different analytical approach is recommended. No obvious relationship was found between operational landfills or transfer stations against division land area, or population density. A similar conclusion was reached for number of operational landfills with respect to division annual budget. The results suggest that the current MSWM system in Saskatchewan is suboptimal and could be further strengthened. On the other hand, the number of transfer stations appears sensitive to division annual budget and population density.
In general, divisions with more operational landfills in 2017 witnessed larger drop in operational landfills in 2020. An average of 72% reduction in LTR was observed between 2017 and 2020. The results of this study provide insights into the design and operation of MSWM in Saskatchewan. The proposed analytical framework also helps policy makers to better monitor and implement MSWM strategies in their regions of governance.