Navigation – Plan du site

AccueilRubriquesCartographie, Imagerie, SIG2022Urban climatic maps for environme...

2022
1013

Urban climatic maps for environmental planning in Manizales, Colombia

Cartes climatiques urbaines pour la planification environnementale à Manizales, Colombie
Mapas climáticos urbanos para la planificación ambiental en Manizales, Colombia
Dalia N. Roncancio et Iain D. Stewart

Résumés

La pollution atmosphérique constitue un grave problème environnemental dans les villes d'Amérique latine. Bien que le problème soit largement étudié, la collaboration entre urbanistes et climatologues urbains a été lente à se développer. Dans cet article, nous démontrons la gravité du problème dans la ville de montagne de Manizales, en Colombie, et proposons une "cartographie climatique urbaine" comme outil pour faciliter la communication entre scientifiques et urbanistes. Nous utilisons une approche globale impliquant des ensembles de données topographiques, démographiques et atmosphériques qui sont assemblés et analysés dans un environnement SIG. Nos cartes montrent que les modèles de vent montagne-vallée à Manizales sont d'une importance cruciale pour la ventilation de la ville et l'amélioration de la qualité de l'air. Nous faisons donc des recommandations en matière de planification environnementale pour protéger et améliorer les bassins atmosphériques naturels et les systèmes de circulation d'air de Manizales en modifiant la forme bâtie. Des recommandations sont également émises pour réduire la charge polluante dans les rues étroites du centre historique et pour étendre le réseau de stations de surveillance environnementale à travers la ville. L'expansion du réseau est conçue conformément aux directives internationales pour l'implantation des instruments météorologiques dans les zones urbaines et est destinée à servir de nouvelles utilisations pour les données du réseau, pour inclure la planification communautaire, la cartographie environnementale et l'éducation au climat urbain.

Haut de page

Texte intégral

We extend our sincere thanks to the members of the Instituto de Estudios Ambientales (IDEA) for supporting this work: in particular, Fernando Mejía for offering valuable comments on the manuscript; Andrea Cuesta, Alexander Pachón, and the late Beatriz Aristizábal for sharing environmental datasets; and Freddy Leonardo Franco, Jeannette Zambrano, and Jorge Julián Vélez for engaging helpful discussions on the climate of Manizales.

1. Introduction

1Local air pollution is a serious environmental problem in Latin American cities. Although the problem is widely studied, collaborative solutions from city planners and urban climatologists have been slow to evolve (Henriquez and Romero, 2019). In Colombia, the process of city planning falls within a framework called the Plan de Ordenamiento Territorial (POT). The POT is a master document that describes a national set of goals, strategies, and policies to guide future development of urban and rural areas (Const., 1997). The POT’s urban” chapter gives a normative model for land-use zoning and city expansion, specifying suitable locations for roads, buildings, industries, and parks. However, a major shortcoming of the POT is that it fails to consider the effects of urbanisation on the local climate, or on the physical environment more broadly. This gap between urban climate and urban planning is not unique to Colombia: it exists worldwide and has been a barrier to knowledge exchange for many decades (Oke, 1988; Ng, 2012).

2Helping to bridge these two professions is the aim of urban climatic maps,” or UCMaps. These were developed in the mid-twentieth century in industrialised cities of Germany (Scherer et al., 1999; Ng & Ren, 2015). Their purpose is to convey simple, graphical information about the local climate to city planners and local decision makers. Two types of UCMaps are recognised: (1) analysis maps,” which combine climatic and land cover / land use information to show temporal and spatial patterns of urban environmental conditions; and (2) recommendation maps,” which give instructions for planners to preserve (or modify) these patterns where they are deemed favourable (or unfavourable) to the urban environment. The maps are accessible and highly credible, and thus provide an important climate service to urban planners, building architects, policy makers, and property developers. However, despite the utility of UCMapping, the approach does not involve recommendations for the design of air monitoring networks to record the data needed for producing the maps. Furthermore, while many dozens of cities worldwide have processed their own UCMaps for integration with city master plans (Ren et al., 2011), most of these are located in mid-latitude or higher-income regions. Few such cases exist for Latin America where many of the social, economic, and environmental challenges are unique. Some examples include the Brazilian cities of Salvador (Andrade et al., 2015), Bello Horizonte (Gomes et al, 2017), and Recife (Freitas et al., 2021), and the Argentinian city of Tandil (Picone and Campo, 2019).

3In this article, we employ UCMapping to gather, analyse, and communicate urban climatic information for environmental planners in Manizales, Colombia. The recommendations we reach are specific to Manizales, but the approach is generalizable to other mountain cities of Latin America where air monitoring may be non-compliant with international standards. In Manizales, planners have yet to operationalise UCMaps in their work. A first step to initiate this process is to leverage existing partnerships between Universidad Nacional de Colombia (UNAL) and the municipal offices of Manizales. This is eased by the fact that both UNAL and the city administration are stakeholders in the local network of environmental monitoring stations—UNAL as its manager and the city as its funding body (section 3.1). The local network is thus a focal point of our work. Accordingly, the main objectives of our work are (1) to map and analyse local climate effects in Manizales and their relation with land cover, built form, population density, and socio-economic status; (2) to develop planning recommendations to mitigate the negative outcomes of these relations; and (3) to design an expanded network of monitoring stations to collect reliable, spatially extensive, and scale-appropriate data for future uses.

2. Study area

4Manizales is an equatorial city located at high elevation (2,126 m) on the western slopes of the Andean Central Mountain Range (Fig. 1). Its regional climate is highland tropical, characterised by abundant rainfall and moderate temperatures year-round. Mean annual rainfall is 2,000 mm, and mean annual temperature is 18 °C. Within the municipal area of Manizales, annual temperatures vary from 13 °C at high elevations to 23 °C at low elevations (Zambrano et al., 2020). The city experiences a bi-modal precipitation regime with two periods of low rainfall (mid-December to mid-March; mid-June to mid-September), and two periods of high rainfall (mid-March to mid-June; mid-September to mid-December). Seasonal rainfall in the city of Manizales is due to its location in the Intertropical Convergence Zone (ITCZ). Annually, the climate of Manizales is strongly influenced by the El Niño Southern Oscillation (ENSO), whose events are associated with warm/dry years (El Niño), cool/wet years (La Niña), and neutral years.

Figure 1. Left: location of Manizales in the Americas. Right: the eleven comunas of Manizales.

Figure 1. Left: location of Manizales in the Americas. Right: the eleven comunas of Manizales.

5The physical terrain of Manizales is rugged with deep-cut valleys, steep slopes, and long ridge-lines. Much of the city is built across a long, east/west-oriented ridge, with highest elevations to the north and east (> 2,300 m), and lowest elevations to the south and west (< 1,900 m) (Fig. 2). Situated in the Pacific Ring of Fire, Manizales is seismically active and susceptible to earthquakes and volcanic eruptions. The nearest active volcano (Nevado del Ruiz) is located 28 km southeast of the city, rising 5,300 m above sea level. Emissions of volcanic ash originating from Nevado del Ruiz are regularly observed in Manizales (e.g., Cuesta et al., 2018). Volcanic soils are abundant in the region, and thus slopes are unstable and vulnerable to landslides, mudflows, and flash floods. Such events are made worse by urbanisation and local deforestation (Cardona et al., 2020).

Figure 2. Elevation contours, municipal roads, and natural drainage of Manizales.

Figure 2. Elevation contours, municipal roads, and natural drainage of Manizales.

The long, east/west-oriented ridge supports the main transportation routes of the city.

Data source: EOS Land Viewer (https://eos.com/​products/​landviewer).

6Manizales has a population of 400,000 and an annual growth rate of 0.4 % (Alcaldía de Manizales, 2017). The city is divided into 11 comunas (or districts) defined by their social and physical geography (Fig. 1) (Arteaga, 2015). The lowest-density and highest-income comuna is Palogrande (5,000 capita km-2), which is located atop the central ridge to the east of the city centre. The most densely populated comunas (with > 20,000 capita km-2) are San Jose, Cumanday, and La Fuente, which straddle the central ridge. The built form of these comunas is mostly compact lowrise (Fig. 3). Throughout Manizales, there are very few locations of informal housing where essential infrastructure is lacking.

Figure 3. Built form and land cover of Manizales.

Figure 3. Built form and land cover of Manizales.

Categories in legend refer to the Local Climate Zone (LCZ) classification scheme (Stewart and Oke, 2012), and are indicative of population and housing densities. The map was created with Landsat 8 imagery and GIS software (as described in Bechtel et al., 2015). Also shown are the sites of climate observing stations. See Table 1 for station metadata.

7The city’s municipal road network straddles the east-west ridge (Figs. 2 and 4a). Road vehicles in Manizales are responsible for more than 90 % of the city’s air pollution emissions, with diesel-run transit buses accounting for nearly one-half of the total PM10 count (González et al., 2017). Air quality in the city is worsened by industrial pollutants from local factories (metal recycling, food processing, waste incineration; Fig. 4d), and by volcanic emissions from Nevado del Ruiz. At commercial and industrial sites, air quality is frequently degraded below recommended levels of the World Health Organization standards (González et al., 2015).

Figure 4. Views of Manizales: (a) the central ridge running east toward the city centre; (b) a wall” of residential towers across the valley floor; (c) ecologically protected areas northeast of the city; (d) a looping” pollution plume near the city centre; (e) a rooftop climate station in the neighbourhood of Bosques del Norte (see Table 1); (f) a street canyon” descending the central ridge.

Figure 4. Views of Manizales: (a) the central ridge running east toward the city centre; (b) a “wall” of residential towers across the valley floor; (c) ecologically protected areas northeast of the city; (d) a “looping” pollution plume near the city centre; (e) a rooftop climate station in the neighbourhood of Bosques del Norte (see Table 1); (f) a “street canyon” descending the central ridge.

8The urbanised area of Manizales covers 57 km2 of land. Outward expansion of the area is limited by steep and unstable slopes. However, verticalisation of the built form (i.e., construction of tall towers) allows for upward rather than outward growth (Fig. 4b). The rural terrain of Manizales is a patchwork of forested and cleared slopes used mainly for agriculture (e.g., coffee, plantains, pastureland). Very little of the area surrounding the city is protected from urban or industrial development (Fig. 4c).

3. Methods and data sources

9GIS software was used to create a comprehensive set of urban climatic maps for Manizales. Air temperature and precipitation data were plotted with an “inverse distance weighting” (IDW) method, which interpolates the values of unknown points based on the distance and values of known points. The IDW method is advantageous for mapping spatially continuous variables, such as temperature and precipitation, because it assumes the values are governed by local variation. Compared to more data-intensive methods (e.g., statistical regression), IDW is preferred where spatial data are limited and where model input originates from a relatively small number of sites. IDW is nevertheless disadvantaged by its sensitivity to spatial clustering and the presence of outliers. Social and environmental data to build the urban climatic maps were obtained from the climate observing stations of Manizales, and from Google Earth Engine, the U.S. National Weather Service, and the local environmental literature.

3.1 Network of climate observing stations

10Datasets of air temperature, precipitation, wind, and relative humidity in Manizales were assembled from a statewide telemetric network of monitoring stations called SIMAC (Sistema Integrado de Monitoreo Ambiental de Caldas). SIMAC was originally established through public-private partnerships and is operated and managed by the Instituto de Estudios Ambientales (IDEA, Universidad Nacional de Colombia, Manizales). The network gathers real-time meteorological and hydrological data at dozens of sites in the city, primarily for managing risks related to air, land, and water hazards. It is considered to be one of the most sophisticated networks in Colombia (Mejía et al., 2006). Network data are freely available through CDIAC (Centro de Datos e Indicadores Ambientales de Caldas: http://cdiac.manizales.unal.edu.co/​inicio).

11Twenty-one meteorological and hydro-meteorological stations from the SIMAC network were used in this work, based on the location and spatial representativeness of their measurements (Fig. 3; Table 1). “Representativeness” was judged according to international criteria given in Oke (2006) and Stewart and Oke (2012). Station metadata generated by SIMAC provided useful information with which to assess the microscale exposure of sensors. Personal visits to the measurement sites also helped with this process. Each of the 21 sites is equipped with an automated Davis Vantage Pro2 wireless weather station. Measurement frequency is 5 minutes at each station, and the measurement uncertainty is ± 0.3 °C for air temperature, ± 2 % for relative humidity, ± 3 % for (total) rainfall, ± 3 ° for wind direction, and ± 0.9 m s-1 for wind speed. Temperature and humidity sensors are housed in gill-type radiation shields; at each station, the shields are exposed to ambient airflow but are not mechanically aspirated. The station network is routinely checked for instrument operation and calibration by field technicians. Data are digitally transmitted from the monitoring stations to a receiving station at IDEA, where they are processed and evaluated by specialised personnel and computer software (Mejía et al., 2006). Adding to the SIMAC data are the synoptic wind records of La Nubia Airport, which are stored on the Iowa Environmental Mesonet website (https://mesonet.agron.iastate.edu).

Table 1. Metadata for climate observing stations in Manizales. Stations are rank ordered by elevation. See Fig. 3 for station locations.

No.

Station

name

Elevation

(m)

Instrument placement

Local Climate Zone

(LCZ)

1

Skinco

2,350

Ground level

LCZ 8

2

Niza

2,256

Ground level

LCZ B

3

Ingeominas

2,226

Building rooftop

LCZs 3 & A

4

Posgrados

2,195

Building rooftop

LCZ 5

5

Hospital

2,183

Building rooftop

LCZs 3 & 5

6

Yarumos

2,179

Building rooftop

LCZs A & 9

7

Liceo

2,156

Building rooftop

LCZs 3 & 5

8

Bosques del Norte

2,126

Building rooftop

LCZ 3

9

Carmen

2,112

Building rooftop

LCZ 3

10

Peralonso

2,094

Ground level

LCZs 3 & A

11

La Nubia‒Campus

2,092

Building rooftop

Unclassified

12

Airport

2,080

Ground level

Unclassified

13

Aguas of Manizales

2,064

Ground level

Unclassified

14

Emas

2,060

Ground level

LCZ B

15

Alcazares

2,057

Building rooftop

LCZs 3 & A

16

Mirador Villapilar

2,046

Ground level

LCZs 5 & A

17

Bosque Popular

1,970

Ground level

LCZs 9 & A

18

La Palma

1,965

Building rooftop

LCZs 9 & A

19

Chec

1,940

Ground level

LCZs 8 & D

20

Aranjuez

1,915

Building rooftop

LCZs 3 & A

21

Terminal de Buses

1,890

Ground level

LCZs 8 & A

For ground-level stations, instruments are mounted 2‒3 m above ground; the exception is Airport station (no. 12), where the anemometer is mounted 10 m above ground.

LCZ classes describe the dominant land cover and built form surrounding the station, to a distance of 250 m. See Stewart and Oke (2012) for further explanation. Stations were classified according to in-situ and remote assessment of sites using Google Earth. LCZ 3 = compact lowrise buildings; LCZ 5 = open midrise buildings; LCZ 8 = large lowrise buildings; LCZ 9 = sparsely built; LCZ A = dense trees; LCZ B = scattered trees; LCZ D = low plants; Unclassified = no dominant land cover or built form.

12The year 2017 was chosen for map-making and climate analysis because it contains the most complete record of observational data in recent years. All data were inspected for inaccurate or implausible values associated with instrumental or environmental errors. Erroneous data were removed from analysis or else flagged for further inspection. Data for July and November were chosen to represent the low- and high-rainfall seasons, respectively, based on the completeness of their monthly records. Climatologically, 2017 is associated with a neutral ENSO episode for January to August, and a La Niña episode for September to December. Maps representing La Niña (section 4) therefore depict climatic conditions that were cooler and wetter than El Niño episodes in 2014–16 and 2018–19.

3.1.1 Limitations of the network

13The network of climate observing stations in Manizales is extensive; however, the horizontal and vertical placement of stations has left discernible gaps in spatial coverage. This is largely due to restrictions on instrument security, equipment acquisition, and site accessibility (Mejía et al., 2006). The stations are variously sited at ground level and on building rooftops (1–3 m above the roofing surface) (Table 1; Fig. 4e). The data are therefore not easily compared due to notable differences in meteorological elements at rooftop and ground-level sites (Oke, 2006). Rooftops are well exposed to the sky and are generally hotter during the day and cooler at night, as compared to the sheltered streets below the building canopy (see Fig. 7 for station locations relative to hot roofs in Manizales). Rooftops are also exposed to the microscale turbulence effects of buildings, which can distort the recorded wind and rainfall amounts (WMO, 2008). Rooftop sensors in Manizales do not sample the “blended atmosphere” high above the building canopy, where air properties are mixed and therefore representative of the local scale (Oke, 1987).

3.2 Google Earth Engine

14Land surface temperatures (LST) for Manizales were retrieved from Google Earth Engine (GEE) and processed with QGIS software. GEE hosts a catalog of open-source satellite imagery and geospatial datasets for locations worldwide. Landsat 8 imagery was acquired for Manizales, and its top-of-atmosphere radiance values then converted to land surface temperatures using the Statistical Mono-Window algorithm in GEE (Ermida et al., 2020). The algorithm assumes blackbody (or full radiator) emissivities for all surfaces. LST values are therefore considered to be “surface brightness temperatures,” not corrected land surface temperatures. In most cities, the correction is unnecessary because typical emissivities of natural and manufactured materials (ε ≈ 0.95) are nearly equal to that of a blackbody (ε = 1.0) (Oke et al., 2017). A notable exception is corrugated iron (ε = 0.13–0.28); however, this material is rarely used in Manizales. Given the prevalence of cloudy weather in the tropics, and the infrequent passage of Landsat 8 over fixed points on Earth (every 16 days at approximately 1000 hr), few dates were associated with clear-sky conditions for analysing LST in Manizales. GEE’s catalog was searched for the clearest scenes available, which were found to be 18 December 2017 and 3 September 2019.

3.3 U.S. National Weather Service

15Numerous indexes are available for quantifying thermal comfort in urban environments. Many of these are based on the human energy balance and afford comprehensive assessments of thermal stress (e.g., Universal Thermal Climate Index) (Blazejczyk et al., 2012). However, a major drawback of these indexes is their demand for input, which often entails sophisticated measurements and use of complex models. The indexes are therefore difficult to comprehend for non-specialists. To assess the potential for thermal stress in Manizales, a simplified quantity called the Heat Index (HI) is used. HI measures the perceived (not actual) temperature felt by humans, taking into account the effects of air temperature, relative humidity, wind speed, and solar radiation (Steadman, 1979a). Its computation is simple (i.e., based on standard meteorological data), and its output is directly relatable to public health outcomes (Table 2). However, the drawback with HI is its inability to indicate the causes or microscale patterns of heat stress in urban environments, and thus its diagnostic value is limited. The formulation of HI is based on a “typical person” of standard height (1.7 m) and weight (66 kg), wearing long trousers and a short-sleeved shirt, and walking outdoors in the shade while exposed to a light breeze (~ 2.5 m s-1) (Rothfusz, 1990). Online HI calculators provided by the U.S. National Weather Service (https://www.weather.gov/​safety/​heat-index) were used to compute HI values for the afternoon conditions of July 2017 in Manizales. Required inputs are local air temperature and relative humidity, which were sourced from the city’s climate observing network. To estimate HI values in full sunshine, 7 °C was added to the standard (shade) values, as suggested by Steadman (1979b) in theoretical work on the topic.

3.4 Local literature

16To supplement wind measurements obtained from the climate stations of Manizales, additional data on mesoscale airflow were sourced from the numerical simulations of González et al. (2018). In that work, the model “WRF Chem” (Weather Research and Forecasting with Chemistry) was used to simulate wind fields and air pollution concentrations in and around Manizales. For local air pollution data, these were sourced from Cuesta et al. (2018). For gaseous airborne NOx and SO2 concentrations, the data were originally collected by Cuesta et al. at nine monitoring stations (permanent and temporary) during a 4-week observation period in November 2015. The stations were equipped with passive air samplers, which were installed at commercial, industrial, and residential sites. Due to the variable height-placement of the passive samplers (e.g., rooftops, ground level), comparability of the data is limited. Lastly, demographic and socio-economic data at the comuna level were sourced from Arteaga (2015).

4. Results

4.1 Urban climatic maps

4.1.1 Air temperature and heat index (HI)

17Air temperature (Tair) in Manizales varies at small spatial scales due to sharp changes in elevation, land cover, and built form. During the month of July 2017, the mean daily Tair at 1500 hr (i.e., the approximate time of daytime maxima) was < 21 °C for most of the urban area. In the south of the city where elevation drops, Tair at 1500 hr rises to 24‒28 °C (Fig. 5a). The combination of air temperature and relative humidity at 1500 hr does not yield dangerous HI values (Fig. 5b). At low elevations to the south, the highest HI values are just 30 °C; in these conditions, pedestrians and outdoor workers are advised to “take caution” with prolonged exposure to heat (Table 2). If those same HI values are adjusted for full sunshine, most of the municipal area exceeds 27 °C (Fig. 5c). At the lowest elevations, HI values exceed 31 °C, raising the alert-level to “extreme caution.” For the particular case of 25 July 2017, when daytime maxima in Manizales exceeded 26 °C, HI values for full sunshine required pedestrians to “take caution” in all sunlit areas of the city (Fig. 5d).

18For the nighttime temperature field of Manizales, the case of 25 July 2017 is again demonstrated. Tair at 2200 hr is < 20 °C for the entire city, except for the lowest elevations to the south (Fig. 6). In the city centre, nighttime Tair is 18‒19 °C, but at entry points to the northeast, where colder air from higher elevations flows downslope (katabatic) into the urban area, Tair is < 16 °C. Nighttime Tair increases with decreasing elevation to the south and west. At 2200 hr, the alert level for heat stress in Manizales does not require pedestrians to “take caution” anywhere in the city, regardless of the relative humidity.

Figure 5. Daytime air temperature (Tair) and Heat Index (HI) values for Manizales: (a) mean Tair at 1500 hr for July 2017; (b) mean HI values (in shade) at 1500 hr for July 2017; (c) mean HI values (in sunshine) at 1500 hr for July 2017; and (d) HI values (sunshine) for July 25 at the time of daily maximum temperature (Tmax). See Table 2 for summary of public health outcomes associated with HI values.

Figure 5. Daytime air temperature (Tair) and Heat Index (HI) values for Manizales: (a) mean Tair at 1500 hr for July 2017; (b) mean HI values (in shade) at 1500 hr for July 2017; (c) mean HI values (in sunshine) at 1500 hr for July 2017; and (d) HI values (sunshine) for July 25 at the time of daily maximum temperature (Tmax). See Table 2 for summary of public health outcomes associated with HI values.

Table 2. Heat Index (HI) values and public health outcomes (Source: U.S. National Weather Service).

HI (°C)

Alert level and heat disorders associated with prolonged exposure

or strenuous activity

27–32

Caution: fatigue is possible with prolonged exposure and activity. Continuing activity could result in heat cramps.

32–41

Extreme caution: heat cramps and heat exhaustion are possible. Continuing activity could result in heat stroke.

41–54

Danger: heat cramps and heat exhaustion are likely. Heat stroke is probable with continued activity.

> 54

Extreme danger: heat stroke is imminent.

Figure 6. Nighttime air temperatures for Manizales at 2200 hr, 25 July 2017.

Figure 6. Nighttime air temperatures for Manizales at 2200 hr, 25 July 2017.

4.1.2 Land surface temperature

19Satellite images of land surface temperature (LST) represent a “bird’s eye” (or 2-D) view of the surface. The LST values are therefore biased to rooftops and treetops because the walls of buildings cannot be “seen” by a satellite radiometer. Daytime LST distributions for September and December in Manizales are characteristic of a clear, dry weather pattern (Fig. 7). The two distributions are similar but the latter has slightly higher LST, on average, than the former. Building rooftops and paved surfaces in compact urban areas are warmest (> 35 °C), especially at low elevations, as compared to the regional air temperature (21 °C). In contrast, high-gradient and densely treed slopes are coolest (< 20 °C), particularly at higher elevations. Intermediate LST values (22–32 °C) are associated with natural surfaces, such as low grasses and scattered trees, as well as low-gradient slopes of the peri-urban region (e.g., crops, pasturelands). Open urban areas with green spaces also exhibit intermediate LST values. Notable “hot spots” in the city include the historical centre (compact mid- and lowrise buildings); densely populated residential neighbourhoods (compact lowrise buildings); and industrial facilities (large lowrise buildings) (Fig. 7).

4.1.3 Wind

20Wind in Manizales is presented at three spatial scales. Regionally, observations from the Airport show a clear pattern of diurnal wind direction and speed (0600 to 1800 hr), indicating a mountain-valley circulation (Fig. 8). At sunrise (0600 hr), winds are easterly with cool mountain air flowing downslope toward the city. The average speed is 1.8 m s-1 (i.e., “light breeze”; Table 3), with 26 % of the recorded values described as “calm” (< 0.2 m s-1). By mid-morning (1000 hr), wind direction changes to westerly (i.e., warm upslope winds) and the speed increases to 3 m s-1. This marks a transition from nighttime mountain flow to daytime valley flow, which occurs between sunrise and 1000 hr. Winds remain westerly throughout the afternoon, with speeds of 2–3 m s-1. By sunset (1800 hr), winds again reverse from valley to mountain, with a complete turnaround by 2000 hr and a reduction in speed throughout the night.

Figure 7. Land surface temperatures (LST) for Manizales at 1018 hr for (a) 18 December 2017 and (b) 3 September 2019.

Figure 7. Land surface temperatures (LST) for Manizales at 1018 hr for (a) 18 December 2017 and (b) 3 September 2019.

Images are from Landsat 8 and have a spatial resolution of 100 m. Weather conditions during both days were clear, warm, and dry, with a light northwest wind (2–3 m s-1). Map symbols: 1 = densely built commercial core; 2 = compact residential neighbourhoods; 3 = large industrial facilities; 4 = high-gradient slopes with dense trees; 5 = low-gradient slopes with scattered trees. LST for white areas inside the city boundary are obscured by cloud cover.

Figure 8. Time series for the diurnal wind cycle (0600 to 1800 hr) in Manizales.

Figure 8. Time series for the diurnal wind cycle (0600 to 1800 hr) in Manizales.

Data are from 2017 at Airport station, where regional winds are measured 10 m above open, level terrain (see Table 1). Percentage values in the centre of each wind rose indicate the occurrence of “calm” winds (< 0.2 m s-1). Nighttime observations are not available.

Table 3. Beaufort classification of wind speed over land. Scale numbers higher than 5 are not listed in the table. Source: WMO (2008).

Beaufort scale number

Description

Wind speed equivalent at standard height of 10 m above open, level terrain

Specifications for estimating wind speed over land

m s-1

km hr-1

0

Calm

0 – 0.2

< 1

Calm; smoke rises vertically

1

Light

air

0.3 – 1.5

1 – 5

Smoke rises nearly vertically; direction of wind shown by smoke-drift but not by wind vanes

2

Light

breeze

1.6 – 3.3

6 – 11

Wind felt on face; leaves rustle; ordinary vanes moved by wind

3

Gentle breeze

3.4 – 5.4

12 – 19

Leaves and small twigs move constantly; wind extends light flag; hair is disturbed; clothing flaps

4

Moderate breeze

5.5 – 7.9

20 – 28

Wind raises dust and loose paper; small branches are moved; hair disarranged

5

Fresh breeze

8.0 – 10.7

29 – 38

Small trees begin to sway; force of wind felt on body

21At the local scale, wind is influenced by regional flows, and thus the observed patterns at the two scales are generally consistent, i.e., west by day, east by night (Figs. 8 and 9). However, spatial differences arise due to variations in (1) the aerodynamic roughness of the land, especially in areas with buildings and trees; and (2) the radiant heating and cooling of the land due to soil moisture, slope orientation, and sky exposure. The directional wind pattern in Manizales is well sampled by the climate observing stations (Fig. 3; Table 1). From mid-morning to sunset, flow is generally west to east, following the upslope movement of warm, buoyant air from low elevations to high elevations (Fig. 9a). The expected wind speed is greatest where the land is dry, exposed to the sun, and aerodynamically smooth (e.g., pasture, grassland, arable land), and least where it is wet, shaded, and/or aerodynamically rough (forests, built-up areas) (Geiger, 1965; Yoshino, 1984). From sunset to mid-morning, the flow is generally east to west, following the downslope movement of cold, dense air from high elevations to low elevations (Fig. 9b). The expected wind speed at night is greatest where surface cooling is rapid, e.g., smooth, dry slopes exposed to the sky. East-facing slopes may be cooler than west-facing slopes because they store less daytime heat.

Figure 9. Wind fields for Manizales (2017), representing (a) daytime hours (0900–1800 h) and (b) nighttime hours (1900–0700 h).

Figure 9. Wind fields for Manizales (2017), representing (a) daytime hours (0900–1800 h) and (b) nighttime hours (1900–0700 h).

Position of wind arrows is determined by measurements at the climate observing stations of Manizales, as well as simulations from numerical climate models by González et al. (2018). Microscale winds are theoretical, i.e., based on the laws of slope flow (e.g., Geiger, 1965; Yoshino, 1984). Wind arrows indicate direction not speed. Contour interval for elevation is 50 m.

22At the microscale, winds flow upslope into the urban area during the day, mostly from west to east but with southerly and northerly aspects (Fig. 9a). In the city centre, air is channeled through east/west and north/south “street canyons.” During the night, air drains downslope from higher ground to the east and north, through the wide street-canyons and westward across the ridge. If sky conditions are clear and calm, air at the surface will descend from the ridgetop, through the network of north-south streets (Fig. 4f), and into the valley bottom (Geiger, 1965).

4.1.4 Precipitation

23Monthly rainfall distribution in Manizales shows a characteristic pattern: at lower elevations to the south, west, and east, rainfall is less, while higher elevations across the central ridge receive more (Fig. 10). The pattern is forced by orographic lifting of air masses toward the city (i.e., for weather systems trending upslope, or west to east). In the low-rainfall season, maximum precipitation occurs in the westernmost region of Manizales, but decreases sharply with increasing elevations to the east, which suggests a “rain shadow” effect to the lee of the city (Fig. 10a). In the high-rainfall season, a second area of rainfall maxima occurs downwind and upslope of the city centre, with amounts decreasing further upslope due to rain shadow effects (Fig. 10b).

Figure 10. Total monthly precipitation (2017) for Manizales in (a) July (low-rainfall season) and (b) November (high-rainfall season).

Figure 10. Total monthly precipitation (2017) for Manizales in (a) July (low-rainfall season) and (b) November (high-rainfall season).

Rainfall amounts are typically uneven across urban-rural areas, and thus additional climate stations beyond the mapped region were included in the spatial interpolation of isohyets for Manizales.

4.2 Climate analysis maps

24Climate analysis maps for Manizales depict the prevailing winds (daytime, nighttime), population distribution (comuna level), land cover (urban, rural), and pollution concentrations (gases, aerosols) (Figs. 11 and 12). NOx and SO2 are featured in the maps as gaseous pollutants originating from road vehicles and industrial facilities. In Manizales, atmospheric concentrations of NOx and SO2 are highest near major transportation routes and terminal bus facilities (Cuesta et al., 2018). Areas of highest NOx concentrations (and secondary pollutants) are found in comunas with the largest and/or poorest populations in Manizales, such as Ciudadela del Norte, San Jose, Cumanday, La Fuente, and La Macarena. NOx concentrations are lower in the eastern reaches of the city due to fewer vehicles and lower population densities. SO2 concentrations are greatest in the commercial core (Cumanday), at major industrial factories in the east, and near the regional bus terminal (La Fuente and La Macarena).

Figure 11. NOx concentrations, population distribution, land cover, and daytime wind direction for Manizales. NOx data are from November 2015 and are sourced from Cuesta et al. (2018).

Figure 11. NOx concentrations, population distribution, land cover, and daytime wind direction for Manizales. NOx data are from November 2015 and are sourced from Cuesta et al. (2018).

Figure 12. SO2 concentrations, population distribution, land cover, and nighttime wind direction for Manizales. SO2 data are from November 2015 and are sourced from Cuesta et al. (2018).

Figure 12. SO2 concentrations, population distribution, land cover, and nighttime wind direction for Manizales. SO2 data are from November 2015 and are sourced from Cuesta et al. (2018).

Daytime case

25During the day, air circulations move upslope (anabatic) into the urban area, mainly from the west but with southerly and northerly aspects. The air moves across green rural areas, then ventilates urban areas downwind (Fig. 11). However, in comunas to the lee of high NOx and SO2 concentrations (in congested or industrialised parts of the city), populations are exposed to drifting and recirculating pollutants in the lowest layer of the atmosphere. The exposure risk is partly offset by the instability of the urban atmosphere during daytime hours (Oke et al., 2017). This has been reported for Manizales by González et al. (2015) and Cuesta et al. (2020). In general, instability enhances vertical mixing and dispersal of air pollutants away from their ground-level sources, although in certain conditions (e.g., clear, warm days with light winds) pollutants can re-circulate back to ground level when carried by “looping” plumes from elevated point sources (e.g., factory smokestacks, Fig. 4d) (Bierly and Hewson, 1962). NOx and SO2 emissions from industrial factories in the east of Manizales are dispersed by local winds flowing upslope. The factories are suitably located for daytime transport of pollutants away from populated areas of the city.

Nighttime case

26At night, air circulations move downslope (i.e., katabatic) into the urban area. The air moves across green rural areas, then ventilates the city downwind. Vegetated mountain slopes to the east and northeast are sources of cool, clean air, which helps to disperse pollutants from northern and central comunas, and from the city centre where gaseous (NOx, SO2, O3) and aerosol (PM10) concentrations are high (Fig. 12) (González et al., 2018). The air flows approximately parallel to the well-travelled, east/west-oriented streets. However, to the east of the urban area are major sources of industrial pollutants (e.g., SO2), which are carried downwind to the easternmost neighbourhoods of Manizales and into the city centre (Cuesta et al., 2020). Furthermore, the effectiveness of the mountain winds to carry pollutants away from the city is diminished by a stable (or neutral) atmosphere that prevails in urban areas at night (Oke et al., 2017). A stable atmosphere is poorly mixed and thus ground-level air pollutants are trapped at the surface. Relative to nighttime wind and stability conditions, factories on the east side of Manizales are poorly located.

5. Discussion

5.1 Urban effects on local climate

27It is well known that the water and energy balances of cities have been modified by processes of urbanisation (Oke et al., 2017). The result is an urban atmosphere that is, on average, warmer, drier, more turbulent, and more polluted than the pre-urban atmosphere. Urban effects have been documented for hundreds of cities worldwide, but with less attention to tropical and lower-income regions (Jáuregui, 1986; Roth, 2007). The climatic maps of Manizales are suggestive of these effects at the local level; however, the effects are masked by stronger controls on climate, namely orography. Separating urban and orographic effects is methodologically challenging—yet still critically important—for cities in valley and mountain settings. This is chiefly to separate the influences of urban and global climate change on the local environment of the city.

28Most cities experience an “urban heat island” (UHI) effect, i.e., elevated air and surface temperatures in the urban area. The effect is generated by the storage of heat in urban materials; slowing of winds by trees and buildings; and sealing of natural surfaces with pavement (Oke, 1987). The Tair difference between urban and rural areas is greatest at night due to the large (small) thermal inertia of urban (rural) materials. In Manizales, study of these effects on temperature began only recently with Roncancio’s (2013) investigation of the local heat island, followed by Zambrano et al.’s (2020) statistical work on the spatial and temporal variabilities of air temperature. However, due to the dominating effects of orography, the urban heat island is not shown in isothermal maps of Manizales—urban areas are instead shown to be cooler than the countryside to the north, east, and south of the city (Figs. 5 and 6). The effects of elevation were removed by Roncancio (2013) with temperature-correction schemes, revealing an average urban-rural Tair difference of 1.5 °C for compact neighbourhoods—this is the “corrected” magnitude of the nighttime heat island. One should bear this result in mind when interpreting isothermal maps of Manizales for urban effects on climate. Roncancio’s finding for Manizales is consistent with observations of UHI in other cities of similar size and form (e.g., Unger et al., 2014; Stewart et al., 2014; Leconte et al., 2015). It is also consistent with the projected influence of global warming on Caldas state (1.2 °C by the year 2040) (Corpocaldas, 2019). Local climate conditions in the urban area of Manizales are thus a good proxy for future warming at the state level.

29Unlike the atmospheric heat island whose magnitude peaks at night, the surface heat island peaks in early afternoon when solar receipt on roofs and roads is greatest (Oke et al., 2017). Spatially, the daytime surface heat island of Manizales corresponds with the urban footprint—this is evident in the LST maps of the city (Fig. 7). The maximum LST difference between urban and rural surfaces in Manizales is 25 °C. Roofs and paved surfaces are sources of sensible heat, which adds instability to the lower atmosphere. This can be advantageous for the transport of air pollutants, especially if the pollution plumes are warm and buoyant and thus able to rise high above (and downwind of) the city. With continued growth of the city, the UHI will expand its footprint as the natural land cover is replaced with urban materials of high heat capacity and low permeability.

30In all cities, winds are reduced by frictional drag over the rough urban surface (Oke, 1987). In Manizales, “wind streets” are expected to develop in neighbourhoods where air moves downslope (upslope) through the urban canyons during the night (daytime) (Fig. 4f). The depth of this flow is characteristically small and its movement slow and intermittent (on the order of 1 m s-1), but its ability to disperse air pollutants at street level is crucially important (Geiger, 1965). Also important is a thermal (heat island) circulation called the “country breeze” (Oke, 1987). On clear, calm nights, warm air rises from the heated city, drawing in cooler air from the countryside: the greater the urban-rural Tair and LST gradients, the greater the strength of the circulation cell. In Manizales, daytime valley winds (west to east) are accelerated at the rural-urban boundary by a country breeze trending in the same direction; conversely, at the eastern rural-urban boundary, valley winds are decelerated by a country breeze trending in the opposite direction. At night, these relations are reversed.

31Associated with the country breeze is the “urban rainfall” effect. Urban climatologists have hypothesised for many decades that convective rainfall is enhanced above and downwind of cities. Numerous studies have demonstrated this effect, especially for large industrial cities on level terrain in the middle latitudes (e.g., Lorenz et al., 2019). Three physical processes are at work: (1) forced convection above the city due to the UHI effect; (2) air convergence and uplift over the city due to the frictional effects of buildings; and (3) injection of condensation nuclei into the urban atmosphere from industrial smokestacks. In Manizales, urban-enhanced rainfall from the thermal, chemical, and mechanical influences of the city remains an open question (Fig. 10b) (Cortés, 2010). The question may be intractable, however, because urban and orographic effects are difficult to separate in the precipitation record (Lowry, 1998). Attempts to investigate such cause-effect relations would need controlled experiments to track the movement and intensity of individual storms in the region, while also employing computer models and rain-gauge networks across and downwind of the city (e.g., Chagnon et al., 1971). 

5.2 Planning recommendations

32Due to the highland climate of Manizales, planning recommendations to address heat stress are not compelling, although this might change with global warming and intensification of UHI effects in the future. Our attention is therefore focused on the wind potential of the city to disperse air pollutants. We recommend three planning strategies to improve air quality (and also mitigate future warming) in Manizales: (1) enhance natural ventilation of the city through preservation of local airsheds and wind paths; (2) reduce pollutant loading in street canyons of the city centre; and (3) expand instrumented monitoring of the urban atmosphere.

5.2.1 Enhance natural ventilation of the city

33Our wind analysis maps for Manizales have identified local airsheds—or atmospheric “source areas”—for clean air originating from valley and mountain slopes outside the city (Figs. 9, 11, and 12). Maintaining or enhancing these source areas is necessary for clean air to enter the city and disperse pollutants away from residential neighbourhoods, while also ventilating well-travelled streets. However, this requires legal protection of airsheds that coincide with mountaintops, exposed slopes, and natural drainage channels (Figs. 2 and 4c).

34We recommend that these airsheds not be surfaced or built upon in the future (Fig. 13). Where this may not be possible, new buildings, streets, and planting structures should be arranged parallel to the direction of airflow. Furthermore, the placement of compact highrise or midrise buildings at or near the city-airshed border is discouraged, as this creates a barrier to airflow known as the “wall effect” (Fig. 4b; see inset drawing in Fig. 13). To encourage airflow along natural wind paths, tall buildings on the urban fringe (and throughout the city) should be openly spaced, stepped in height, and aligned with the prevailing wind direction (Ng, 2012).

Figure 13. Planning recommendations to enhance natural ventilation of Manizales. Protection of local airsheds and wind paths from urban development is suggested, as is open spacing of towers.

Figure 13. Planning recommendations to enhance natural ventilation of Manizales. Protection of local airsheds and wind paths from urban development is suggested, as is open spacing of towers.

5.2.2 Reduce pollutant loading in street canyons of the city centre

35Due to the large number of gasoline- and diesel-powered vehicles in Manizales, air quality is generally poor (González et al., 2018). This is worsened in compact streets of the city centre where airflow is mainly of the “skimming” type (see inset drawing in Fig. 14). In these areas, streams of air move horizontally above the building rooftops, with minimal interaction at street level (i.e., poor ventilation of the canyon air space) (Oke, 1988). Emissions from idling or slow-moving vehicles in the canyon are therefore trapped at ground level, exposing pedestrians, vehicle operators, and street vendors to harmful air.

36We recommend a traffic management plan to re-route diesel transit buses from their existing east-west routes through the city centre. The routes should be moved 100 m north to the main avenue, which is wide and well ventilated (Fig. 14). This plan will not reduce net emissions from the city centre, but it will reduce pollutant loading in narrow, congested streets. To continue transit services to the area, planners should allow only low-emission buses (e.g., natural gas) to access these narrow streets. Also recommended is the creation of a “zero-emission zone” 300 m south of the main avenue, along an east/west-oriented street (Fig. 14). This is a popular street for pedestrian traffic because it links many local attractions, such as the library, cathedral, parks, and public squares. The section of street to be designated as a zero-emission zone should be accessible only to pedestrians and cyclists.

Figure 14. Planning recommendations to reduce pollutant loading in street canyons of central Manizales. Re-routing of diesel transit buses is suggested, as is creation of a zero-emission zone.”

Figure 14. Planning recommendations to reduce pollutant loading in street canyons of central Manizales. Re-routing of diesel transit buses is suggested, as is creation of a “zero-emission zone.”

5.2.3 Expand instrumented monitoring of the urban atmosphere

37This recommendation serves three purposes: (1) to densify the existing network of climate stations where spatial gaps exist; (2) to ensure placement of sensors at standard height for monitoring human exposure to heat and pollution; and (3) to promote international guidelines for measuring atmospheric variables in the city. While these aims may be difficult to achieve in full, they do provide a basis for adjusting the network closer to an “ideal” system for recording urban atmospheric data. The aims also address the known limitations of the existing meteorological and air pollution networks (sections 3.1.1 and 3.4). However, the network expansion proposed herein is only suggestive and must therefore be considered alongside other proposals in the local literature (e.g., Mejía et al., 2006; Cortés, 2010; González et al., 2015; Escobar et al., 2016).

38We acknowledge that there are many more sites for sensor installations than we have identified here. These should be considered as new uses emerge for the network, and as the city undergoes physical, social, and environmental change. The network itself may therefore be in a state of expansion for many years, but the choice of new sites to extend its reach (vertically and horizontally) must always begin from these guiding principles:

  1. Select sites that fill existing gaps in the network, relative to urban and rural land uses and built and natural landscapes;

  2. Select sites that need air monitoring for a concerning circumstance, e.g., landscape change or population exposure to heat or pollution;

  3. Select sites that offer instrument security and accessibility, now and in the future.

39In these principles (and those that follow), the overarching goal is not to change the location of existing stations, but to add new stations where practical and advisable. This ensures the continuity of the network and the homogeneity of its data.

Air pollution

40Existing stations for air pollution monitoring in Manizales are located at different heights above ground, in some cases > 15 m (e.g., rooftops of institutional buildings). This is problematic because pollution monitoring is sensitive to the placement of sensors relative to the pollutant source, especially in the vicinity of major streets. As an example, sensors on rooftops do not sample the air at the “breathing height” of pedestrians in the streets below.

41Our recommendations for new pollution monitoring sites are classified according to international criteria developed by Ott (1977) (Fig. 15). First, pedestrian exposure stations are advised for sites in commercial areas where people congregate (e.g., transit stops, shopping areas, public squares, pedestrian crossings). We recommend the placement of sensors at or near normal breathing height of pedestrians (i.e., 2–4 m above ground-level). Second, residential exposure stations are advised for sites in neighbourhoods that are located near industrial facilities, or that house vulnerable populations (low-income). Third, rural background stations are advised for sites in remote (non-urban) areas with no air contamination from urban or industrial activities (e.g., farms, parks, nature reserves). These background sites will set the benchmark for establishing urban or industrial effects on air quality. For all recommended sites, low-cost sensors can be mounted to lamp posts, traffic signals, utility poles, or building walls, and at heights and positions that are representative of the ground-level air. The choice of sites should include private homes and businesses where volunteers (or “citizen scientists”) can crowdsource the data (Lewis et al., 2018).

Figure 15. Planning recommendations to expand air pollution monitoring in Manizales.

Figure 15. Planning recommendations to expand air pollution monitoring in Manizales.

Sensors should be installed at or near the breathing height” of pedestrians (~ 2–4 m above ground) in commercial, residential, and rural areas. Classification of sites is based on Ott (1977).

Air temperature, precipitation, and wind

42A compact array of sensors is needed in Manizales to measure meteorological changes across short distances (Zambrano et al., 2020). The intent of our recommendations is therefore to densify the existing network and advise suitable placements of the instruments. Of considerable importance to these recommendations is that the placement of sensors is consistent with international guidelines for meteorological measurements in urban areas. For air temperature, sensors are to be installed at or near the standard height of 2 m above ground, as recommended by the World Meteorological Organization (Oke, 2006; WMO, 2008). In built-up areas, sensors should be placed in residential and commercial street-canyons that are representative of the surrounding neighbourhood (Fig. 16) (Stewart and Oke, 2012; Stewart and Mills, 2021). This ensures that the air properties encountered by pedestrians are actually being measured. Low-cost sensors can be mounted to lamp posts, traffic signals, and utility poles that are secure and well ventilated. Additionally, the property of private citizens can be used for siting air temperature sensors.

Figure 16. Planning recommendations to expand meteorological monitoring in Manizales.

Figure 16. Planning recommendations to expand meteorological monitoring in Manizales.

Installation of temperature sensors at or near standard height (~ 2 m above ground) in commercial and residential street-canyons is suggested, as is the placement of rain gauges (~ 0.5–1.5 m above ground) at open urban and rural sites. A selection of existing rooftop sites is identified for installing new (i.e., heightened) wind sensors. These should be placed well above the turbulence effects of buildings and trees. See Fig. 3 and Table 1 for site metadata about existing sites.

43For rainfall, the WMO recommends precipitation gauges be mounted 0.5–1.5 m above the ground, on level, grassy terrain and away from buildings and trees (which should be no closer to the gauge than 2–4 times their height) (WMO, 2008). This condition is easily met in rural areas, but not in cities. Suitable sites for rain gauges include lightly constructed areas with open-set buildings, such as urban parks, sports fields, work yards, or airport grounds (Oke, 2006). To enhance the existing rain-gauge network in Manizales, and to increase its density inside the urban area, we recommend placing three additional gauges at upwind, mid-wind, and downwind sites, relative to the city and its mountain-valley circulation (Fig. 16). At each site, gauges should be securely installed at ground level and in open areas that are sheltered from strong winds.

44The recommended height-placement for standard wind sensors is 10 m above ground, with no buildings or trees closer to the sensor than 10x their mean height (WMO, 2008). This condition is rarely met in cities, especially where the built form is compact. Oke (2006) therefore suggested that rooftop sensors be raised on masts to a height of at least 2x the mean building height of the local area. This ensures the sensor is free of strong wind-distortion effects from buildings and trees. In Manizales, none of the existing wind sensors meet this requirement. We therefore recommend the addition of heightened sensors at several of the rooftop sites (Fig. 16). The suggested sites are near the geographic centre of Manizales (Hospital station), and at the urban-rural boundary where winds from the countryside enter the city (Ingeominas, Yarumos, and Bosques del Norte) (Figs. 4e and 13; Table 1). The installation of new sensors at 1.5–2x mean building height will complement the existing sensors at roof level.

6. Conclusions

45We close this article with four points to underscore the value of UCMaps to environmental planning in Manizales. First, green areas and thermally driven circulations around the city must be protected. This is to ensure the long-term supply of fresh air to the city, and to mitigate the unwanted effects of urban development (i.e., air pollution) and global warming (extreme heat). Second, the replacement of fossil fuels with clean energy in road vehicles must be accelerated, along with changes to public transit routes for diesel buses in the historical centre of the city. Third, environmental monitoring in Manizales must be expanded to places where atmospheric data are critically needed. Citizen scientists and private property should be included in the expansion process to enable installation of sensors at standard height above ground. In Colombia, a successful example of this approach is the Seedbed of Scientific Citizens (Semillero de Ciudadanos Científicos) of the SIATA Early Warning System in Medellín and the Aburrá Valley. To further support the need for expanded monitoring in Manizales, citizen-based education programs at the comuna level, such as “Biociudadanos” (Bio-citizens), should be revived by IDEA. Data from these programs can help to diversity local uses of the station network, to include spatial planning, climatic mapping, and urban climate education. Lastly, knowledge gained from UCMapping must be transferred to municipal offices via existing institutional partnerships. This will initiate a long but hopeful process toward eventual integration of urban climatic maps into the Plan de Ordenamiento Territorial for Manizales.

Haut de page

Bibliographie

Alcaldía de Manizales, 2017, Boletín estadístico, perfil del sector educativo año 2017 [Statsitical bulletin, educational profile for the year 2017],” Manizales, Colombia.

Andrade T., Nery J., Moura T., Katzschner L., 2015, Urban climatic map studies in Brazil: Salvador.” In: Ng E., Ren C. (eds.), 2015, The Urban Climatic Map: A Methodology for Sustainable Urban Planning, London, Routledge.

Arteaga Botero G.B., 2015, “Análisis de indicadores de densidad en el municipio de Manizales, Colombia [Analysis of density indicators in the town of Manizales, Colombia],” Revista Ciudades, Estados y Política, Vol.2, No.1, 44–53.

Bechtel B., Alexander P.J., Böhner J., Ching J., Conrad O., Feddema J., Mills G., See L., Stewart I.D., 2015, “Mapping local climate zones for a worldwide database of the form and function of cities,” ISPRS International Journal of Geo-Information, Vol.4, 199–219.

Bierly EW, Hewson EW., 1962. “Some restrictive meteorological conditions to be considered in the design of stacks,” Journal of Applied Meteorology and Climatology, Vol.1, No.3, 383–390.

Blazejczyk K., Epstein Y., Jendritzky G., Staiger H., Tinz B., 2012, “Comparison of UTCI to selected thermal indices,” International Journal of Biometeorology, Vol.56, No.3, 515–535. 

Cardona Arboleda O.D., Carreño Tibaduiza M.L., Mendes Arraiol K.C., Alcántara-Ayala I., Saito S.M., 2020, “Inestabilidad de laderas—Deslizamientos [Slope instability—Landslides].” In: Moreno J.M., Laguna-Defi C., Barros V., Calvo Buendía E., Marengo J.A., Oswald Spring U. (eds.), Adaptación frente a los riesgos del cambio climático en los países iberoamericanos—Informe RIOCCADAPT [Adaptation to the risks of climate change in Ibero-American countries—RIOCCADAPT report], Madrid, McGraw-Hill.

Chagnon S.A., Huff F.A., Semonin R.G., 1971, “METROMEX: An investigation of inadvertent weather modification,” Bulletin of the American Meteorological Society, Vol.52, No.10, 958–967.

Const. (Constitución Política de Colombia), 1997, Artícle 9, Law 388 of 1997, 2nd ed., Legis.

Corpocaldas (Corporación Regional de Caldas), 2019, Plan Integral de Gestión de Cambio Climático del Departamento de Caldas: Documento para Responsables de Política [Comprehensive Climate Change Management Plan of the Department of Caldas: Policy-Makers Document], Manizales, Gobernación de Caldas.

Cortés Cortés A.C., 2010, Análisis de la variabilidad especial y temporal de la precipitatición en una ciudad de media montaña Andina. Caso de studio: Manizales [Analysis of the spatial and temporal variability of precipitation in a mid-Andean mountain city. Case of study: Manizales],” MSc thesis, Universidad Nacional de Colombia, Sede Manizales.

Cuesta-Mosquera A.P., González-Duque C.M., Velasco-García M., Aristizábal B.H., 2018, “Distribución espacial de concentraciones de SO2, NOX y O3 en el aire ambiente de Manizales [Spatial distribution of concentrations of SO2, NOX and O3 in the ambient air of Manizales],” Revista Internacional de Contaminación Ambiental, Vol.34, No.3, 489–504.

Cuesta-Mosquera A.P., Wahl M., Acosta-López J.B., García-Reynoso J.A., Aristizábal-Zuluaga B.H., 2020, Mixing layer height and slope wind oscillation: Factors that control ambient air SO2 in a tropical mountain city,” Sustainable Cities and Society, Vol.52, 101852.

Ermida S.L., Soares P., Mantas V., Göttsche F-M., Trigo I.F., 2020, Google Earth Engine open-source code for Land Surface Temperature estimation from the Landsat series,” Remote Sensing, Vol.12, No.9, 1471.

Escobar D.A., Martínez S., Moncada C.A., 2016, Relación entre PM10 y condiciones de accesibilidad territorial urbana en Manizales (Colombia) [Relationship between PM10 and conditions of urban territorial accessibility in Manizales (Colombia)],” Información Tecnológica, Vol.27, No.6, 273–284.

Freitas R.F.M., Azerêdo J., Carvalho L.T., Costa R.F., 2021, Climate map as an instrument for urban planning,” Revista Brasileira de Estudos Urbanos e Regionais, Vol.23, e202108en.

Geiger R., 1965, The Climate Near the Ground, 5th edn., Cambridge, Harvard University Press.

Gomes Ferreira D., Assis E., Katzschner L., 2017, Construção de um mapa climático analítico para a cidade de Belo Horizonte [Building an urban climatic analysis map for Belo Horizonte city, Brazil],” Revista Brasileira de Gestão Urbana, Vol.9, Supl. 1, 255–270.

González-Duque C.M., Cortés-Araujo J., Aristizábal-Zuluaga B.H., 2015, Influence of meteorology and source variation on airborne PM10 levels in a high relief tropical Andean city,” Revista Facultad de Ingenieria Universidad de Antioquia, No.74, March, 200–212.

González C.M., Gómez C.D. , Rojas N.Y., Acevedo H., Aristizábal B.H., 2017, Relative impact of on-road vehicular and point-source industrial emissions of air pollutants in a medium-sized Andean city,” Atmospheric Environment, Vol.152, 279–289.

González C.M., Ynoue R.Y., Vara-Vela A., Rojas N.Y., Aristizábal B.H., 2018, High-resolution air quality modeling in a medium-sized city in the tropical Andes: Assessment of local and global emissions in understanding ozone and PM10 dynamics,” Atmospheric Pollution Research, Vol.9, No.5, 934–948.

Henriquez Ruiz C., Romero H. (eds), 2019, Urban Climates in Latin America, Switzerland, Springer International.

Jáuregui E., 1986, Tropical urban climates: Review and assessment.” In: Urban Climatology and its Applications with Special Regard to Tropical Areas, WMO No.652, Geneva, World Meteorological Organization.

Leconte F., Bouyer J., Claverie R., Pétrissans M., 2015, Using Local Climate Zone scheme for UHI assessment: Evaluation of the method using mobile measurements,” Building and Environment, Vol.83, 39–49.

Lewis A.C., Peltier W., Schneidemesser E. (eds.), 2018, Low-cost sensors for the measurement of atmospheric composition: Overview of topic and future applications,” WMO-No.1215, Geneva, World Meteorological Organization.

Lorenz J.M., Kronenberg R., Bernhofer C., Niyogi D., 2019, Urban rainfall modification: Observational climatology over Berlin, Germany,” Journal of Geophysical Research: Atmospheres, Vol.124, No.2, 731–746.

Lowry W.P., 1998, Urban effects on precipitation amount,” Progress in Physical Geography, Vol.22, No.4, 477–520.

Mejía Fernández M., Londoño Linares J.P., Pachón Gómez J.A., 2006, Network of meteorological stations for prevention of disasters in Manizales—Caldas (Colombia)” [in Spanish], Taller International Workshop on Risk Management at the Local Level: The Case of Manizales, Colombia, September 28-29, Manizales, Colombia.

Ng E, 2012, Towards planning and practical understanding of the need for meteorological and climatic information in the design of high-density cities: A case-based study of Hong Kong,” International Journal of Climatology, Vol.32, No.4, 582–598.

Ng E., Ren C. (eds.), 2015, The Urban Climatic Map: A Methodology for Sustainable Urban Planning, London, Routledge.

Oke T.R., 1987, Boundary Layer Climates, New York, Methuen and Co.

Oke T.R., 1988, Street design and urban canopy layer climate,” Energy and Buildings, Vol.11, No.1–3, 103–113.

Oke T.R., 2006, Initial Guidance to Obtain Representative Meteorological Observations at Urban Sites, IOM Report 81, Geneva, World Meteorological Organization.

Oke T.R., Mills G., Christen A., Voogt J.A., 2017, Urban Climates, Cambridge University Press.

Ott W.R., 1977, Development of criteria for siting air monitoring stations,” Journal of the Air Pollution Control Association, Vol.27, No.6, 543–547.

Picone N., Campo A.M., 2019, Improving urban planning in a middle temperate Argentinian city: Combining urban climate mapping with Local Climate Zones.” In: Henriquez Ruiz C., Romero H. (eds), Urban Climates in Latin America, Switzerland, Springer International.

Ren C., Ng E., Katzschner L., 2011, Urban climatic map studies: A review,” International Journal of Climatology, Vol.31, No.15, 2213–2233.

Roncancio D.N., 2013, Study of heat island phenomenon in an Andean Colombian tropical city, case of study: Manizales-Caldas, Colombia,” MSc thesis, Universidad Nacional de Colombia, Sede Manizales.

Roth M., 2007, Review of urban climate research in (sub)tropical regions,” International Journal of Climatology, Vol.27, No.14, 1859–1873.

Rothfusz L.P., 1990, The Heat Index Equation” (or, more than you ever wanted to know about Heat Index), Fort Worth, Scientific Services Division, National Weather Service.

Scherer D., Fehrenbach U., Beha H-D., Parlow E., 1999, Improved concepts and methods in analysis and evaluation of the urban climate for optimizing urban climate processes,” Atmospheric Environment, Vol.33, 4185–4193.

Steadman R.G., 1979a, The assessment of sultriness. Part I: A temperature-humidity index based on human physiology and clothing science,” Journal of Applied Meteorology and Climatology, Vol.18, No.7, 861–873.

Steadman R.G., 1979b, The assessment of sultriness. Part II: Effects of wind, extra radiation and barometric pressure on apparent temperature,” Journal of Applied Meteorology and Climatology, Vol.18, No.17, 874–885.

Stewart I.D., Oke T.R., 2012, Local climate zones for urban temperature studies,” Bulletin of the American Meteorological Society, Vol.93, No.12, 1879–1900.

Stewart I.D., Oke T.R., Krayenhoff E.S., 2014, Evaluation of the ‘local climate zone’ scheme using temperature observations and model simulations,” International Journal of Climatology, Vol.34, No.4, 1062–1080.

Stewart I.D., Mills G., 2021, The Urban Heat Island—A Guidebook, Amsterdam, NL, Elsevier.

Unger J., Lelovics E., Gál T., 2014, Local Climate Zone mapping using GIS methods in Szeged,” Hungarian Geographical Bulletin, Vol.63, No.1, 29–41.

WMO (World Meteorological Organization), 2008, Guide to Meteorological Instruments and Methods of Observations, 7th ed. WMO-No.8, Geneva.

Yoshino M., 1984, Thermal belt and cold air drainage on the mountain slope and cold air lake in the basin at quiet, clear night,” GeoJournal, Vol.8, No.3, 235–250.

Zambrano Nájera J., Delgado V., Vélez Upegui J.J., 2020, "Short-term temperature variability in a tropical Andean city: Manizales, Colombia," Revista Vinculos, Vol.17, No.2.

Haut de page

Table des illustrations

Titre Figure 1. Left: location of Manizales in the Americas. Right: the eleven comunas of Manizales.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-1.png
Fichier image/png, 434k
Titre Figure 2. Elevation contours, municipal roads, and natural drainage of Manizales.
Légende The long, east/west-oriented ridge supports the main transportation routes of the city.
Crédits Data source: EOS Land Viewer (https://eos.com/​products/​landviewer).
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-2.png
Fichier image/png, 1,3M
Titre Figure 3. Built form and land cover of Manizales.
Légende Categories in legend refer to the Local Climate Zone (LCZ) classification scheme (Stewart and Oke, 2012), and are indicative of population and housing densities. The map was created with Landsat 8 imagery and GIS software (as described in Bechtel et al., 2015). Also shown are the sites of climate observing stations. See Table 1 for station metadata.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-3.png
Fichier image/png, 1,1M
Titre Figure 4. Views of Manizales: (a) the central ridge running east toward the city centre; (b) a “wall” of residential towers across the valley floor; (c) ecologically protected areas northeast of the city; (d) a “looping” pollution plume near the city centre; (e) a rooftop climate station in the neighbourhood of Bosques del Norte (see Table 1); (f) a “street canyon” descending the central ridge.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-4.jpg
Fichier image/jpeg, 612k
Titre Figure 5. Daytime air temperature (Tair) and Heat Index (HI) values for Manizales: (a) mean Tair at 1500 hr for July 2017; (b) mean HI values (in shade) at 1500 hr for July 2017; (c) mean HI values (in sunshine) at 1500 hr for July 2017; and (d) HI values (sunshine) for July 25 at the time of daily maximum temperature (Tmax). See Table 2 for summary of public health outcomes associated with HI values.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-5.png
Fichier image/png, 1,0M
Titre Figure 6. Nighttime air temperatures for Manizales at 2200 hr, 25 July 2017.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-6.png
Fichier image/png, 503k
Titre Figure 7. Land surface temperatures (LST) for Manizales at 1018 hr for (a) 18 December 2017 and (b) 3 September 2019.
Légende Images are from Landsat 8 and have a spatial resolution of 100 m. Weather conditions during both days were clear, warm, and dry, with a light northwest wind (2–3 m s-1). Map symbols: 1 = densely built commercial core; 2 = compact residential neighbourhoods; 3 = large industrial facilities; 4 = high-gradient slopes with dense trees; 5 = low-gradient slopes with scattered trees. LST for white areas inside the city boundary are obscured by cloud cover.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-7.png
Fichier image/png, 2,0M
Titre Figure 8. Time series for the diurnal wind cycle (0600 to 1800 hr) in Manizales.
Légende Data are from 2017 at Airport station, where regional winds are measured 10 m above open, level terrain (see Table 1). Percentage values in the centre of each wind rose indicate the occurrence of “calm” winds (< 0.2 m s-1). Nighttime observations are not available.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-8.png
Fichier image/png, 338k
Titre Figure 9. Wind fields for Manizales (2017), representing (a) daytime hours (0900–1800 h) and (b) nighttime hours (1900–0700 h).
Légende Position of wind arrows is determined by measurements at the climate observing stations of Manizales, as well as simulations from numerical climate models by González et al. (2018). Microscale winds are theoretical, i.e., based on the laws of slope flow (e.g., Geiger, 1965; Yoshino, 1984). Wind arrows indicate direction not speed. Contour interval for elevation is 50 m.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-9.png
Fichier image/png, 811k
Titre Figure 10. Total monthly precipitation (2017) for Manizales in (a) July (low-rainfall season) and (b) November (high-rainfall season).
Légende Rainfall amounts are typically uneven across urban-rural areas, and thus additional climate stations beyond the mapped region were included in the spatial interpolation of isohyets for Manizales.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-10.png
Fichier image/png, 1,1M
Titre Figure 11. NOx concentrations, population distribution, land cover, and daytime wind direction for Manizales. NOx data are from November 2015 and are sourced from Cuesta et al. (2018).
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-11.png
Fichier image/png, 885k
Titre Figure 12. SO2 concentrations, population distribution, land cover, and nighttime wind direction for Manizales. SO2 data are from November 2015 and are sourced from Cuesta et al. (2018).
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-12.png
Fichier image/png, 898k
Titre Figure 13. Planning recommendations to enhance natural ventilation of Manizales. Protection of local airsheds and wind paths from urban development is suggested, as is open spacing of towers.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-13.png
Fichier image/png, 468k
Titre Figure 14. Planning recommendations to reduce pollutant loading in street canyons of central Manizales. Re-routing of diesel transit buses is suggested, as is creation of a “zero-emission zone.”
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-14.png
Fichier image/png, 663k
Titre Figure 15. Planning recommendations to expand air pollution monitoring in Manizales.
Légende Sensors should be installed at or near the “breathing height” of pedestrians (~ 2–4 m above ground) in commercial, residential, and rural areas. Classification of sites is based on Ott (1977).
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-15.png
Fichier image/png, 344k
Titre Figure 16. Planning recommendations to expand meteorological monitoring in Manizales.
Légende Installation of temperature sensors at or near standard height (~ 2 m above ground) in commercial and residential street-canyons is suggested, as is the placement of rain gauges (~ 0.5–1.5 m above ground) at open urban and rural sites. A selection of existing rooftop sites is identified for installing new (i.e., heightened) wind sensors. These should be placed well above the turbulence effects of buildings and trees. See Fig. 3 and Table 1 for site metadata about existing sites.
URL http://journals.openedition.org/cybergeo/docannexe/image/38885/img-16.png
Fichier image/png, 332k
Haut de page

Pour citer cet article

Référence électronique

Dalia N. Roncancio et Iain D. Stewart, « Urban climatic maps for environmental planning in Manizales, Colombia », Cybergeo: European Journal of Geography [En ligne], Cartographie, Imagerie, SIG, document 1013, mis en ligne le 04 mai 2022, consulté le 19 mai 2022. URL : http://journals.openedition.org/cybergeo/38885 ; DOI : https://doi.org/10.4000/cybergeo.38885

Haut de page

Auteurs

Dalia N. Roncancio

Instructor and Researcher, Instituto de Estudios Ambientales (IDEA), Universidad Nacional de Colombia, sede Manizales
dnroncancior@unal.edu.co

Iain D. Stewart

Research Fellow, Global Cities Institute, University of Toronto, Canada
iain.stewart@utoronto.ca

Haut de page

Droits d’auteur

Licence Creative Commons
La revue Cybergeo est mise à disposition selon les termes de la Licence Creative Commons Attribution 4.0 International.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search