Study area

The study area is located in the city of Victoria de Durango, capital of the state of Durango, Mexico, within the Guadiana Valley. It has an altitude of 1 880 m (Figure 1). The city's climate varies from semi-dry temperate to temperate sub-humid, with little rain in semi-cold summer and sub-humid in winter. The average annual temperature is 17.5 °C and the average annual precipitation is 522 mm, with maximum precipitation values and regular rains from June to October (^{Conagua, 2014}). The population is 688 697 inhabitants, which represents 37.6 % of the total population of the State. The total area of the urban area is 9 285 ha (^{Inegi, 2021}). The surface area occupied by public green areas in the Durango city corresponds to 2.1 million m² (includes parks, gardens, squares and boulevards); in this estimate, a population of 518 709 inhabitants was considered, which is equivalent to a density of 3.7 m² of green areas per inhabitant (^{Blancarte-Siqueiros et al., 2019}).

Source: Own information, with support from ESRI public images.

Figure 1 Location of the study area and the trees evaluated. 

Identification and characterization of leafy trees

The characterization of the leafy trees was done with the following criteria: trees with a diameter at breast height (DBH, 1.30 m above ground level) greater than 80 cm, total height greater than 15 m and a crown diameter greater than 10 m. In order to identify possible areas in the location of said trees in the city, a satellite image from the WorldView-2 sensor from 2015 was used, with a spatial resolution of 0.5 m (^{Anderson and Marchisio, 2012}).

Subsequently, a supervised classification was done to identify the areas with vegetation in the city. The following classes were generated: urban area, vegetation, bare soil, pavement and bodies of water. Classification was performed using the maximum likelihood algorithm, which assumes that the statistics for each class in each band are normally distributed and the probability that a given pixel belongs to a specific class is calculated, that is, the maximum probability (^{Richards, 2013}). This process was executed in ERDAS Imagine software (^{Nelson and Khorram, 2018}). Once the classification was executed, only the vegetation class was selected to identify the green areas of the city and visit the sites to verify the leafy trees. Afterwards, a forest inventory of the identified trees was prepared, for which a brigade was organized with three people trained to collect field information.

Forest inventory of urban trees

In order to obtain the measurements and phytosanitary information of leafy trees, field trips were carried out in the City. Based on the classified image obtained from the vegetation class, the leafy trees were located and the variables described in Table 1 were directly measured for each tree.

To capture the data, a field format was designed and used in which each tree was identified by genus and species, and labeled with an identification code to facilitate its geographical location.

To measure the DBH of the trees, Forestry Suppliers^® diameter tape was used, the total height was measured with a model PM-5 360 PC Suunto^® clinometer, plus a 30 m tape measure model Fluor MEDID^® for the horizontal distance at which said height would be taken (usually 20 m). The height of the crown was also measured, starting from the first branch of the tree and up to the tip of the crown. In addition, the height of the clean stem (from the base of the tree to the beginning of the crown) was calculated with the difference between the total height and the crown height. Likewise, the average diameter of the crown was calculated using the compass included in the clinometer and the 30 m measuring tape, always at the tangential friction of the tree trunk and in a north-south and then east-west direction; there, the longest branches were measured in both directions. The tree crown was also evaluated at different angles to determine the percentages of the dead crown and missing crown variables (^{Nowak, 2021}).

Ecosystem services and economic valuation

Once the inventory data was obtained, a database was generated that was sent to the i-Tree Eco v6 platform for the analysis of ecosystem services and economic valuation. i-Tree Eco v6 is a digital platform developed by the United States Department of Agriculture Forest Service and is in the public domain (^{Nowak, 2021}). Its configuration is simple and compatible with some cities in the country including Monterrey, Guadalajara, Mexico City and Durango city. The software uses inventory data by species, local weather and air pollution information by hour. After sending the field information, the server returns the diagnostic report containing estimates of carbon dioxide storage and capture, air quality, runoff control, and its economic value.

The calculations are carried out through equations taken from the literature for each species or by genus (^{Nowak, 2021}); due to space limitations, these equations and calculation processes are not described here, but are available in the manual and server page (https://www.itreetools.org/). In the case of economic valuation, the user can add local price data such as m³ of water, kWh and price per ton of CO₂.

The limitations of the software are that the air pollution data dates back to 2015, and currently there may be variations. Furthermore, there are no economic valuation models for the removal of contaminants; these are calculated with methods used in the United States and the user makes the conversion to national currency, based on the purchasing power parity of each country.

Environmental and social evaluation

For the environmental evaluation, the variables of temperature and solar radiation were taken. Outdoor temperature data were recorded for each tree, taken with a TER-150 Steren^® digital thermometer for indoors and outdoors with humidity sensor. In order to observe the temperature change caused by the trees individually. Two measurements were considered for each tree, the first was taken under the shade of the trees and the second directly under the sun at the same time. The measurements were taken between 2:00 p. m. and 5:00 p. m. when the sun radiates with greater intensity (^{Roque, 2018}). The thermometer was placed on the surface and the temperature was waited for to stabilize (approximately 15 minutes) at each point. To determine the solar radiation (Watts m²) intercepted by the foliage of the trees, an SM206 Walfront^® solarimeter was used. Like temperature, two measurements were considered for each tree, radiation was measured under the shade of the tree and directly under the sun at the same time.

In order to know the perception that society has about long-lived and leafy trees, a social evaluation was carried out through the application of personal surveys. Social benefits such as recreation, mental health, provision of shade and historical, educational or landscape relevance were included. Also, environmental benefits were included such as CO₂ capture, soil retention, habitat for wildlife, microclimatic regulation and avoided runoff, in addition to risks and damages such as falling branches, weak trunk, exposed roots, damage to permanent structures, among others. Additionally, sociodemographic variables such as income level, education, age, etc. were captured.

The survey used a five-point Likert scale (^{Matas, 2018}) to identify the degree of agreement on the importance of these trees in the city, where 1 was the minimum value assigned and represents total disagreement and minimum degree of importance, and 5 represents the maximum degree of conformity and greater importance. A random sample of 20 trees was made and 5 people were surveyed in each one, who always had the vision of the tree.

The information derived from the surveys was analyzed using the statistical technique of factor analysis to identify latent or unobservable variables. The number of latent variables or factors is generally less than the number of observed variables, so this technique also helps to reduce the number of correlated variables that are originally observed. Those that are highly correlated (positively or negatively) can be linked by the same factors and those that do not exhibit correlation can be associated by different factors (^{Méndez and Rondón, 2012}). The varimax rotation technique, Bartlett's test of sphericity and the Kaiser-Meyer-Olkin coefficient of sampling adequacy were applied to test the veracity of the factor analysis (^{Tabachnick and Fidell, 2013}).

Evaluation of ecosystem services

According to the results of i-Tree, the leafy trees of Durango city store more than 465 tons of CO₂ and annually sequester 1 296.1 kg of CO₂ through the process of photosynthesis. The calculation of gross carbon dioxide sequestration per year was carried out considering the average growth of the DBH per year and the leaf index, as well as the general condition of the trees. Eucalyptus (6.31 tons), Taxodium (5.68 tons) and Populus (4.16 ton) showed the greatest CO₂ accumulation, because these three genera were the most abundant in the sample. Figure 2A illustrates the storage averages per tree in each of them.

A = Average CO₂ storage by genus; B = Average annual CO₂ sequestration

Figure 2 Most abundant genera (>3 individuals) in Durango city. 

Regarding the annual sequestration of CO₂, as shown in Figure 2B, in average values per tree, the Populus genus is the one that contributes the most to the removal of carbon dioxide from the environment (41.43 kg yr^-1) followed by Eucalyptus (7.98 kg yr^-1) and Taxodium mucronatum (6.27 kg yr^-1).

There are significant differences between genders in both the current storage and annual sequestration of CO₂. Individuals of the Populus genus have the lowest carbon dioxide storage but their annual sequestration rate is the highest. The differences may be due to the leaf index and photosynthetic capacity that certain individuals have to convert solar energy, water and CO₂ into carbohydrates and, eventually, biomass (^{Muraoka et al., 2010}; ^{Pompa-García et al., 2023}).

The price per ton of CO₂ is constantly changing; for this analysis, the price per metric ton of CO₂ was determined at MXN $1 000, based on updated market references in the month of March 2023 (^{SendeCO2, 2023}). Therefore, given that the monetary value of annual carbon storage and sequestration is equal to 1kg=MXN $1.00, this results in MXN $465 204.00 in total and MXN $1 296.00 per year, respectively, for all trees.

The total removal of air pollutants by the 83 trees sampled was estimated to be 588.8 kg annually. The most abundant polluting materials in the air that can harm human health are suspended particles smaller than 10 µm (PM₁₀), and gases such as ozone, carbon monoxide, nitrogen dioxide and sulfur dioxide.

Table 3 shows that PM₁₀ removal on average is highest in Eucalyptus spp (5 217.4 g), followed by Taxodium mucronatum (4 838.9 g). According to the i-Tree report, the 83 trees annually intercept a total of 374 kg of PM₁₀, 85 kg of carbon monoxide (CO), 89 kg of ozone (O₃), 4.8 kg of nitrogen dioxide (NO₂) and 35.4 kg of sulfur dioxide (SO₂). The polluting gases removed by leafy trees in Durango city amount to a value of MXN $234 262.

It was estimated that the replacement value of leafy trees in the city of Durango is approximately MXN $23.6 million. This amount, which refers to the intrinsic value of trees (^{Piccolo, 2017}), is composed of the total cumulative value of CO₂ storage (MXN $465 000), annual functional values of ecosystem services such as CO₂ sequestration (MXN $1 300.00), avoided runoff (MXN $5 190) and the removal of polluting gases from the air (MXN $234 000). The relationship that exists between the replacement value of these trees with the DBH and total height is positive.

The 83 trees sampled contribute 286 m³ per year of avoided runoff, that is, the portion of the precipitation that infiltrates into the subsoil. This represents an economic benefit of MXN $5 186 per year. To obtain these values, costs as of January 2023 were used, coming from the fees and rates applied by the municipal office in charge of managing drinking water (Municipal Waters of Durango). Precipitation data from meteorological station 10 092 located in of Durango city were also used. Figure 3A shows the averages of water intercepted by Eucalyptus spp. (19.84 m³ yr^-1) and Taxodium mucronatum (18.41 m³ yr^-1); non-significant differences were observed between both genera compared to Populus spp. (11.05 m³ yr^-1). Figure 3B illustrates the averages of stormwater runoff avoided by Eucalyptus spp. (3.99 m³ yr^-1), Taxodium mucronatum (3.69 m³ yr^-1) and Populus spp. (2.22 m³ yr^-1). The latter presented significant differences with the other two concurrent genera (F=13.3, p<0.01).

A = Average intercepted water in m³; B = Average avoided runoff in m³.

Figure 3 Most abundant genera in the sample in Durango city. 

Regarding the effect of tree canopies on radiation and ambient temperature, it was found that there is a large difference between the data taken under the tree canopy and those taken directly under the sun. Regarding radiation, the range extends from 400 Wm² to 1 200 Wm². While the temperature ranges between 6 and 20 °C; this contrast is reflected in the immediate surroundings of the tree. The more abundant the foliage, the less radiation reaches the ground and therefore the temperature is reduced. The results coincide with other studies that evaluated differences in temperature. ^{Borelli et al. (2018)} mentions that shaded surfaces can be from 11 to 25 °C cooler than the temperatures of places exposed to the sun.

Survey results

The factor analysis revealed three important factors that together accumulated 83 % of the total variation. The first factor (damages or disservices, by its definition in English) included variables such as weak branches and trunk, exposed roots and damage to infrastructure, and recorded 42.8 % of the variation; the second (benefits) included variables such as recreation, shade, and mental health, and accumulated 32.1 % of the total variation; and the third (existence value) included variables such as presence, coverage and distribution in the city, with 8.4 %. The scores from each factor suggest that the people surveyed attribute slightly greater importance to the harms than to the benefits that these trees represent. However, the t-student test did not show significant differences between the two factors (t=1.81, p=0.09).

Through one-way analysis of variance, it was evaluated whether the geographical location of the trees is a key indicator to express their perceptions through the factor scores of each block. The analysis showed that only in the case of damages, significant differences were found in terms of individual perception (F=4.07, p=0.023).

The trees in the central area of the city had values close to zero, while those on the periphery gave negative values. That is, the damage caused by the trees was less noticeable on the outskirts of the city than in the central area. This may be because they are more dispersed, in smaller numbers and with less damage to infrastructure.

Comparison with other studies

There are few studies that have described the importance of this type of trees inside and outside Mexico using similar methodologies. ^{Benegas et al. (2021)} determined that the trees of Turrialba, Costa Rica, generate more than US $60 000 per year through their existence value and carbon sequestration. One of the first cases that used i-Tree Eco in Mexico for the evaluation of urban trees was in the city of Mérida, Yucatán State by ^{de la Concha et al. (2017)}. They concluded that the average diameter of the 2.3 million trees evaluated was 13 cm, which sequester around 16 637 tons per year and their economic value was estimated at US $2.5 million per year. ^{Martínez-Trinidad et al. (2021)}, in their study on the urban trees of the city of Texcoco, State of Mexico, and also with the i-Tree platform, reported that of the 391 trees of different sizes that were evaluated, their total economic value was MXN $269 481 or MXN $875 per tree (prices updated as of May 2023). The average data for these trees were DBH=25.3 cm, total height=8 m and crown diameter=3.7 m. In the case of Durango, a value of approximately MXN $8 505 per tree was obtained. These comparisons indicate that the value of leafy trees is almost 10 times greater than those evaluated in the city of Texcoco.

The results of this research suggest that the density and diversity of leafy tree species in Durango city is low, compared to other cities in Mexico. In the Durango city, 83 leafy trees belonging to eight different genera were found in about 9 200 ha, which shows a very low density. The most frequent taxa were: Eucalyptus spp., Taxodium mucronatum and Populus spp. Climate and geographical location are factors that not only limit the quantity and diversity of tree species, but also their photosynthetic capacity, carbon accumulation and productivity (^{Pompa-García et al., 2023}). ^{Leal et al. (2018)} recorded in the city of Linares, Nuevo León State, a total of 41 species of urban trees, where trees with DBH greater than 80 cm had a density of 13 individuals per hectare.

As ^{Jim (2018)} mentioned, the study of leafy trees can help raise awareness of the benefits they generate for society and obtain support to increase their number in appropriate places and improve their management. ^{Villanueva et al. (2010)} estimated that a tree of this type (particularly Taxodium mucronatum) takes at least 200 years to reach this classification. During that time, many benefits are obtained, and many times they cannot be replaced by young individuals. Therefore, it is important to identify those trees that can be treated as leafy or heritage and place records that detail their physiognomic characteristics and, where appropriate, the history associated with social events. Its dissemination on social networks or related media can also help in its conservation.