Introduction

A cartographic model is a raster or vectorial Geographic Information System, with spatial and geospatial information based on analytical procedures, through graphical representation of data, for analysis and structure of procedures it shows cartographic information through maps (^{Chuvieco, 1996}; ²⁰⁰²; ^{DeMers, 2002}). Analysis of geospatial data uses statistical operations. As spatial features of geospatial information are based on coordinate analysis techniques (^{Luneta and Elvidge, 1988}; ^{Ramachandran and Abrans, 2011}).

Texcoco municipality has undergone a very dynamic change related to the morphology of the landscape and a reconditioning as a geographic space. The landscapes of the municipality are changing according to the logic of the development model and territorial order. Texcoco, is immersed in the dynamics and influence of Mexico City. Basic reason for the elaboration of a cartographic model of the spatial change of the soil that allows to know the transformation, and due to the complexity of Texcoco municipality the study was carried out through sub-basins in the period from 1977 to 2000.

For the development of the research, three Landsat MSS images of 1977, TM 1989 and ETM 2000 were used, which had geometric, radiometric, and topographic correction procedures. The geometric correction was done through geometric adjustment and georeference assignment, with the support of an ortho digital photo elaborated by ^{INEGI (1996)} with a resolution of 2 m, the method used was the image/image adjustment using control points, with a third order model and a bicubic equation, with a mean square error smaller than 0.5 pixel size.

Regarding to radiometric and atmospheric corrections, the digital levels were calibrated absolutely to convert them to reflectivity levels (^{Fernández et al., 2015}). Subsequently, the topographic correction was performed by means of a digital model for each image using the ^{Civco formula (1989)}. Once the satellite images were calibrated, the classification by class was made, which consisted of distinguishing the separation between each of the classes, resulting in eleven classes. Based on field work, it was corroborated that the thematic categories coincided with the classification of the Instituto Nacional de Estadística Geografía e Informática (INEGI). Then, the graph of the standard deviation was used where it was not found overlaying between classes.

Then, to the land use maps obtained, the classification error was calculated in two ways. The first one was to apply the ERmap algorithm of the Idrisi program, where by means of another image classified with a different methodology the error of the classification between the maps of land use was compared. In the second option, the INEGI information of the land use charter of 1978 was consulted for the 1977 map. For the land use map of 1989 and 2000, the photo was interpreted as the digital photo of the year 1996 of INEGI, using the INEGI photo interpretation methodology of 1978.

When running the algorithm, a classification error of ±95% of the three classified satellite images was obtained. Finally, 170 additional control points were established to verify in the field the specific thematic classes of the 2000 land use map. At the end of this procedure, a Gaussian filter was used to standardize the information. And the land occupation maps of 1977, 1989, 2000 were obtained. A subsequent phase was the delimitation of the hydrological basins, digitizing the sub-basins of the cards scanned in tiff (Tagged Image File Format, format of high resolution images) of INEGI of the 1996 year of the topographical charts E14B21 Texcoco and E14B31 Chalco.

The ITC-Netherlands methodology for basins delineation was applied, using the digital elevation model, contours, drainage pattern, and communication channels. The basins delineation was obtained at 1:500 000 scale, obtaining seven sub-basins and delimiting the watersheds between the Basin of Mexico and the section corresponding to the Puebla-Tlaxcala basin.

Once the sub-basins were obtained, the change matrix was performed with the crossing of each sub-basin for each land occupation map per year, using the change matrix algorithm, based on the number of total pixels per column and row, to obtain the transition areas into three categories, stable areas, loss and gain. Based on the dates of 1977 and 2000. Based on these data, dynamic maps of soil occupation by sub-basin were carried out from 1977 to 2000. The cartography resulting from the processes is shown in a scale of 1: 200 000.

Based on the land cover map of 1977 and 2000, the soil cover dynamics were obtained in the seven sub-basins and the watersheds between the Basin of Mexico and the section of the Puebla-Tlaxcala Basin were delimited, wich are located within the territory of Texcoco municipality (Figure 1)

Figure 1 Map of the main hydrological sub-basins of Texcoco municipality. 

Hondo River sub-basin. This sub-basin, located at an altitude of 3 900 m, did not show great changes in relation to the pasture and forest surface. However, in the middle part of the sub-basin human activity is greater and there is an increase in pasture and in the area of rainfed agriculture. The sub-basin has a total area of 52.4 km2 of which 32 km2 correspont to stable surface, so the alteration in the landscape is minimal.

Jalapango River sub-basin. The pasture showed an increase of 9.2 km2. Rainfed agriculture increased its area by 9.3 km2, which indicates a penetration and advance of the rainfed agricultural activity. The forest decreased to 3.18 km2. Irrigated agriculture showed a loss of 6.26 km2 and a stable surface of 1.58 km2 , which indicates a loss of irrigation agricultural activity and a change in soil occupation. The urban area had an increase of 1.7 km2 within the sub-basin, and in the sub-basin of 1.48 km2.

Coxcacoaco River sub-basin. This sub-basin showed a remarkable dynamics in the types of land occupation. Previously there were some wetlands that have disappeared to date (^{INEGI, 1977}, ¹⁹⁸⁰, ¹⁹⁹⁰, ²⁰⁰⁰), and these are now used for irrigated agriculture. In the case of the pasture, it showed an increase of 4.58 km2, mainly in the mountainous part, this increase has a direct relation with the rainfed agricultural activity that develops and is introduced in the forest zone. The area of rainfed agriculture increased to 6.58 km2 in the whole sub-basin. In the case of the forest, a stable area of 16.13 km2 is maintained, and this situation is very important for this sub-basin, since from 1977 to 2000; only 2.57 km2 have changed.

Irrigation agriculture is the area that showed most changes because it currently has only 5.31 km2 of stable irrigated agricultural land, and over the course of 23 years its surface fell to 9.38 km2. The bare soil showed a decrease of 0.54 km2 and is mainly due to the soil conservation works of the restoration project of Texcoco Lake. Recently the mining activity and the exploitation of material banks showed an area of 0.53 km2. Compared to the urban area it showed an increase of 5.12 km2 mainly occupying agricultural soils under irrigation within the sub-basin. In addition, it is the most densely populated sub-basin and where the municipal head is located.

Texcoco River sub-basin. This sub-basin also showed some wetlands (^{INEGI, 1977}, ¹⁹⁸⁰, ¹⁹⁹⁰, ²⁰⁰⁰), which are currently occupied by irrigated agriculture. In the case of the pasture, it showed a surface gain of 3.18 km2, although the analysis indicates that in 23 years it showed a reduction of 1.74 km2; that is, this pasture change involves other sites within the sub-basin. The forest area is practically stable when accounting for 14.01 km2 of the 15.14 km2 recorded for the 2000, and only 1.5 km2 of forest area loss has been recorded.

Irrigated agriculture had a loss of 5.12 km2. Just as the Coxcacoaco River sub-basin that lost a quarter of its area from 1977 to date. In the rainfed agriculture the reverse phenomenon occurs because it showes a gain of 5.9 km2, and this is important to emphasize because it indicates in some way the change that are suffering other types of coverage are incorporated to the rainfed agriculture. The urban growth had an increase of 2.31 km2. The configuration of the topography with the sub-basin of the Coxcacoaco River, make these two sub-basins to concentrate the largest urban conglomerate of the municipality.

Chapingo River sub-basin. The grassland cover, has a stable surface of 1.49 km2 and reports a gain of 3.97 km2; that is, the grassland surface is incorporating other lands and, furthermore, it is moving from its original coverage. The forest recorded an increase of 1.25 km2 and the stable surface is maintained with a small variation, but the important thing is that this increase in the forest area can be observed, due to the reforestation that is carried out by means of the recovery plan of the Texcoco Lake plan. Irrigation agriculture in the sub-basin experienced a reduction of 2.76 km2 in 23 years and only recorded a stable surface of 2.2 km2. The surface of water sheets is the same, there is no variation.

San Bernardino River sub-basin. The grassland surface has a gain of 6.29 km2. The scrubland has a gain of 0.21 km2, its presence is mainly in small elevations less than 100 meters, where human activity does not alter this type of vegetation. The forest also shows a very small increase of 1.1 km2 and a stable surface in 2.02 km2, which is related to the soil conservation works and reforestation of the Texcoco Lake plan.

The location of this sub-basin is in the center of the municipality, reason why the irrigation agriculture shows only 2.74 km2 of stable surface and a loss of 2.35 km2. Compared to rainfed agriculture, there is a loss area of 6.37 km2. The surface of water remains stable, while the bare soil decreased 0.99 km2. Rock outcrops have a slight increase. The mining activity of open extraction is present in this sub-basin. The urban area had an increase of 0.86 km2.

Tejocote-Santa Monica River sub-basin. In 1977 this sub-basin showed wetlands (^{INEGI, 1977}; ¹⁹⁸⁰; ¹⁹⁹⁰; ²⁰⁰⁰), that have now disappeared as in the previous sub-basins. The grassland increased its area to 7.98 km2 and there are new areas within the forest and also showed a surface affected by the fires of the 1998 year. The scrub is one of the landscape elements in this sub-basin that shows a decrease of 1.22 km2.

The forest area is very important for this sub-basin and maintains a stable area of 26 km2 and recorded a loss of 4.12 km2, there was a loss of forest area mainly due to the forest fires that occurred in the 1998 year of those still there is no assessment of the affected area. In this sub-basin, it can be clearly seen that it is the one most affected by the loss of irrigated agricultural land showing a loss of 7.51 km2, almost half of the original surface it had in 1977.

Rainfed agriculture has increased to 6.26 km2 and is related to the areas that are being generated in the forest area. Regarding to the bare soil it has a loss of 0.51 km2 and rock outcrops maintain its original surface. The exploitation of open pit mines is also a new element in the sub-basin by registering 0.28 km2. The urban element has a gain of 2.23 km2 by increasinng four times more the original surface it had for the 1977 year.

Manzano River sub-basin and other sub-basins. The name of “other sub-basins” is for the reason that there is not a name in the cartography of the topographic chart of INEGI in the head of these rivers, although in the following topographical charts that join the Texcoco and Chalco charts must appear the name of each one. They consist of three specific zones, part of the Manzano river is located in the south end that limits with the municipality of San Vicente Chicoloapan and with Ixtapaluca, the remaining two are located in the section corresponding to the Sierra Nevada.

These two sub-basins form part of the Puebla-Tlaxcala basin but are within the natural limits. Two of these sub-basins are of great importance because they have the largest forest area of the municipality with 37.69 km2, mainly due to the accessibility of this area and the relief configuration. And they have no other type of spatial variation in soil occupation. The following variations occur in the sub-basin of the Manzano River that delimits to the south with the Texcoco municipality, where the pasture has an increase of 2.72 km2, and the scrubland maintains some stable areas in the hills product of lahar with a height less than 100 m, with an increase of 0.71 km2.

This sub-basin shows a reduction in the irrigation agriculture in its surface of 3.58 km2 and the rainfed agriculture is maintained in 2.96 km2 of constant surface in its corresponding area. In relation to the bare soil there are areas without vegetation with a gain of 0.202 km2 with presence of rock outcrops with an increase of 0.11 km2. And in this sub-basin there is also open-pit mining activity. The urban area, in a period of 23 years, had an increase of 1.33 km2 in this sub-basin, and it should be noted that this sub-basin is bordered by the municipality of San Vicente Chicoloapan where important urbanization works have recently been carried out, putting into operation the management plan of the metropolitan area of Mexico valley, Government of Estado de Mexico and Secretaría de Desarrollo Social and Governments of the Distrito Federal, 1999), this situation must have all the attention from the municipal government.

Former Lake of Texcoco basin. In the zone corresponding to the basin of Texcoco lake we locate the last areas of tular and some wetlands that are conserved in this zone, the stable surface is very small 0.007 km2, but with the accomplishment of the restoration works of the Texcoco lake’s plan there was able to gain an area of 1.2 km2 that mainly are areas that develop the vegetation of the tular, which is the habitat of the birds that still survive in this zone. The management of restoration of the Texcoco lake’s plan achieved the introduction of the halophilic pasture and covered the soil of lacustrine origin, by the 2000 year the pasture surface showed a gain of 32.03 km2.

In relation to the bare soil this has decreased to 40.16 km2 not only due to the pasture but also by the hydraulic infrastructures. The management plan introduced a forest reforestation zone with an area of 0.44 km2 . Irrigation agriculture is maintained with the same surface with a small increase of 0.74 km2. Rainfed agriculture shows an increase in the lake area of 4.23 km2. The realization of the hydraulic infrastructures and with the putting into operation of the four lakes, the zone of the Texcoco lake gained 10.44 km2 in sheets of water.

Finally, urban areas are shown in this area with a gain of 0.44 km2, where the urban development plan (City Hall of Texcoco municipality, 2006) shows that in these areas it is necessary to regulate and order them territorially, because its colonization and growth do not obey an order but in a disordered, irregular and spontaneous way. This growth takes place in an area that has ejidal property and federal zone. The change in land use in the Texcoco municipality was mainly due to the increase in the urban area, pasture, scrubland and forest, as well as in irrigated and rainfed agricultural land (Figure 2).

Figure 2 Stable and change zones of Texcoco municipality. 

Conclusions

It was observed that the subbasins with the greatest impact were: Hondo, Xalapango, Coxcacoaco, Texcoco, Chapingo and San Bernardino Rivers, since they showed more changes and alterations in their land occupation area in the last 23 years, mainly in the vegetation, pasture, scrubland and forest, as well as in the agricultural areas of irrigation and rainfed.

The Manzano River subbasin in the study period maintained a relative balance in the areas of natural vegetation and productive areas. The El Tejocote-Santa Monica subbasin is considered based on data as an area with no alteration in its land use. The area of the Texcoco Lake basin shows greater alteration with respect to the whole area of study. pest damage, weather phenomena and low temperatures, requiring a precise date of transplant field definition.

The method to be used to analyze the dynamics of land use change generated by urban growth depends on the analyst’s knowledge, skills and ability on the methods of detecting existing changes and on the image data used, as well as the characteristics of the study area. In addition, the selection of the technique will depend on the aspect to be evaluated, on the quality of information to be generated and on the implementation cost. Thus, no method is applicable in all cases, but it is essential to use one or several of them so that the authority and the decision makers can know the scope of the changes registered, the risks involved, and if possible to identify the agents that cause the change aiming to monitore the territorial planning.