**4. Geography and GIScience**

At present, we are facing a paradoxical situation. On one hand, there is the emergence of neogeography and the proliferation of user-generated geographic content

and more precisely volunteered geographic information (VGI). On the other hand, we are witnessing a new incursion of physics into social sciences, with a growing motivation for the use of physical models in the analysis of social systems, e.g., cities [22–25] and social networks [26].

We are turning our scope to the traditional latent tension between macrogeography, centered in general principles (law-seeking), and microgeography (description-seeking), which proliferated in most of the twentieth century. The former advocates a nomothetic geographic knowledge and the later an idiographic one [27]. This is not the first time, and surely not the last, we see this, and we have differential modeling to prove it [28]. GISc holds both, the first in algorithms and methods and the second in data. One can easily draw a connection between physics and GISc, passing through spatial analysis [29]. This is the first insight that we should retain, and this is particularly true for human geographers, in which spatial context does matter.

Despite all the efforts, human geography has denied the importance of physical principle-based models, which makes it clear that the interactions between low-level system components can produce new characteristics that proved to be unpredictable, even if we fully understand the central laws that rule the system, i.e., emergency. These characteristics are obvious human systems of (auto-)organized complexity and sophisticated feedback loops (positive and negative) driving to emergent behaviors and counterintuitive effects, e.g., cities [30]. Here, spatial context is fundamental, because it influences the local pattern of interactions between system components and consequently the system dynamics that emerge from individual behaviors [31] in a phenomenon where one could designate aggregate complexity.

The use of models in any branch of geography research has proven to be more efficient than the traditional techniques of data analysis. Stating so, one do not intend to appeal to the rejection of any other technique that has demonstrated its usefulness, neither do we expect that geographers change their research objectives. The use of models at all levels is presently a different question. Models are so effective and the functional explanation of a system is much more effective than a set of disconnected facts that the use of models will prove to be of the most importance in the long run.

To advocate the use of models is not necessarily to insist on the study of general geography, spatial distribution, or the purpose of formulating general laws. In the long term, the most significant result of developing geographic models will be the establishment of genuine principles, distinct from superficial generalizations. These models are supported by a stronger basis and are enhanced with values that emerge at higher levels. For instance, considering spatial interaction models, or one of its outmost representatives: gravity model, it was first drawn as an analogy to the third Newton's gravity Law, then strengthened through entropy maximization [32, 33], and finally turned spatially explicit. The outcome was a refined panoply of spatial interaction models, including the ones focused on origin-destination spatial context [34].

This is where geography is different from natural sciences. It is where the exceptions, the discrepancies, and the uniqueness cannot be ignored because of the fact that the Earth's surface is not an isotropic space. One should not end up studying the discrepancies between the normative model and the real case, instead one should insist in finishing the study of the particular cases, of the normative element, of the special element, and of all of them as one. Nowadays, we have the knowledge and necessary techniques to study what appear to be parts of a more general system. Geography will just achieve internal consistency by being able to abstract himself from reality.

**9**

**Figure 3.**

*From geography to GIS [36].*

*Introductory Chapter: Geographic Information Systems and Science*

The choice is not only between human and physical, regional and general geography, nor is it only between regional differentiation, landscape evolution, human ecology, and spatial distribution and the recognition of spatial patterns, but may also be between objective academic studies and practical spatial planning. It may be between pointing out objectives to build coherent theories to search for order in a complex world and trying to understand those parts. In both, models are inevitable,

One may enquire whether we are looking for truth or utility: a pertinent question currently in geography. Many simulation models used for prediction and planning are black box models and may be useful in a short-term practical application even presenting a false assumption of the systems nature. We can name the abductive reasoning with this type of logical inference, which starts with data and finishes by drawing a hypothesis that best fits that data. This kind of tentative knowledge (might be true) is more fragile than induction (what is true) and deduction (what must be true) knowledge. Therefore, it should be validated and sustained by far-

One factor that has gained importance and helped to reinforce the role of quantitative geography (and to dispel fears) about model implementation is the growing mathematical component, sometimes leading to the designation of mathematical geography. At present, we recognize geomatics, i.e., mathematical geography, as a branch of geography. Thus, the concept establishes a parallelism between geography and algorithms. Hence, it can be said that quantitative geography, well seconded by the diffusion and rapid growth of the computational technology, has unfolded toward what would be the logical evolution, the appearance of the GIS. This branch has expanded too many other disciplines that use GIS and remote sensing (**Figure 3**) [36]. **Figure 3** expresses advances pushing geographic science beyond cartography into a far more multipurpose and dominant vision of "maps" that has its reflection

If GIS traditionally has a singular connexion to geography, as it has to other sciences that deal with georeferenced data, e.g., engineering and landscape architecture, GISc,

*DOI: http://dx.doi.org/10.5772/intechopen.86121*

stimulating, and economic.

reaching theories [35].

on many sciences and/or technologies.

#### *Introductory Chapter: Geographic Information Systems and Science DOI: http://dx.doi.org/10.5772/intechopen.86121*

*Geographic Information Systems and Science*

[22–25] and social networks [26].

context does matter.

gate complexity.

the long run.

spatial context [34].

and more precisely volunteered geographic information (VGI). On the other hand, we are witnessing a new incursion of physics into social sciences, with a growing motivation for the use of physical models in the analysis of social systems, e.g., cities

We are turning our scope to the traditional latent tension between macrogeography, centered in general principles (law-seeking), and microgeography (description-seeking), which proliferated in most of the twentieth century. The former advocates a nomothetic geographic knowledge and the later an idiographic one [27]. This is not the first time, and surely not the last, we see this, and we have differential modeling to prove it [28]. GISc holds both, the first in algorithms and methods and the second in data. One can easily draw a connection between physics and GISc, passing through spatial analysis [29]. This is the first insight that we should retain, and this is particularly true for human geographers, in which spatial

Despite all the efforts, human geography has denied the importance of physical principle-based models, which makes it clear that the interactions between low-level system components can produce new characteristics that proved to be unpredictable, even if we fully understand the central laws that rule the system, i.e., emergency. These characteristics are obvious human systems of (auto-)organized complexity and sophisticated feedback loops (positive and negative) driving to emergent behaviors and counterintuitive effects, e.g., cities [30]. Here, spatial context is fundamental, because it influences the local pattern of interactions between system components and consequently the system dynamics that emerge from individual behaviors [31] in a phenomenon where one could designate aggre-

The use of models in any branch of geography research has proven to be more efficient than the traditional techniques of data analysis. Stating so, one do not intend to appeal to the rejection of any other technique that has demonstrated its usefulness, neither do we expect that geographers change their research objectives. The use of models at all levels is presently a different question. Models are so effective and the functional explanation of a system is much more effective than a set of disconnected facts that the use of models will prove to be of the most importance in

To advocate the use of models is not necessarily to insist on the study of general geography, spatial distribution, or the purpose of formulating general laws. In the long term, the most significant result of developing geographic models will be the establishment of genuine principles, distinct from superficial generalizations. These models are supported by a stronger basis and are enhanced with values that emerge at higher levels. For instance, considering spatial interaction models, or one of its outmost representatives: gravity model, it was first drawn as an analogy to the third Newton's gravity Law, then strengthened through entropy maximization [32, 33], and finally turned spatially explicit. The outcome was a refined panoply of spatial interaction models, including the ones focused on origin-destination

This is where geography is different from natural sciences. It is where the exceptions, the discrepancies, and the uniqueness cannot be ignored because of the fact that the Earth's surface is not an isotropic space. One should not end up studying the discrepancies between the normative model and the real case, instead one should insist in finishing the study of the particular cases, of the normative element, of the special element, and of all of them as one. Nowadays, we have the knowledge and necessary techniques to study what appear to be parts of a more general system. Geography will just achieve internal consistency by being able to abstract himself

**8**

from reality.

The choice is not only between human and physical, regional and general geography, nor is it only between regional differentiation, landscape evolution, human ecology, and spatial distribution and the recognition of spatial patterns, but may also be between objective academic studies and practical spatial planning. It may be between pointing out objectives to build coherent theories to search for order in a complex world and trying to understand those parts. In both, models are inevitable, stimulating, and economic.

One may enquire whether we are looking for truth or utility: a pertinent question currently in geography. Many simulation models used for prediction and planning are black box models and may be useful in a short-term practical application even presenting a false assumption of the systems nature. We can name the abductive reasoning with this type of logical inference, which starts with data and finishes by drawing a hypothesis that best fits that data. This kind of tentative knowledge (might be true) is more fragile than induction (what is true) and deduction (what must be true) knowledge. Therefore, it should be validated and sustained by farreaching theories [35].

One factor that has gained importance and helped to reinforce the role of quantitative geography (and to dispel fears) about model implementation is the growing mathematical component, sometimes leading to the designation of mathematical geography. At present, we recognize geomatics, i.e., mathematical geography, as a branch of geography. Thus, the concept establishes a parallelism between geography and algorithms. Hence, it can be said that quantitative geography, well seconded by the diffusion and rapid growth of the computational technology, has unfolded toward what would be the logical evolution, the appearance of the GIS. This branch has expanded too many other disciplines that use GIS and remote sensing (**Figure 3**) [36]. **Figure 3** expresses advances pushing geographic science beyond cartography into a far more multipurpose and dominant vision of "maps" that has its reflection on many sciences and/or technologies.

If GIS traditionally has a singular connexion to geography, as it has to other sciences that deal with georeferenced data, e.g., engineering and landscape architecture, GISc,

**Figure 3.** *From geography to GIS [36].*

being an original set of concepts and expertise with an extensive applicability, has captured the attention of several traditional sciences, e.g., physics, philosophy, mathematics, etc. (**Figure 3**). As follows, GISc is many times labeled as interdisciplinary, multidisciplinary, transdisciplinary, and multiparadigmatic. Despite having different meanings, these definitions usually are used as synonymous. But, why this debate of science, subscience, or multidisciplinary field remembering the outdated—or perhaps not—discussion about if GIS and quantitative analysis are geography?

Some researchers still say that GISc is just a response to GIS technology and not a state-of-the-art science, giving the example of critical GIS, today fully integrated in GISc but initially the focus of pronounced disagreement [37]. This bifurcated propensity is clear in two polemic and completely distinct understandings [37]: one defending a mathematical and formal perspective of GISc, mainly focusing on quantitative and computational aspects [38], and other antiessentialist and antiinterpretative [39]. The latest one, being a deflationary method, allows joining a role of essential ideologies. However, taken to the extreme, it can lead to a totally amorphous field of research.
