**2.4 Implement of lower level terrestrial ecoregion classification in Yukon territory of Canada**

The major Canadian publications about territorial ecosystem classification or ecoregion classification were designed and generalized as a hierarchical, nested framework with systematic, nested hierarchical layers in the upper four layers (**Table 1**) in [38, 39].

In second case analysis, we validated the Environment Yukon's data and documental report [40–43] with our field observation. The territory of Yukon

#### **Figure 4.**

*(A) Flow diagram of ecosystem classification of land from bolson segments to vegetation stands. (B) Map of the ecological sites in sampling area.*

is approximately 483,450 km2, about 2.2 times that of Utah State in the US, and intersects with Southern Arctic, Taiga Plain, Taiga Cordillera, Boreal Cordillera, and Pacific Maritime Ecozone. Yukon's 23 Ecoregions **of 32, 51, 53, 66, 166–182,184** were described and reported (**Figure 5A**) in [40]. The Yukon Ecosystem and Landscape Classification Framework in [43] provided a classifying tool and method for mapping and implementing ecosystem classification under the Canada Ecozones and Ecoregions.

The research and field work focused on displaying and describing bioclimate features such as the horizontal distribution from south to north and vertical distribution from lower to high (**Figure 5B**). The study was characterized the broad areas influenced by similar climates into a hierarchy of bioclimate zone to lower level classification. Thus, Boreal Low (BOL), Boreal High (BOH), Subalpine (SUB), Taiga Wooded (TAW), Taiga Shrub (TAS), Tundra (TUN), Alpine (ALP) were identified as Bioclimate Zones. The broad ecosystem types by slope position and the phases by plant community dominant species were identified in the nested multiple layers and simplified in **Table 3** in Refs. [41–43]. Field survey and road investigation were carried out at the eleven observation points in 2021 summer (**Figure 5B**). The broad ecosystem types were classified by relative moisture regime as dry, moist, and wet, which can be functionally represented and retrieved the relationship by the generalized the Edatopic Grid as **Figure 6**, and using indexes of Hydrodynamic, aquatic and actual moisture, PH, similarly to it in report [43].

A DEM is a derivative product of the CanVec topographic data set. In Yukon, DEM is available for the entire territory. The generalized GIS model in Keno town area was established to generalize the lower level's bioclimate board ecosystem

#### **Figure 5.**

*(A) Yukon ecozones and ecoregions. Data source from Ecological Stratification Working Group and Smith et al. editors [38, 40]. (B) Yukon bioclimate zones, red dot – observation points. Background source from Environment Yukon [43].*

*Implement and Analysis on Current Ecosystem Classification in Western Utah of the United… DOI: http://dx.doi.org/10.5772/intechopen.100557*


*Note: bioclimatic Zone: TAW- Taiga Wooded, BOL-Boreal Low, BOH-Boreal High, SUB-Subalpine, TUN- Tundra, ALP- Alpine.*

*Bioclimatic subzones: Yukon Plateau North, Eagle Plains, North Ogilvie Mountains etc.*

*Canada ecodistrit can be searched and viewed https://databasin.org/maps/new/#datasets=8dca767690af48e6ae558 1b34612a19d*

#### **Table 3.**

*Yukon's board ecosystem classification and nested lower levels' ECL.*

#### **Figure 6.**

*Broad ecosystem gernerated with edaptopic grid scheme and slope position. The board ecosystem types can be identfied in a lanform position.*

#### **Figure 7.**

*Vegetation distribution along Keno Hill slope, Yukon.*

classification. Predictive ecosystem mapping relayed on digital elevation models (DEM) to represent landform slope and aspect conditions. These conditions provided and informed soil moisture, a primary determinant of ecosystem pattern. A demonstration was the slope survey completed near Keno city up to Monument hill (**Figure 7**). Subalpine shrub appeared above elevation 1530 m, and Salix + Carex shrub grasses from 1600 m to 1730 m. Homogenous Carex + Litchen alpine vegetation located at 1780 m become biological indicator where was near the ice valley or cold environment. Gravels + Carex + gravels belt located at 1825 m indicated that the seasonal frozen condition was occurred constantly.

### **3. Discussion**

By analyzing the upper level of ECLs in the United States and Canada, we realized that the ecosystem classification of land was a special methodology to explore and classify the ecoregions in the different countries. Bailey classified upper-level Ecosystem Classification of Land (Domain, Division, and Province), in which Domain was based on Köppen climate system classification [1–3]. Bailey, in Ref. [34], indicated that the differences in the climatic regime distinguish the natural ecosystems. The principle is that climate, as a source of energy and moisture, acts as the primary control for the ecosystem. Whether or not using Bailey's Domain as the top level of Canada's territorial Domain remained a further comparison between the United States and Canada. At least, the upper four levels' ecosystem classification and detail descriptions of Canada (see **Table 1**) would be the best fulfillment and data source. Technically, the vector and raster data can be retrieved and integrated into GIS software [14, 44–46].

The Ecological Framework of Canada in Refs. [37–39] used different classification schemes and presented the upper four levels of ecosystem classification with features of hierarchy structure in a subcontinent scale. Canada's top-level fifteen Ecozones have overlaid and intersected with Bailey's 100 Polar Domain, 200 Humid Temperate Domain, and 300 Dry Domain. For instance, Bailey's 100 Polar Domain overlays the area of Canadian eight Ecozones, Bailey's 200 Humid Temperate Domain covers the area of Canadian six Ecozones. In addition, the Prairies in Canada is extended from 200 Humid Temperate Domain to 300 Dry Domain in the US.

ECOMAP defined by the National Hierarchy of Ecological Unit (NHEU), had presented the "top-down" approach of Ecosystem Classification of Land in the United States. Western Utah's project had proved that it was a cost matter through a complete ECL's field survey. Another consequence of the strictly top-down nested hierarchical design of ECOMAP is that progressively smaller and unique polygons

#### *Implement and Analysis on Current Ecosystem Classification in Western Utah of the United… DOI: http://dx.doi.org/10.5772/intechopen.100557*

are created for each level. In other words, the ECOMAP process applied so far prevents one from easily relating features at one location to those within other landform units or bolson segments. Thus, ECOMAP is a top-down regionalization with hierarchically nested features for an explicitly geographic area. At the same time, these futures allow the ecosystem classification units to be used for various needs, from local to national. These features in the NHEU are the perimeters of outer polygons created at lower levels have to be vertically integrated with the delineation of polygons occurring at upper levels.

The limitation is for this "top-down" process; if the lowest levels are produced independently from higher levels, we still cannot answer whether the similarity of the same label polygon or unit is the same until a field survey is conducted or references available.

Much information for local managers and management companies, not all information very useful for Ecological land of classification. We did not expect any ecological research had funding to complete for mapping as to details. The project in a dry domain area with a 10 level classification would be more theoretical than practical management.

While network linked rather than nested hierarchically could be employed, we propose a simpler, more straightforward solution. Our actions were carried out a complete hierarchical land classification from a top-down approach. Ideally, we treated the ecosystem like an "organism" and separated it into components, following a top-down nested hierarchy to its finest subdivisions, and countered in common sense and practicality. Thus, a terrestrial ecosystem is considered as a volume of earth space with organic contents. We separated it from its neighbors by reasonable divisions by the empirical observation and knowledge in climatology, geography, ecology, soil, and physiography in [47–51].

While it is recognized that the National Ecological Framework with the terrestrial ecoregions in **Table 1** is a referential part of the Yukon ELC Framework, maintaining these layers for Yukon as attributive layers and data in the GIS model that is recommended in [40–43]. Specially, using 100 Domain as a top level ELC. Canada's Ecozone was considered as second level ELC. Canada's Ecoprovince in Yukon Territory was equivalent to the Bioclimate Zone, and Ecoregion was equivalent to the Bioclimate Subzones. Canada's Ecodistrict was established and can be used as identical fifth ELC layer. The sixth and seventh ELCs were related to Bioclimatic Board Ecosystem in terms of slope position and plant population important index. Canada's eight ELC was objectively defined Ecological Site or bioclimatic Ecosite. Thus, we established a complete ELC in Yukon Territory (**Table 3**).

The management approach and applications for the broad ecosystem classification and mapping are listed in **Table 4**.


#### **Table 4.**

*Broad ecosystem classification mapping and applications.*

Practically, the lower level cases of Canada territorial Ecosystem Classification had preferred more practice and objective. The researchers can use GIS technology and Spatial Analysis Modeling to efficiently produce the different maps for the landowner, management companies, and government agencies. In addition, plant ecologists had sophistical experiences in [18, 30, 33, 44, 52–57] to develop the vegetation classification and ecoregion map with a nested structure using biogeoclimatic principles. The map products were delivered by the scaled-based ecosystem classification and represented them with a high relation among the long-term climate condition, climax vegetation, and dominant plant species.

In addition to Bioclimatic Board Ecosystem Classification, Biogeoclimatic Ecosystem Classification (BEC) approach was often demonstrated as a quick approach and identified as an ecological framework for vegetation classification, mapping, and monitoring vegetation dynamics in [33, 44, 53–55, 58]. BEC approach has been used in many provinces in Canada, and the association-based ecological units of BEC are the fundamental units, for example, that the boreal vegetation association was integrated for its boundary justification. Also, the BEC approach delineated ecologically equivalent climatic regions and displayed the site conditions in the Edatopic Grid with a relationship between soil nutrient regime and soil moisture regime in [53, 54].

Ecologists studied different computational models in ecological classification such as LeNet, AlexNet, VGG models, residual neural network, and inception models in Refs. [16, 17, 24, 28]. The biggest challenge was faced in the need for an extensive training dataset to achieve high accuracy. Examples trained algorithms and the machine can only detect what criteria have been previously shown and selected. Deep learning, or machine learning algorithms, was going on method for analyzing nonlinear data with complex interactions. Moreover, they can achieve remarkable accuracy for identification and classification tasks. As a result, achieving proper ecological predictions is more feasible now. Increasing data availability is highly related to using GIS, remote sensing, and international research networks in Refs. [45, 46, 56, 57]. Furthermore, a fundamental change in research culture is towards making ecological data open access publically. All of these developments are important factors behind deep learning and development in ecology.

With further understanding, the ecosystem classification approaches and ecological modeling experiences in [14, 44, 46, 56, 57, 59] and objectively defined

**Figure 8.** *Objectively defined ecosystem classification.*

*Implement and Analysis on Current Ecosystem Classification in Western Utah of the United… DOI: http://dx.doi.org/10.5772/intechopen.100557*

ecosystem classification can be integrated by using a computer algorithm to develop efficient tools and affordable applications (**Figure 8**) without losing hierarchical structure feature in [30]. The ECL menu had input data function by getting upper-level Domain, Division, Province, and Section digital format data, and carried out a deliverable application associated with a scaled lower level ECLs. The objective analysis generated internal function outputs and combined them in the Deep Learning Algorism. The slope model, landform model, was running based on objective needs; vegetation, soil, and geology data could be considered attribute data sources depending on the study area.

We did not discuss landscape-scale changes and boundary issues that influenced ecosystem classification, which authors already presented in Refs. [1, 2, 11, 15, 25, 31, 48, 49]. Second study case demonstrated that a full ECL generally included three components: Bailey's upper level ECL, Broad Ecosystem classification, and bottom level Ecological site. With assessment, justification, and testing, we completed a full Ecosystem Classification in a Yukon ecoregion.

Why do we use western Utah's ECL to compare with Yukon's? The direct reason is that these two ecoregions had fewer human activities and had more broad original nature ecosystems in North America. In the meantime, the climate conditions are between a Dry Domain and a Polar Domain in these two ecoregions. Our study cases led the research and study with a complete ELC in Bailey's 300 Dry Domain and 100 Polar Domain.

## **4. Conclusions**

Canada's continental upper level ecoregion framework defined the ecological Mozaic on a sub-continental scale, representing an area of the earth's ecological units characterized by interactive and adjusting abiotic and biotic factors. Therefore, using Bailey's Domain as the top level of Canada's territorial ecoregion was recommended. Similarly, many users suggested that they examined the popularity and characteristics in a study area linked to the continental and global scales in [1, 8, 59–62] whenever necessary and integrated to delineate and identify the regional ecosystem. Ecological regionalization is an abstraction from global to a local site-level, contributing to understanding nature and providing differentiated guidance to sustainable environmental management. It recommended that using the global ecoregion scheme offers the guidelines for biodiversity conservation, but it still faces obstacles in improving ecosystem services and substantial uses. We had reviewed and analyzed the regionalization process, implements in two ecoregions, and some practices. With the critical consideration of ecosystem services, global environmental change and human activities should be followed in functionalized ecological regionalization. Ecosystem regionalization is a scale-based approach to classifying land surfaces, combined with regional and continental data. We should have understood more about taking geology, landform, soils, vegetation, and climate into account to classify the regionalization in different scales and ecosystem levels for a global-wide scheme when the ecosystem studies and services have grown in the research, publication and practice.

## **Acknowledgements**

Correspondence author collaborated in USU's ECL project with Prof. Neil West (Referred West et al., 2005), and conducted the Yukon Ecosystem Classification

Project. The final study was supported by Instant Calling Spatial Arch Lab, Burnaby, BC, Canada. Thanks to Prof. Neil West for his past advice. Thanks to Simon Fraser University Library funds for eligible open access publication.
