Mapping groundwater-dependent ecosystems by means of multi-layer supervised classification

Martínez‐Santos, Pedro; Díaz-Alcaide, S.; Portillo, África de la Hera; Gómez-Escalonilla, Víctor

doi:10.1016/j.jhydrol.2021.126873

Cited by 15 publications

(3 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The remaining area, corresponding essentially to the weathered mantles, was predicted as the medium potential class (class-1) which is consistent with the results outlined in Table 6. Martínez-Santos et al (2021) argue that the results obtained by machine learning methods should be verified beyond standard machine learning indicators whenever possible. In this case, the results of the agreement map were cross-checked with variables such as the number of boreholes, the average yield of the boreholes, the percentage of wells exceeding 10 m 3 /h, and the average success rate of boreholes located in the different yield domains.…”

Section: Borehole Yield Potential Maps and Limitationsmentioning

confidence: 99%

Multiclass spatial predictions of borehole yield in southern Mali by means of machine learning classifiers

Gómez-Escalonilla

Diancoumba²,

Traoré³

et al. 2022

Journal of Hydrology: Regional Studies

Self Cite

View full text Add to dashboard Cite

Section: Borehole Yield Potential Maps and Limitationsmentioning

confidence: 99%

Multiclass spatial predictions of borehole yield in southern Mali by means of machine learning classifiers

Gómez-Escalonilla

Diancoumba²,

Traoré³

et al. 2022

Journal of Hydrology: Regional Studies

Self Cite

View full text Add to dashboard Cite

“…Depending upon intended level of detail, global mapping of GDEs can thus be associated with considerable effort in terms of data volume and processing and might be affected by inconsistent data availability and quality across regions and between different data types (Eamus and Froend 2006, Kuginis et al 2016, Pérez Hoyos et al 2016, UNDP 2022. The current state of the art in GDE mapping addressing larger regions of regional to continental scale, refers to studies associated with the US (Howard et al 2010, Brown et al 2011, Mathie et al 2011, Gou et al 2015, South Africa (Colvin et al 2002, Münch andConrad 2007), the Iberian Peninsula (Marques et al 2019, Páscoa et al 2020, Martínez-Santos et al 2021, Central Asia (Liu et al 2021), as well as the Australian continent (Doody et al 2017). These classify GDEs at different resolutions, ranging from watershed level at slightly below 100 km 2 (Howard et al 2010, Brown et al 2011) to 25 m spatial resolution (Münch and Conrad 2007, Doody et al 2017, Marques et al 2019.…”

Section: Introductionmentioning

confidence: 99%

Groundwater-dependent ecosystems at risk – global hotspot analysis and implications

Link¹,

El-Hokayem²,

Usman³

et al. 2023

Environ. Res. Lett.

View full text Add to dashboard Cite

Many land-based ecosystems are dependent on groundwater and could be threatened by human groundwater abstraction. One key challenge for the description of associated impacts is the initial localisation of groundwater-dependent ecosystems (GDEs). This usually requires a mixture of extensive site-specific data collection and the use of geospatial datasets and remote sensing techniques. To date, no study has succeeded in identifying different types of GDEs in parallel worldwide. The main objective of this work is to perform a global screening analysis to identify GDE potentials rather than GDE locations. In addition, potential risks to GDEs from groundwater abstraction shall be identified. We defined nine key indicators that capture GDE potentials and associated risks on a global grid of 0.5° spatial resolution. Groundwater-dependent streams, wetlands and vegetation were covered, and a GDE index was formulated incorporating the following three aspects: the extent of groundwater use per GDE type, GDE diversity and GDE presence by land cover. The results show that GDE potentials are widely distributed across the globe, but with different distribution patterns depending on the type of ecosystem. The highest overall potential for GDEs is found in tropical regions, followed by arid and temperate climates. The GDE potentials were validated against regional studies, which showed a trend of increasing matching characteristics towards higher GDE potentials, but also inconsistencies upon closer analysis. Thus, the results can be used as first-order estimates only, which would need to be explored in the context of more site-specific analyses. Identified risks to GDEs from groundwater abstraction are more geographically limited and concentrated in the US and Mexico, the Iberian Peninsula and the Maghreb, as well as Central, South and East Asia. The derived findings on GDEs and associated risks can be useful for prioritising future research and can be integrated into sustainability-related tools such as the water footprint.

show abstract

“…The mapping of GEF can be achieved through ecosystem fieldwork [41], and remote sensing data are necessary and feasible for studying GEF in regions. The identification methods based on remote sensing include supervised classification [42], unsupervised classification [43], index classification [44], and standardized NDVI classification [45,46]. For example, Australia carried out catchment-scale mapping research on groundwater-dependent ecosystems to provide a scientific basis for natural resource management in 2015 [47].…”

Section: Introductionmentioning

confidence: 99%

Study on Index of Groundwater Ecological Function Crisis Classification and Early Warning in Northwest China

Cui

Zhang

Wang

et al. 2022

Water

View full text Add to dashboard Cite

The natural oases in the plain area of the northwest inland basin strongly depend on the groundwater depth. With the overexploitation and utilization of groundwater, natural oases are faced with the problems of serious degradation and rehabilitation. How to evaluate the degree of the degeneration crisis of groundwater ecological function has become one of the key scientific and technological problems to be solved. In this paper, the Shiyang River basin of Gansu Province was selected as a typical research area. The remote sensing interpretation, groundwater–soil ecology comprehensive investigation, and groundwater in situ monitoring were adopted to carry out the research. Based on the correlation analysis method of natural ecology and groundwater, the interactive relationship between the natural ecological environment and groundwater depth in different ecological types of the region were studied: (1) under the arid climate condition in northwest China, the relationships between the ecological situation and the groundwater depth in different ecological types of the region were obviously different, and as a result, the optimal or limit ecological water level of groundwater in different ecological types was also different; (2) in the natural wetland area, the suitable ecological water level of groundwater was between 0.5 m to 1.5 m, and the limit ecological water level was 8.0 m; in the natural vegetation area, the suitable ecological water level was between 3.0 m to 5.0 m, and the limit ecological water level was 10.0 m; and in the farmland area, the suitable ecological water level was between 2.0 m to 5.0 m, and the limit ecological water level was 2.0 m; (3) in order to effectively protect the natural ecology in different ecological types, a five-level early warning and control index system should be established for the ecological function degeneration crisis of groundwater. It may be beneficial to promote restoration and protection of the groundwater ecological function and natural ecology in the inland area of northwest China.

show abstract

Mapping groundwater-dependent ecosystems by means of multi-layer supervised classification

Cited by 15 publications

References 38 publications

Multiclass spatial predictions of borehole yield in southern Mali by means of machine learning classifiers

Multiclass spatial predictions of borehole yield in southern Mali by means of machine learning classifiers

Groundwater-dependent ecosystems at risk – global hotspot analysis and implications

Study on Index of Groundwater Ecological Function Crisis Classification and Early Warning in Northwest China

Contact Info

Product

Resources

About