1.2 Contact details of resource persons and institutions involved in the assessment and documentation of the Technology

{'additional_translations': {}, 'value': 1049, 'label': 'Name of project which facilitated the documentation/ evaluation of the Technology (if relevant)', 'text': 'Integrated Biodiversity Management, South Caucasus (IBiS)', 'template': 'raw'} {'additional_translations': {}, 'value': 6110, 'label': 'Name of the institution(s) which facilitated the documentation/ evaluation of the Technology (if relevant)', 'text': 'Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ)', 'template': 'raw'}

1.3 Conditions regarding the use of data documented through WOCAT

The compiler and key resource person(s) accept the conditions regarding the use of data documented through WOCAT:

Yes

1.4 Declaration on sustainability of the described Technology

Is the Technology described here problematic with regard to land degradation, so that it cannot be declared a sustainable land management technology?

No

1.5 Reference to Questionnaire(s) on SLM Approaches (documented using WOCAT)

2.1 Short description of the Technology

Definition of the Technology:

Land degradation contributes to biodiversity loss and the impoverishment of rural livelihoods in Tusheti. Above all, however, land degradation are triggered by climate change as traditional land use practise might not be adapted to new climate conditions which can cause or speed up degradation processes significantly. On the other hand, degraded land often leads to low biomass volumes and this reduces the ecosystem capability to stabilise local climate conditions. The concept of Land Degradation Neutrality (LDN) and the method of using remote sensing for monitoring land degradation are tools to identify the need for local planning processes. This showcase describes the LDN monitoring concept, national targets and the technology to assess indicators, mechanism and incentives for LDN.

2.2 Detailed description of the Technology

Description:

Purpose
The continuing global degradation of land resources threatens food security and the functioning of ecosystem services by reducing or losing their biological or economic productivity. Unsustainable land-use practices such as deforestation, overgrazing and inappropriate agricultural management systems trigger the loss and degradation of valuable land resources in Georgia. These effects are visible in all countries of the South Caucasus. About 35% of the agricultural land in Georgia is severely degraded, 60% is of low to middle production quality.

Land Degradation Neutrality (LDN)
LDN is a new international concept to combat the ongoing degradation of valuable soil resources. The LDN concept was developed by the UNCCD to encourage countries to take measures to avoid, reduce or reverse land degradation, with the vision of achieving a zero-net loss of productive land. To combat land degradation in Georgia, in 2017, the national LDN Working Group set voluntary national targets to address specific aspects of LDN, and submitted them to the UNCCD Secretariat.

To effectively set up counter measures to combat land degradation it is important to have detailed spatial information on land cover and land cover changes as well as on trends in degradation (like size of areas effected by erosion). Therefore a remote sensing toolset was developed and tested in the pilot are of Tusheti protected landscapes in the High Caucasus in Georgia. This region shows increasing soil erosion problems by uneven distribution of grazing activities and was selected for developing erosion control measures within the Integrated Biodiversity Management in the South Caucasus Program (IBiS) funded by the Deutsche Gesellschaft für internationale Zusammenarbeit (GIZ).

Sensitivity Model
The Integrated Biodiversity Management in the South Caucasus (IBiS) project in cooperation with national experts in Georgia, developed and applied a remote sensing toolset called "Erosion Sensitivity Model". This remote sensing toolset helps to assess the current state and the general erosion risk. The sensitivity model is based on the RUSLE – Revised Universal Soil Loss Equation. The tool allows the calculation of erosion caused by rainfall and surface run-off. The RUSLE equation incorporates a combination of different input factors such as precipitation (R), soil type (K), slope (LS), vegetation cover (C) and protection measures (P). In this way, the estimated average soil loss in tonnes per acre per year (A) can be calculated as follows: A = R * K * LS * C * P.

The rainfall factor (R) results from a quotient from the monthly and annual mean value of precipitation. The data come from the data platform “CHELSA – Climatologies at high resolution for the earth’s land surface areas”. For the soil type factor (K), a soil map of 1:200,000 was taken. Then, depending on the soil type, different contents of sand, silt, loam and clay were used to calculate the K factor. The slope length and steepness factor (LS) is calculated from a digital elevation model (DEM) with a raster resolution of 10x10m. The DEM is derived from the topographic map 1:25,000. The global elevation model derived from SRTM data (Shuttle Radar Topography Mission) has a resolution of 30x30 m and is available worldwide free of charge. The land cover factor (C) describes the vegetation cover that protects the soil from erosion. The vegetation cover slows down the speed of the raindrops and reduces the erosive effect of the rain. It slows down surface water runoff and stabilises the soil through root systems. The main indicators, land cover and productivity, can be assessed by remote sensing. The data from satellites need to be classified and calibrated by field data (ground truthing). The technology for the assessment of these indicators with Sentinel 2 satellite images was developed and applied in 2016 to 2018 in the Tusheti region (Akhmeta municipality) in the framework of the GIZ-IBiS project. Based on spectral information from airborne or satellite images, the density of the vegetation was calculated and mapped. There are well developed vegetation indices and classification systems to derive different land cover types and vegetation densities (mainly described by the Leaf Area Index LAI or biomass indices). The LAI is the area of the leaf surface (in square meters) per square meter ground surface. Since the real surface area of the leaves is hardly measurable, the amount of biomass is a proxy for the LAI. The P-factor is rarely considered in large-scale modelling of soil erosion risk as it is difficult to estimate it with very high accuracy. Therefore, to refine the model, a more detailed DEM (digital elevation model) is required (e.g., from satellite images). Based on the input factors, a soil erosion risk map was calculated for the whole territory of the Tusheti Protected Areas (113,660 ha). Based on the different spectral bands of the Sentinel 2 satellite image, a land cover map was calculated using the Support Vector Machine (SVM) technology and spectral image information.
The results have been integrated in the development of pasture management plans ("pasture passports"). This maps and documents are indicating areas of high erosion risk that need to be excluded from grazing and the maximum number of livestock has been calculated based on the biomass maps and will be integrated into the lease contracts.
The repetition of the remote sensing after some years (e.g. 5 years) will help to evaluate, if the measures in the pasture management have been successful to stop the degradation processes.

2.3 Photos of the Technology

2.5 Country/ region/ locations where the Technology has been applied and which are covered by this assessment

2.6 Date of implementation

Indicate year of implementation:

2016

2.7 Introduction of the Technology

Specify how the Technology was introduced:

• through projects/ external interventions

3.1 Main purpose(s) of the Technology

• improve production
• reduce, prevent, restore land degradation
• preserve/ improve biodiversity

• provide information to make a spatial-territorial planning

3.2 Current land use type(s) where the Technology is applied

Land use mixed within the same land unit:

Yes

Specify mixed land use (crops/ grazing/ trees):

• Agro-pastoralism (incl. integrated crop-livestock)

3.3 Has land use changed due to the implementation of the Technology?

Has land use changed due to the implementation of the Technology?

• No (Continue with question 3.4)

3.4 Water supply

other (e.g. post-flooding):

• rainfed and mixed rained-irrigation

3.5 SLM group to which the Technology belongs

• pastoralism and grazing land management
• improved ground/ vegetation cover

3.6 SLM measures comprising the Technology

Comments:

On some pilot plots technologies to control erosion and stop land degradation have been tested. This includes fencing, rotational pasture management, mulching and installing check dams to stop gully erosion.

3.7 Main types of land degradation addressed by the Technology

Comments:

The main drivers of land degradation in the pilot area are overgrazing and trampling, off-road driving as well as infrastructure development (especially inappropriate road construction in steep slopes).

3.8 Prevention, reduction, or restoration of land degradation

Specify the goal of the Technology with regard to land degradation:

• prevent land degradation
• reduce land degradation

Comments:

The monitoring tools presented here help to monitor the development of land degradation and to evaluate measures and development trends.

4.1 Technical drawing of the Technology

4.2 General information regarding the calculation of inputs and costs

Specify how costs and inputs were calculated:

• per Technology area

Specify currency used for cost calculations:

• USD

4.3 Establishment activities

	Activity	Timing (season)
1.	National level. Baseline: Field assessment for remote sensing calibration (1x/20 years)	2017
2.	Sentinel satellite image classification (multi temporal data from 2017)	2017
3.	Statistical data from GEOSTAT Agricultural census	2014-2016
4.	Analysis of soil carbon content from existing profiles	2003 - 2006
5.	Conduct ongoing monitoring	5 years intervals
6.	Update sentinel satellite image classification	1x year
7.	Update statistical data from GEOSTAT Agricultural census	4x/year
8.	Resampling of soil carbon content near existing profiles	1x/5 years
9.	Municipal level. Spatial planning: Assessment of current stage of land degradation, anticipated gains and losses	1x/10 years
10.	Revision of spatial planning on Municipal level.	1x / 5 years

4.4 Costs and inputs needed for establishment

	Specify input	Unit	Quantity	Costs per Unit	Total costs per input	% of costs borne by land users
Labour	Remote Sensing analysis by Sentinel Satellite data	person days	50.0	200.0	10000.0
Labour	Collecting field data for satellite image callibration	person days	40.0	200.0	8000.0
Labour	Soil sampling (for carbon content)	person days	20.0	200.0	4000.0
Labour	Including results in spatial planning	person days	10.0	200.0	2000.0
Total costs for establishment of the Technology					24000.0
Total costs for establishment of the Technology in USD					24000.0

Comments:

This covers the implementation of the baseline. Calibrating the model for erosion risk and land cover classification is an big investment but can be extended to larger areas than 1000 km² with similar resources.

4.5 Maintenance/ recurrent activities

	Activity	Timing/ frequency
1.	Repeating the application of the calibrated remote sensing model for monitoring repitition	with 5 years interval
2.	Repetition of soil samples for assessing soil carbon content	with 5 years interval
3.	Analysing the results and integrate them in spatial planning and policy making	with 5 years interval

4.6 Costs and inputs needed for maintenance/ recurrent activities (per year)

	Specify input	Unit	Quantity	Costs per Unit	Total costs per input	% of costs borne by land users
Labour	Applying the calibrated remote sensing model for monitoring repetition	person days	20.0	200.0	4000.0
Labour	Repetition of soil samples for assessing soil carbon content	person days	10.0	200.0	2000.0
Labour	Analysing results and integrating in spatial planning	person days	10.0	200.0	2000.0
Total costs for maintenance of the Technology					8000.0
Total costs for maintenance of the Technology in USD					8000.0

Comments:

For the repetition of the remote sensing no new calibration of the GIS-model is needed. Only the field samples for soil carbon need to be repeated.

4.7 Most important factors affecting the costs

Describe the most determinate factors affecting the costs:

Field sample collection;
Remote sensing experts.

5.1 Climate

5.2 Topography

Indicate if the Technology is specifically applied in:

• not relevant

Comments and further specifications on topography:

The remote sensing approach was applied for the total landscape of Tusheti, including a great variety of land-forms, altitudes ranging from 1600-4000 m a.s.l.

5.3 Soils

5.4 Water availability and quality

Is water salinity a problem?

No

Is flooding of the area occurring?

No

5.5 Biodiversity

5.6 Characteristics of land users applying the Technology

Indicate other relevant characteristics of the land users:

The technology is applied by the Government.

5.7 Average area of land used by land users applying the Technology

5.8 Land ownership, land use rights, and water use rights

Land ownership:

• state

Land use rights:

• communal (organized)
• leased

Water use rights:

• open access (unorganized)

Are land use rights based on a traditional legal system?

No

5.9 Access to services and infrastructure

6.1 On-site impacts the Technology has shown

Ecological impacts

Soil

Other ecological impacts

Specify assessment of on-site impacts (measurements):

The monitoring technology was applied for the first time to draw a baseline. Based on the results, activities have been planned and pilot measures have been implemented (exclusion from grazing, reforestation, regulation of grazing intensity). Future replications of the monitoring will show changes and evaluate success of measures. The technologies to control erosion are described separately in the WOCAT database (Community-based Erosion Control [Azerbaijan]; Pasture-weed control by thistle cutting [Georgia]; High-altitude afforestation for erosion control [Armenia]; Slope erosion control using wooden pile walls [Armenia])

6.2 Off-site impacts the Technology has shown

Specify assessment of off-site impacts (measurements):

The technology is only about the monitoring (see above).

6.3 Exposure and sensitivity of the Technology to gradual climate change and climate-related extremes/ disasters (as perceived by land users)

Gradual climate change

Gradual climate change

	Season	increase or decrease	How does the Technology cope with it?
seasonal rainfall	summer	decrease	very well

Comments:

Technology is sensitive, it shows the climate change, the impact of the global change locally. The technology itself is not affected by climatic changes.

6.4 Cost-benefit analysis

How do the benefits compare with the establishment costs (from land users’ perspective)?

How do the benefits compare with the maintenance/ recurrent costs (from land users' perspective)?

Comments:

The monitoring technology was applied for the first time to draw a baseline. Based on the results, activities have been planned and pilot measures have been implemented (exclusion from grazing, reforestation, regulation of grazing intensity). Future replications of the monitoring will show changes and evaluate success of measures. The technologies to control erosion are described separately in the WOCAT database (Community-based Erosion Control [Azerbaijan]; Pasture-weed control by thistle cutting [Georgia]; High-altitude afforestation for erosion control [Armenia]; Slope erosion control using wooden pile walls [Armenia]).
The costs of the remote sensing approach have not been invested by the land owners but by GIZ and the Ministry. Therefore there are no direct negative impact caused by the investment. The maintenance will be covered by public authorities as well. The positive impact for the land users are the clearly delineated pasture unit giving the exact area of grassland and the accessible amount of fodder biomass. By this, the lease-rate can be found according to the productivity and the number of livestock can be adapted to the carrying capacity of the land within the lease contract.

6.5 Adoption of the Technology

• single cases/ experimental

If available, quantify (no. of households and/ or area covered):

The technology is desigend to be applied by national or regional addministrations and not by land owners themselves.

6.6 Adaptation

Has the Technology been modified recently to adapt to changing conditions?

No

6.7 Strengths/ advantages/ opportunities of the Technology

Strengths/ advantages/ opportunities in the land user’s view
The monitoring technology can help to find erosion and degradation hot spots and based on this spatial information counter measures can be applied to save the productivity of land. As the income from agricultural activities and livestock breeding is of high priority in this pilot region, the protection of the productivity of land is of high importance to the local land users.

Strengths/ advantages/ opportunities in the compiler’s or other key resource person’s view

The presented remote sensing technologies are a cost efficient and objective way to monitor land degradation and land use changes on large areas on long time periods. Based on this spatial data, land use regulations can be integrated in spatial planning and other legal and practical frameworks (e.g. pasture lease contracts) to counter act the degradation processes. The success of the measures and the development of degradation and rehabilitation can be monitored by the same toolset.

6.8 Weaknesses/ disadvantages/ risks of the Technology and ways of overcoming them

Weaknesses/ disadvantages/ risks in the land user’s view	How can they be overcome?
The technology is complex and cannot be applied by the land user her-/himself and is sometimes hard to understand. Therefore they might mistrust in the results and are not eager to accept regulations and measures to stop degradation.	Transparent documentation of the technology and regular field visits to evaluate together with the land owners and users the remote sensing results in the field.

Weaknesses/ disadvantages/ risks in the compiler’s or other key resource person’s view	How can they be overcome?
The institutional setup on the national level for the regular application of the remote sensing technology and the storage and management of the monitoring data is not established yet. GIS, remote sensing and soil experts are of limited availability.	Institutional capacity building and academic training courses provided at the Georgian universities can help to overcome these limitations.
Field data for calibration of satellite images (biomass volumes, classified land cover types, soil types, land management types) with exact information on the spatial location are rare and costly to be created.	Such data and information should be organised and gathered on national level across different sectors (agriculture, forestry, spatial planing, nature conservation ...). This would help to reduce significantly the costs and remote sensing could be applied on much larger areas.

7.1 Methods/ sources of information

7.2 References to available publications

Title, author, year, ISBN:

Land Degradation Neutrality 25.10.2017

Available from where? Costs?

https://e-c-o.at/files/publications/downloads/D00813_ECO_policy_brief_LDN_Georgia_171025.pdf

7.3 Links to relevant online information

Title/ description:

Tools for satellite image analysis

URL:

http://step.esa.int/main/snap-2-0-out-now/

Title/ description:

UNCCD Good Practice Guidance on SDG Indicator 15.31. (Sims et al. 2017)

URL:

https://www.unccd.int/sites/default/files/relevant-links/2017-10/Good%20Practice%20Guidance_SDG%20Indicator%2015.3.1_Version%201.0.pdf

7.4 General comments

UNCCD Good Practice Guidance on SDG Indicator 15.31. (Sims et al. 2017) gives a detailed technical overview on methods and approaches to calculate LDN indicators by means of remote sensing data.