Documentation and Help

About the UNDPPA

The United Nations Department of Political and Peacebuilding Affairs plays a central role in United Nations efforts to prevent deadly conflict and build sustainable peace around the world. The DPPA monitors and assesses global political developments with an eye to detecting potential crises and devising effective responses. The department provides support to the Secretary-General and his envoys in their peace initiatives, as well as to the UN political missions around the world.

About the UNDPPA Innovation Cell

In January 2020, the DPPA launched the Innovation Cell, an interdisciplinary team dedicated to helping the Department and its field presences to understand and explore, pilot, and scale new technologies, tools, and practices in conflict prevention, mediation, and peacebuilding. Responding to the Secretary-General’s call on the UN system to accelerate its uptake of innovative methods, the Innovation Cell catalyses innovation in peace and security, while providing a forum for colleagues at UNHQ and in the field to engage collaboratively in human-centered design and problem solving. In response to the Secretary-General’s mandate to strengthen environmental and humanitarian security, the UN DPPA Innovation Cell established a Geospatial Portfolio to create tools for self-reliance and resilience building. The geospatial portfolio uses data-driven, non-invasive methods to address conflict and insecurity in areas where ground truth and in-situ research is difficult to obtain or conduct. Some examples of non-invasive methods include the use of satellite imagery and augmented or virtual reality tools.

About the Iraq Water Security Board

The United Nations Assistance Mission for Iraq (UNAMI) has engaged with the DPPA’s Innovation Cell to create the Iraq Water Security board for use by desk officers, in-situ practitioners, and decision makers. The Iraq Water Security board leverages environmentally-purposed remote sensing data with the mission to predict and prevent conflict and to enable self-reliant peacebuilding and peacekeeping.

Methodology and FAQs

This section of the Documentation and Help Page describes our methodology behind creating the Iraq Water Security geoboard, including the selection of indicators and datasets, dataset processing, and various other Frequently Asked Questions derived from user testing sessions. Preceding the start of development, a literature review was conducted by the Innovation Cell’s research team. The research team found and analyzed environmentally-purposed satellite datasets in order to derive the best combination of indicators and supporting datasets, given the intended purpose of the Iraq Water Security board.

Indicators and Data Sets

What are our indicators and datasets and why were they selected?

Please review the table below:

Indicator	Dataset	Temporal Availability	Temporal Resolution	Purpose
Total Water Change	GRACE (The Gravity Recovery and Climate Experiment)	2002-Present	Monthly	Measures changes in the Earth’s Mass; derived total water mass
Total Water Change	Project relevance: Measuring water abundance or absence over time provides an important baseline for overall water within a region.
Land Water Change	GRACE	2002-Present	Monthly	Measures changes in land water levels
Land Water Change	Project relevance: Land water change can indicate natural and human activities on Earth.
Ground Water Change	GRACE	2002-Present	Monthly	Measures changes in groundwater levels
Ground Water Change	Project relevance: Measuring groundwater abundance or absence over time provides an important baseline for overall water within a region.
Precipitation	CMORPH (Gravity Recovery and Climate Experiment)	1999-Present	Daily	Measures fluctuations in rainfall and potential drought or flooding
Precipitation	Project relevance: Seasonal and long term changes in precipitation influence potable water availability and agricultural viability.
Soil Moisture	SMAP (Soil Moisture Active Passive)	2015-Present	Daily	Measures soil water levels and fertility
Soil Moisture	Project relevance: Soil moisture estimates are important for monitoring droughts, predicting floods, and assisting agriculture.
Surface Water Extent	GSWE	1999-2020	Annual	Measures changes in total surface water
Surface Water Extent	Project relevance: This dataset measures the dynamics of terrestrial surface-water features such as ponds, rivers, and wetlands, which is valuable for the purposes of land use, water management, and ecosystems and services.
Vegetation	NOAA CDR & NDVI (Normalized Difference Vegetation Index)	1981-Present (updated every 10 days)	Daily	Measures vegetation change and classifies pixels by crop-type
Vegetation	Project relevance: Land cover change to support agriculture increases water demand. The degree of this demand is determined by irrigation versus precipitation dependency during cultivation. Although agriculture may support development and therefore contribute to stability, it may conversely increase water insecurity.
Project relevance: Changing population density and distribution impact and are impacted by water availability.

How did we acquire our datasets?

Our project leverages environmentally proposed remote sensing data from over 20 satellites. This open source data is collected from international satellite missions ranging from Landsat to Sentinel. Data acquisition was led by our commercial geospatial partners at Element 84.

How do we determine demand versus supply indicators?

Factors that contribute to water availability are classified as supply indicators, while those that strain water quality or quantity are classified as indicators of water demand.

Do we classify land used for agriculture as a demand or supply indicator?

Because agriculture is a water-intensive sector, it is classified as a water demand indicator. Out of the range of land cover classifications pertaining to vegetation, we selected a classification that most closely approximates crop presence. This was chosen because land use for agriculture relative to land use for grasslands or shrubbery presents a greater water demand.

How should a user interpret the Vegetation indicator?

The NDVI classifies land into 24 land-cover types. Central africa has three predominant land cover types today: irrigated cropland, rainfed cropland, and arid desert.

Where do we source our NDVI data from and what specific corrections are applied to it?

We source our NDVI data from https://www.ncei.noaa.gov/pub/data/sds/cdr/CDRs/Normalized%20Difference%20Vegetation%20Index/AlgorithmDescription_01B-20b.pdf. The corrections applied to the NOAA CDR NDVI product are: cloud masking; geometry-based shadow masking; atmospheric correction, which uses water vapor, digital elevation model, ozon, stratospheric aerosol climatology, and tropospheric aerosol corrections; and BRDF correction.

Data Processing, Calculations, and Standards
How do we process the data?
Download data from open source satellites
Normalize data to produce monthly values for each indicator
Generate a historical average for each month for each region
Calculate variance values for each month for each region based on a raw value’s variance from the historical average
Aggregate all indicator values to master geodatabase
Why do we aggregate all values by the month and what is data normalization?
Normalization is the process of creating a common denominator from which different datasets can be compared. Each dataset has a unique temporal range, temporal resolution, and spatial resolution; i.e. because some data is collected by satellites daily, while others annually, we average all values across the common denominator of a month. This creates a datapoint for every month-year combination for each dataset so that indicators may be cross-compared more effectively.
What is historical average and how is it calculated?
Water availability fluctuates throughout the year, according to natural seasonal cycles. In order to understand when water is abnormally low or high, we calculated historical average values for each month in each region. For example, all of the Julys in the Ninawa governorate can be compared to each other to identify years of particular water abundance or scarcity. The historical average provides a baseline approximation for what is considered normal or expected for a given month.
What is historical variance and how is it calculated?
Each raw datapoint is compared to the historical average value for that month and region. The resulting value is the historical variance and is a percent ranging from -1 to 1, indicating decreases or increases from the expected baseline value for that month. Data on the dashboard is presented in historical variance format rather than in raw value format to generate a standard axis on which indicators measured in different units can be cross-compared and easily understood. If we did not do this, the specific units of each indicator would be difficult to compare, e.g. cubic centimeters to human population counts. The historical variance calculation solves this.
What is “% Variance” and how is it calculated?
Percent Variance is how we display variances from historical averages. This allows the viewer to efficiently analyze the significance of fluctuations over time. This scalar model reflects the magnitude of change and provides a standard baseline of analysis in which all indicators may be graphed simultaneously. This allows the end-user to compare indicators against each other, despite their differing units and resolutions. Additionally, it presents the data in a more digestible way for end-users who are not experts of each indicator. For example, a 7mm increase in precipitation per year may hold little meaning to users who are policy makers, not hydrologists.
What kind of data standards are applied?
Prior to selecting datasets, a literature review is conducted by our research partners at Stanford and the UN. We vigorously analyze data sets and their integrity through an extensive process to validate its usefulness, given our intended purpose. There are a few standard corrections applied to satellite datasets such as cloud coverage and BRDF correction. These are proven scientific models that identify or correct for pesky cloud coverage and light reflectance that obscure satellite sensors and imagery.
Who is the map provider?
The map used in the Iraq Water Security board is generated by Mapbox using their basemap and the GADM political boundaries. Because the official UN map has not yet been approved by the Iraq delegation, we cannot incorporate it.

Contact