Pastoralization

Help pastoralist communities around the world preserve their livelihood, by providing them access to information and resources (e.g grass land, water, safe routes) to prepare and support their migration journey.

The objective of the project is to design a working prototype for a mobile application which provides pastoralists community with information on land monitoring: water availability, grassland, and assist communities in networking.

project presentation: https://drive.google.com/open?id=0B69dNePSUlZ6NUNic1o2el9aUkk

We use the results of the analysis from satellite image (MODIS) to provide the information about water resource and vegetation health of grassland to pastoralist communities and help them relocate to the best grassland with available water and high quality of grass (within 200km).

Index to show the health of grass is obtained from MOD13 (https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mod13a1). It is call NDVI (normalized difference vegetation index), which can be used to assess whether the target being observed contains live green vegetation or not. For more information about NDVI (https://en.wikipedia.org/wiki/Normalized_Difference_Vegetation_Index).

The index to identify water body is NDWI (Normalized Difference Water Index), which can be calculated from MOD09 reflectance data (http://modis.gsfc.nasa.gov/data/dataprod/mod09.php). NDWI offers a means to view water bodies and may be useful for assessing flooding impacts. For more information about NDWI (http://deltas.usgs.gov/fm/data/data_ndwi.aspx).

The information of land use type to make sure whether the relocated place is grassland or not is get from MOD12 Land cover type data (https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12c1).

With the help of above three products from MODIS, we use ArcGIS 10.3 software to create a GIS layer showing the distribution of water bodies on grassland, and the grass health condition around these water bodies. These results will upload and store in our database server every 16 days. For each time when MODIS update its products, we do the analysis and upload it to the server. Therefore, the pastoralist communities are informed the up-to-date situation about the grassland where they current live, and the available water and vegetation resources near them.

People from pastoralist communities may worry about how to read satellite image. But there is no need to worry about it! We will translate the numbers into the easily understand words. We will use the normal distribution pattern of each index, and give each health condition, or water body condition a threshold value. We set up 4 levels of health condition of grass: low, normal, good and great. We will give pastoralist a warning when the grass quality they current occupied is low, such as "Grass Quality: Low! WARNING: You'd better move to a new place as soon as possible since insufficient resource!" Therefore, pastoralist will easily understand their current grassland condition and decide to relocate or stay.

If pastoralist decide to relocate, after we read their current coordinate, we will provide them the all the best relocated places with available water and high quality of grass within 200km. We could guarantee that there are enough water and grass resources in the relocated places. Among these relocated places with water resources, we also offer them the rank order of grass health from best to worst to help them choose which place they want to move to. Once the pastoralist choose the relocated place, we add the google api to offer them the safely migration route.

All in all, our app have the benefits in 4 ways:

1. Safe: we guarantee the water and grass resources in the relocated place;
2. Efficient: we provide the best relocated place in a minimum range;
3. Knowledge: we help pastoralist understand grass quality at their current place, and warn them be ready to move when grass health is declining below normal grass quality level;
4. Uptodate: We update the database every 16 days.

Calculated data from MODIS: https://drive.google.com/folderview?id=0B69dNePSUlZ6TkduXzViUWV5WHc&usp=sharing (we only use the data in the part of Australia, including Victoria state from 2008 to 2009 as an example)

solution code: https://github.com/pastrolmate/PastoralMate

Background data:

MOD09 Surface Reflectance: http://modis.gsfc.nasa.gov/data/dataprod/mod09.php

MOD13 Vegetation Indices: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mod13a1

MOD12 Land Cover Type: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mcd12c1