The human footprint on water : agricultural, industrial, and urban impacts on the quality of available water globally and in the Andean region


Published on

Presentation at Aguas2009, November 2009, Cali, Colombia. Mark Mulligan.

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

The human footprint on water : agricultural, industrial, and urban impacts on the quality of available water globally and in the Andean region

  1. 1. The human footprint on water : agricultural, industrial, and urban impacts on the quality of available water globally and in the Andean region Mark Mulligan, King’s College London, UNEP-WCMC [30 mins]
  2. 2. The issue • Land use and cover change affects hydrological processes and thus downstream users of water • With increasing populations and human appropriation of land careful management of these impacts is necessary • PES schemes are one mechanism by which downstream beneficiaries can pay upstream land managers for the hydrological services provided • The hydrological services considered are generally water quantity, flow regulation and water quality • Whilst there are reasonable data and spatial models for water quantity and flow regulation (in part because of the availability of remotely sensed data... • Spatial water quality data and models are much less developed
  3. 3. Rules of thumb for the water service benefits of ‘protected’ areas Water quantity services •Protected ecosystems do not necessarily generate more rainfall than agricultural land uses. •Protected ecosystems may have higher evapotranspiration and thus lower water yields Thus quantity benefits difficult to prove Water regulation services •Protected ecosystems do not protect against the most destructive floods •For ‘normal’ events they do encourage more subsurface flow and thus more seasonally regular flow regimes Likely benefits especially in highly seasonal environments Water quality services (quantity for a purpose) •Protected ecosystems encourage infiltration leading to lower soil erosion and sedimentation •Unprotected land will tend to have higher inputs of pesticides, herbicides, fertilisers ... Clear benefits of PA’s: generation of higher quality water than non- protected areas
  4. 4. Quality determines quantity 1. Water quality = water availability (for a purpose) 2. Quantity and access can be high but if quality is not sufficient then water scarcity can still exist 3. Countries like Colombia have a lot of water but to what extent is it all usable without expensive water treatment?
  5. 5. Q. How can we understand the impact of human activity on water quality when water quality cannot be assessed from remote sensing? A1. identify the proportion of your water originating in upstream protected areas A2. Identify the point and non point sources and calculate the ‘upstream human influence’ on river water
  6. 6. Protected areas : nature’s water filter •12-14% of the terrestrial surface is nominally protected •34% of ice free areas are used for agriculture and grazing •The rest is ice, desert,urban or unprotected wilderness •Mean management budget: $8.75 per km2* 1872 *estimated on the basis of James, A.N., Green, M.J.B. and Paine, J.R. 1999. A Global Review of Protected Area Budgets and Staffing. WCMC – World Conservation Press, Cambridge, UK. vi + 46pp
  7. 7. Quantifying the hydrological value of global protected areas Protected areas may help with water quantity and regulation functions and certainly help with water quality functions. Protected areas provide a ‘purification function’ on the basis that they tend to have lower human influence on water. Assumption: Water draining from a protected area is better (higher quality, better regulated) than water that drains from non-protected areas Method 1. Combine global rainfall dataset (1km resolution) with global dataset of flow directions (Hydro1k, HydroSheds) 2. Route rainfall down the flow network 3. For each pixel downstream calculate the proportion of runoff in that pixel derived from protected areas upstream 4. Combine with population and urban areas datasets (CIESIN) and calculate the number of persons benefitting from runoff originating in protected areas 5. Put online in Google maps/Earth : see 6. Repeat for other ‘contributing’ areas (forests, mountain forests, non-protected but non- agricultural areas etc.)
  8. 8. % of water originating in a protected area – WDPA 2009 (Colombia) [gl_pc_wc_fin] Protected areas provide a ‘purification function’ on the basis that they tend to have lower human influence on water. As you travel downstream from the protected areas their contribution to flow diminishes as rivers are swamped with water from non-protected areas see
  9. 9. Modelling the human footprint on water 0.1*Pr 1.0*Pm P non-human-influenced Frac*Pa Frac*Pg 1.0*Pog 1.0*Pu P non-human-influenced Frac*Pc Human Footprint on Water=∑Ppolluting /∑Ptotal
  10. 10. Roads Mines Point sources Urban areas Oil and gas
  11. 11. Pasture Cropland Non-point sources + Protected areas Unprotected agriculture - =
  12. 12. % of water that is human impacted At the global scale dominated by the human agricultural footprint
  13. 13. % of water that is human impacted At the continental scale the influence of roads and protected areas becomes more obvious
  14. 14. % of water that is human impacted At the national scale the downstream decay of influence away from agricultural and urban areas is clearer. We might expect the human influence to be reflected in higher sediment loads, organic and inorganic contaminants, incl. pesticides and fertiliser etc. This decay results from the dilution of human influenced water with runoff from less influenced areas.
  15. 15. % of water that is human impacted At the regional scale the distance decay is clear with some rivers still being 25% influenced by upstream polluting activities some 150 km downstream. The extent of the influence depends on the area of the polluting activity (though in reality the intensity of pollution will also be important. Protected areas ‘purify’ through dilution Oil wells may have a locally intense influence but it is soon diluted and fades quickly Takes no account of exposure levels : a measure of influence not of toxicity
  16. 16. % of water that is human impacted (transparent=negligible influence) The human footprint on water is concentrated around human populations (roads, agric, industry, urban tend to coalesce). Thus around major cities water quality could be heavily influenced Strategic positioning of protected areas can have positive impacts on the water supply of urban areas
  17. 17. % of water supply to urban areas that is human impacted At the local scale supply to urban areas is influenced by water diversions, aqueducts, transfers etc. for which there are no data globally. But, if we assume that rivers running into urban areas supply those areas with water then we can map the heavily impacted urbanisations Many of these cities will use intensive water treatment to offset these impacts
  18. 18. % of water supply to urban areas that is human impacted Although Colombia has a lot of water it also has high population, large urbanisation and intensive agriculture in the Andes. Colombia’s urban water supplies are thus heavily influenced by upstream human activity This necessitates costly diversion schemes or water treatment. Putting a natural PA buffer between populations and the mess they create can deliver clear water quality benefits at low cost.
  19. 19. Thank you Questions?
  20. 20. % of water supply to urban areas that is human impacted
  21. 21. Methods 1. Map global distribution of threat factors (1km spatial resolution GIS database) for point and non point sources 2. Integrate global maps of water inputs i.e. Rainfall (from WorldClim and TRMM) 3. Integrate global maps of flow directions (Hydro1k, Hydrosheds) 4. Calculate influence of each upstream point and non-point source as : Area of point source polluting activities+ = (mines + oilandgas + roads++*0.1 + urban) Area of unprotected agricultural land+ = (pasture + cropland) * (1-protected) Total polluting area per pixel (P) = max (1.0, point source polluting activities, unprotected agricultural land) These were summed downstream along the flow network top give Pd The rainfall (Rf) falling on all areas was also calculated and summed downstream to give Rfd. Human influence (%) = (Pd/Rfd) * 100 (the percentage of flow in a given pixel that fell as rain on an upstream human influenced area: a measure of the potential upstream influences on water quality). +Mines= mine according to Hearn et al. (2003), Oilandgas = oil and gas field according to Hearn et al. (2003) (binary), Roads = roads (binary), Urban = urban area according to CIESIN et al. (2004) (binary), Pasture = pasture land according to Ramankutty et al. (2008), Cropland = cropland according to Ramankutty et al. (2008), Protected= nationally or internationally protected areas according to WDPA (2009) (binary). ++ roads , if present, are assumed to occupy 10% of the pixel area