December 13, 2013

Telangana river waters, irrigation & agriculture-12 (Basins- a primer)

A quick recapitulation

In the last several chapters, we tried to decipher water rights especially as applied to trans-boundary water sharing. We looked at the elements of both the Indian water regime and international law.

However, we are still not close to determining what Telangana's "fair share" of river waters, especially in the Krishna basin, should have been. Indian law leaves the job of determining the "fair share" to tribunals while this role is assigned to the Supreme Court in the US. The methodology of doing so is to be decided by the tribunal/court.

CWDT (volume III, page 6) summarizes the position: "No doubt, the principle in respect of equitable apportionment of the water is settled, but what shall be the equitable apportionment in respect of different riparian States so far the water of such inter-State river is concerned is itself a big question". GWDT (page 123) concurs: "There is no rigid formula for the equitable apportionment of waters of a river. Each river system has its own peculiarities".

Before embarking on a study of agreements, tribunal reports and case law, it is useful to look at the hydrological factors. This will help us understand these technical aspects better. This work therefore takes a "hydrology detour" before returning to water law.

River basin

Most people understand (or believe they understand) rivers quite well. However even this is not without controversy. In Kansas v. Colorado, Colorado claimed the Arkansas is in fact two rivers and that its irrigators were confined to the "Colorado Arkansas". Justice Brewer rejected the contention and held it to be one river, even if it was "broken".

The nature of river systems (i.e. a river together with all its components including tributaries) is more difficult but still within the perception of "lay men". Understanding what constitutes a basin (sometimes called river basin, river valley, drainage basin or catchment area) is more complex and may require the assistance of experts.

Hydrologists are unanimous in their agreement that a basin is an indivisible entity. Every basin is separated from the neighboring basins. Every drop of stream runoff whether resulting from snowmelt or rainfall ends up in the appropriate river system. Every square meter of land is necessarily a part of an appropriate basin and only that one basin.

Basins, like contour lines, do not cross each other. Basin boundaries are much more rigid than "real world" features like rivers or "conceptual lines" like seashore. The fact that basins are not yet accurately mapped can not change or distort this scientific position.

Basins are often divided into sub-basins. For instance, Krishna sub-basins are labeled K-1 through K-12. These divisions do have a physical connotation (and occasionally legal implications too) although the primary purpose is convenience.

KWDT (volume I, page 98) goes into the nature of a basin at great detail. KWDT cites WG Moore's Dictionary of Geography: "The entire area drained by the river and its tributaries is called the river basin". KWDT (on the same page) refers to a textbook on Applied Hydrology: "The river basin is necessarily completely bounded by the watershed or divide which separates it from other adjacent basins".

The tribunal accordingly ruled: "The expressions "Krishna basin", "Krishna river basin" and "Krishna drainage basin" used in this Report mean the entire area drained by the Krishna river and its tributaries. The Krishna basin is bounded by the watershed or divide which separates if from other adjacent basins".

This point is elaborated and reiterated as follows (in part relying on HA Smith's The Economic uses of International Rivers):

"River basin an indivisible physical unit. Each river basin is an indivisible physical unit, a more or less self-contained unit of drainage. Nature's laws treat the river and its tributaries as the arteries of a single circulatory system. The surface streams converge, ever seeking a lower level and unite to form one mainstream. All the waters that find their way towards a common outlet form an interconnected and interdependent system, capable of transmitting within itself any disturbance caused by changes affecting water in any part of the basin. Water is a moving resource which implies that changes in quality or quantity of water in one place may directly affect uses of water somewhere else".

Basin vs. boundaries

Political boundaries are defined based on several parameters. These often include social, cultural, economic factors etc. but almost never consider hydrological factors.

Cartographers often use features like rivers or hills when they draw maps. This stems partly from the fact such features are obstacles to transport & movement. This is why rivers & streams often form boundaries between districts, states or countries. The cartographic concept of a "natural boundary" makes it much more likely that basins crisscross political divisions.

The Supreme Court in the 1991 Cauvery presidential reference stated "Though the waters of an inter-State river pass through the territories of the riparian States such waters cannot be said to be located in any one State".

KWDT observes (volume I, page 99): "Division of an inter-State river by the boundaries of several States merely limits the geographic limits of the authority of a given State; but unlike land resources whose distribution among the States is resolved by the very establishment of their boundaries, the water resources of the common river are not subjected to automatic allocation among them by the delineation of their political frontiers".

Basin vs. command area

Because waters within a state are effectively at its command, states distribute these based on their own requirements. This right is recognized by all authorities. For example, KWDT final order Clause XV reads "Nothing in the Order of this Tribunal shall impair the right or power or authority of any State to regulate within its boundaries the use of water or to enjoy the benefit of waters within that State in a manner not inconsistent with the Order of this Tribunal".

States do not limit water use by basins. They base their planning on the concept of command area (sometimes also called "culturable command area") of a project. MoWR defines this as "the area which can be irrigated from a scheme and is fit for cultivation".

The differences between the two concepts is summarized below:

·         Command area is an administrative concept that may have no hydrological or topographic connotation
·         Command area relates to a project while basin relates to a river system
·         Every square meter of land is a part of one basin or another. This is not true of command areas: due to several reasons including government policies, a piece of land may not be a part of any project's command area
·         A command area can crisscross two (or even more) basins if the state resorts to trans-basin diversion

While the scientific definition of a basin is not questioned by any authority, there is some occasional confusion among administrators. AP quoting a couple of cases argued before KWDT that a basin includes "all territories outside the river drainage basin to which the waters of the river may be diverted and beneficially applied". In other words, AP tried to club the basin & command area concepts. KWDT dismissed this "artificial definition" and proceeded to present the correct situation detailed earlier. Trans-basin diversion was held to be permissible "but those areas cannot be regarded as parts of the river basin".

AP's "confusion" appears to persist. Before the second Krishna Water Disputes Tribunal (KWDT-II), AP made (pages 796-800) several contentions that indicate this:

·         A list of fluoride affected inhabitants cutting across basin borders
·         Out of the seven proposed new schemes, three (Srisailam Left Bank Canal, Kalwakurthy & Nettempaddu lift irrigation schemes) totaling 77 TMC are within the Krishna basin. The other four projects totaling 150 TMC are trans-basin diversions
·         When requested for details of drought prone areas, AP contended 67,650 sq. km including areas outside the basin. This is in sharp contrast to Maharashtra & Karnataka who listed only the drought prone areas within the Krishna basin (each state around 50,000 sq. km). Karnataka disputed AP's claim and submitted an estimate of 45,493 sq. km based on information obtained from AP. This led KWDT-II to record "The highest drought prone area in Krishna basin is in the State of Karnataka whereas State of Andhra Pradesh has smallest drought prone area in Krishna basin".

AP claimed its contention was based on KWDT statement (volume II, page 8) "the relevant consideration is the interest of the State as a whole and all its inhabitants and not merely the interest of the basin areas of the State". This is untenable as the context was to the injury caused to a state by the action of its riparian neighbors: "the crucial question is whether the interest of the State or of any of its inhabitants in the waters of the inter-State river and river valley is prejudicially affected by the action of another State".

A reading of ISDA sections 3 & 3a will serve to confirm the above:

"3. Complaints by State Governments as to water disputes. appears to the Government of any State that a water dispute with the Government of another State has arisen or is likely to arise by reason of the fact that the interests of the State, or of any of the inhabitants thereof, in the waters of an inter-State river or river valley have been, or are likely to be, affected prejudicially by--

(a) any executive action or legislation taken or passed, or proposed to be taken or passed, by the other State; or"

It may be noted that AP did not list the entire state drought prone area 89,109 sq. km in its submission. This does not tally with the purported understanding of "state as a whole".

From the above, it emerges AP continues to treat non-basin areas benefitting from trans-basin diversions as a part of the basin. As this contention flies in the face of both accepted science and practice as established by tribunals, it may be concluded that the posturing reflects the state policy. This approach is likely to have impacted the data provided by the state too.

Basins of Andhra Pradesh

SKC page 219 shows a basin map of AP together with district boundaries. While this map's nature (raster) and scale is not amenable to spatial database creation or analysis, it does provide some interesting information based on a visual examination:

·         AP consists of 40 basins including 12 inter-state river systems
·         Khammam is spread in four basins: Godavari (# 20), Thammileru (# 22), Budameru (# 24) & Krishna (# 25)
·         With the exception of the small area in Khammam covered by Thammileru & Budameru basins, the rest of Telangana is fully in either Krishna or Godavari basins
·         Adilabad & Nizamabad districts are totally in the Godavari basin
·         Mahabubnagar & Nalgonda districts are totally in the Krishna basin
·         All other districts crisscross a minimum of two basins
·         Contrary to intuitive perceptions, both Andhra and Rayalaseema are largely outside the Krishna & Godavari basins

The map does have some shortfalls. For instance, it does not delineate sub-basins. It also ignores AP's boundary with Pondicherry (Yanam) and treats Yanam as a part of East Godavari district. This should not detract from the fact the map is otherwise useful.

Andhra Pradesh basin areas

The publication Water Resources Statistical Abstract-2010 (WRSA-2010 or WRSA) published by AP's Irrigation & Command Area Development (I&CAD) department lists (pages 25-26) lists each of the 40 basins with the "catchment" area in the state. The total works out to 262,277 sq. km i.e. 12,768 sq. km less than the state's area!

The same publication (pages 21-22) 58 rivers flowing in the state together with their "catchment" area in the state. The total works out to 318,208 sq. km i.e. 43,163 sq. km more than the state's area!

While the reasons for these discrepancies is not readily ascertainable, this is definitely a serious lapse not to be expected in a work of this nature. It may be stressed here that WRSA appears to be the most important source cited by various contenders including LSP.

In any case, WRSA does not provide region (or district) wise breakup of the river basins. This information is available from KWDT & GWDT reports. The reports provide basin area (in square miles) for each of the basin districts as well as the proportion of the basin area in the district to the entire district.

Converting at the rate of 1 mile= 1.61 km, KWDT's estimates show that AP covers 76,313 sq. km of the Krishna basin. Telangana accounts for 52,274 sq., km i.e. 68.5% of the Krishna basin in the state. Similarly, GWDT estimates show Telangana at 58,006 i.e. 79.2% of AP's 73,261 sq. km. These percentages are widely quoted and so familiar to most people on both sides of the debate.

However, a rather different picture emerges when one tries to validate the tribunal data. Bellary district is shown to be entirely within the Krishna district with an area of 9,915 sq. km whereas the district's current area is only 8,450 sq. km. Khammam district's area after prorating works out to around 11,925 sq. km against the current area of 16,029 sq., km i.e. around 4,100 sq. km short. The area of East Godavari, in contrast, is estimated at 14,967 sq., km, around 4,160 sq. km more than its present area.

On a careful examination of the discrepancies, I am led to believe the tribunal data is based on a pre-1956 estimate. The East Godavari & Khammam discrepancies are almost exact & opposite: this must be the area transferred after 1956 to Khammam. The excess estimate of Bellary district reflects the area transferred to AP in 1956. The other discrepancies are almost certainly traceable to reorganization & other territorial transfers.

Andhra Pradesh basin distribution

As most of Telangana (with the exception of 561 sq. km in two minor basins) is in Krishna & Godavari basins, I classified the AP basins in three categories: Krishna, Godavari & "Others" (i.e. all the 38 basins lumped together).

Srikakulam, Vizianagaram, Prakasam, Nellore, Chittoor & Cuddapah pose no problems as these districts entirely fall in the other basins. Nizamabad & Adilabad are also easy as these districts are entirely within the Godavari basin. Mahabubnagar & Nalgonda are similarly totally in the Krishna basin.

The districts of Visakhapatnam & West Godavari basins are also easy: I assume the GWDT estimate to be correct while the rest of the districts fall in the other basins. The districts of Krishna & Guntur were treated in a similar manner assuming KWDT estimates to be accurate.

The case of the former Hyderabad district was not too difficult either: I assumed the present Hyderabad district to be totally in the Krishna basin and assigned the insignificant Godavari basin area to Ranga Reddy.

I resolved the East Godavari discrepancy by deducting the excess GWDT estimate from its Godavari basin and allocating the balance to other basins. I similarly resolved the Khammam discrepancy by assuming the Krishna & other basin areas to be correct and assigning the balance to Godavari basin.

In the case of Medak, I assumed KWDT estimate to be accurate assigning the balance to Godavari basin. For Warangal, I took the opposite route assuming GWDT estimate to be correct.

The only problem remaining was between Anantapur & Kurnool districts. Bellary district ceded land to both in 1956 with no information on the exact extent. I assigned the Krishna basin area transferred from Bellary to Anantapur. While this is not factually correct, I had no other option as no information is available on these territories. This error should not reflect too much on the study as both Anantapur & Kurnool belong to the same region.

The end result is as follows:

·         Krishna basin: 77,721 sq. km with Telangana at 67.7% (52,587 sq. km)
·         Godavari basin: 72,787 sq. km with Telangana at 84.8% (61,692 sq. km)
·         Other basins: 124,537 sq. km

"Rivers are not human artifacts; they are natural phenomena, integral components of ecological systems, and inextricable parts of the cultural, social, economic and spiritual lives of the communities concerned. They are not pipelines to be cut, turned around, welded and rejoined": R. Ramaswamy Iyer


  1. So T-guys stole 4000 km from TooGoJi :)

    1. The area was transferred legally by the state Govt.

  2. This comment has been removed by a blog administrator.

  3. Dear Jai,
    After reading your blog content, I am impressed with your in depth understanding on water issues/disputes. I want to take legal advise on this issue. Would you call me on my Mobile xxxx to meet you personally.

    1. This is the comment posted by ns above reposted it after eliminating personal info

  4. What are the other basins in AP?

    1. There are totally 40 basins in AP. Pennar (~ 47,000 sq. km in AP) is the biggest after Krishna & Godavari. Gundlakamma, is the fourth largest with over 8,000 sq, km in the state.

      There are 21 other basins with a 1,000+ catchment in the state.

      If you want a full list, please see WRSA 2010. Alternately I can mail you the list if you want. Please note the WRSA numbers have a serious flaw (totals not adding upto state area).

  5. Can you please give ap catchment area by basin, thanks

    1. I have provided the info at the bottom of the post.

      Krishna: 77,721 sq. km (28.3% of state area)
      Godavari: 72,787 sq. km (26.5% of state area)
      Others: 124,537 sq. km (45,3% of state area)

      Region wise breakup (Telangana-Andhra-Seema in %):

      Krishna: 67.7-11.5-20.8
      Godavari: 84.8-15.2-0.0
      Others: 0.5-58.5-41.0


Please be brief. Please respect everyone's privacy and do not reveal any private information about yourself or others.

Suggestions on improving the quality of this blog are always welcome. All other comments should be relevant to the subject of the post. I will delete all spam and messages with abusive or vulgar language.

All material in my blog is original. I will remove any copyrighted material if notified.

You may not use the material from my blog without my permission. I will not refuse any reasonable request as long as you credit me and provide a link to my own post.

If you post rejoinders, rebuttals or supplementary posts in your own blog, please leave a comment with a link.