The 'Ins and Outs' of the Millennium Ecosystem Assessment for headwater wetlands

The Millennium Ecosystem Assessment was published in 2005. That report is a very major assessment of man's impact upon the environment. Four years were invested into its preparation. More than 1300 authors from 95 countries were involved. The MEA had a preparation budget of around 24 million US dollars. Globally significant institutions and some serious names are associated with the final report. With such an investment and provenance, the MEA has frequently been cited as an influential and authoritative source since.

Articles so far: In my previous articles on LinkedIn, I have highlighted that there are essentially two science outlooks on headwater wetlands - a hegemonic model and an alternate. The hegemonic model takes the perspective that headwater wetlands are characterised by substantial storage. Consequently, they are portrayed as acting as sponges, storing rainfall and upstream inflows, bringing benefits of flood attenuation to downstream communities. Further, that they release water to baseflow at critical times, thereby extending flow downstream at critical periods. By this dual role of flood reduction and baseflow sustenance in combination, they are assigned a critical role upon water resources of regulation. A further value is given to their promotion of groundwater recharge. These services and functions are portrayed as being performed at levels much more significant than any regulatory role performed by any other portion of a catchment.

The alternate model presented in previous articles is that headwater wetlands can be characterised by low storage. With soil moisture deficits being rapidly satisfied, the headwater wetlands are - preferentially within the catchment - significant generators of rapid response runoff when saturated. Rapid response runoff does not mean necessarily causing floods of any damaging or devastating proportions. Especially in the tropics and sub-tropics, when potential evaporation exceeds rainfall (as is common for extended periods, up to months), depletion of stored wetland water is mostly by evapotranspiration rather than by release to streamflow. In regions where groundwater is a significant contributor of baseflow, evapotranspiration at wetland margins can actually reduce the amounts and durations of those groundwater baseflow contributions reaching rivers. By roles of rapid response runoff generation and baseflow reduction - individually and in tandem - wetlands are the converse of a regulator of downstream river flows.  As saturated and impermeable barriers between rainfall and groundwater they perform little role in recharging groundwater. Portions of catchments other than the headwater wetlands can be significantly more responsible for performing the vaunted regulatory functions - namely more permeable and unsaturated non-wetland surfaces in respect of buffering rapid response runoff and groundwater in respect of sustaining baseflow.

For the purposes of this discussion, let's call the former the hegemonic/regulatory model and the second the alternate/non-regulatory model.

What this article aims at:

This discussion is to track how those two divergent models have been represented by the Millennium Ecosystem Assessment, at its different stages. Let us start with the main MEA Synthesis report itself, before we then look at material that fed into the MEA. And let us look also at the RAMSAR Conventions products intended to carry forward the main findings from the MEA.

In doing so, let us bear in mind the weight of the scientific evidence. Among around 270 or so independent science conclusions. As discussed in my last article, around 25% of that evidence is supportive of the hegemonic/regulatory model. Around 25% is neutral - ie the wetlands are not significant in hydrological terms. Around 50% are supportive of the alternate/non-regulatory model. Among the evidence from science studies in Africa the proportions are more like 20:20:60. Let us recall also that the extent of African headwater wetlands is about the same as the area of 20 African countries. This is a major land surface type and is of major significant to Africa's water agenda. As per my earlier article on LinkedIn, the case has been made that those citing the hegemonic/regulatory model sometimes will often issue the caveat that not all wetlands perform in that particular way. But rarely give any recognition at all is given to the alternate/non-regulatory case.

The MEA Flagship: 'Ecosystems and Human Well-Being - Synthesis'. Summary for Decision Makers

The principal document of the Millennium Ecosystem Assessment (MEA) leans very heavily to the hegemonic/regulatory model.  The flagship of the MEA is the 'Ecosystems and Human Well-Being - Synthesis'.  In its Summary for Decision Makers, (page 7) wetlands are cited as natural buffers, contributing to natural hazard regulation. On page 40, wetlands in their natural state are closely associated with water regulation. Page 47 cites the storage capacity of the mid-reach Mississippi floodplains and Page 56 also makes the case of flood control by a coastal peat bog in Sri Lanka - both wetland types being distinctly different in catchment location from headwater wetlands. Page 47 makes the case that "protection of wetlands can contribute to flood control". Page 56 discusses in an economic context "the water regulation benefits of wetlands". On page 61, causes of poverty are linked to the loss of ecosystems such as wetlands that "affect the likelihood of flood or storm damage". Commencing on page 117, the section on 'Natural Hazard Regulation - regulating services' points heavily to wetlands ...

               " Ecosystems play important roles in modulating the effects of extreme events on human systems. Ecosystems affect both the probability and severity of events, and they modulate the effects of extreme events. Soils store large amounts of water, facilitate transfer of surface water to groundwater, and prevent or reduce flooding. Barrier beaches, wetlands, and lakes attenuate floods by absorbing runoff peaks and storm surges."

Such is the foundation of the MEA's understanding of hydrological services that the case is made for extension of wetlands into the realm of payment for environmental services 'based on the entire flow regime' (page 128).

So, the synthesis report of the MEA - for Decision-Makers - uniquely refers to the hegemonic/regulatory model, founded on a base of high storage, flood reduction, and flow regulation. No caveats and not even a mention of any alternate/non-regulatory model.  

Assessment 'Chapters' feeding the Synthesis

That MEA Synthesis, published in 2005, is itself drawn from the more substantive Assessment Reports - over 80 in number. So, let us track the origins of the representation of that model from the MEA feeder reports.

One of the feeding Assessment Reports is the so-called Chapter 20 on Inland Water Systems. Page 555 estimates the economic value of flood control/flow regulation services, assigning a value of 464 USD per ha per year, substantially the largest of any material function cited. On hydrological regulation the following synopsis is presented, supported by individual examples:

               " It is well recognized that some inland waters serve as important storage sites, accumulating water during wet periods and providing a reserve of water during dry periods by maintaining base flow in adjacent rivers. Similarly, it is increasingly known that some inland waters, such as lakes and marshes, attenuate floods by retaining water or storing it in the soil and therefore reducing the need for engineered flood control infrastructure". ...

 

               The Section concludes that "Despite a lack of reliable quantitative evaluations, the importance of hydrological regulation by inland water systems is widely recognized around the world" referencing Mitsch and Gosselink 2000. That citation -laying down a marker that the regulation role is recognised around the world - is to Mitsch, W. J. and J.G. Gosselink, 2000: Wetlands. 3rd edn. John Wiley, New York, U.S.A. 920 pp.

That book itself cites a publication by one of the MEA Chapter 20 lead authors - 'Classification and Inventory of the World's Wetlands'. This is a passage from that publication (note that dambos are one type of headwater wetland, very extensive across Southern Africa - see the image associated with my first article. Dambos are probably similar to several other types of headwater wetlands across the continent, such as vleis, inland valley swamps, bolis and others):

               For dambos, Bullock (1992a) has examined critically their role in buffering river flows in 110 gauged catchments of six major hydrological zones. He concluded there was no evidence that dambos maintained dry season flows or moderated most flood events. ... In a related review of dambo function in southern Africa comparing data from Zimbabwe to other published data (Bullock 1992b), it was concluded that dambos often act to reduce dry season flows, as compared to similar catchments without headwater wetland. However, these views are controversial, for dambos in high altitude (eg Lesotho, see Schulze 1979) can exhibit buffering and low flow maintenance. p.74 - 75

[Note for information, the Bullock (1992b) paper had concluded, as per its abstract "Collated statements on the influence of dambos on river flow regimes from 25 reports in seven different countries illustrate an absence of consensus on the influence of dambos on flood flows, low flows and total catchment runoff. A case has been established for dambos to be considered to reduce low flows in certain regions of Africa, by combining results from regional flow regime analysis and regional evaporation data from Zimbabwe, and by re-assessing past studies of dambo hydrology in Malawi and Zambia.]

So, those drafting Chapter 20 of the MEA were very familiar with a substantive review of African evidence that had been published in 1992 in the peer review literature. 25 reports from seven countries that were not supportive of the hegemonic/regulatory model. And this source had been referenced in other publications among those drafting Chapter 20.

But such an assessment was deemed by them to be controversial. To the extent that no mention at all of different science findings was then made in Chapter 20. It was Chapter 20 which fed into the MEA synthesis report for decision makers. As we have seen, that presented a uniquely hegemonic/regulatory case, as its feeder Chapter 20 did. Further similar material also came from Chapter 7, the 'Fresh Water' feeder to the MEA:

               " Wetlands and floodplains act as natural sponges; they expand by absorbing excess water in time of heavy rain and they contract as they release water slowly throughout the dry season to maintain streamflow"

Several citations from 2003 and even from 2004 are made in the Chapter 20 and Chapter 7 feeders. But one paper from early 2003 is totally ignored. Bullock and Acreman (2003) - a global review of the science findings concerning the hydrology of headwater wetlands from 169 different publications. Ignored entirely. No influence at all upon the main MEA Synthesis Report to Decision Makers.

Others such as Professor Ed Maltby have also put the Mitsch and Gosselink and the Bullock and Acreman assessments together, but in doing so have drawn a very different formulation from the evidence than that which was carried into the MEA. Namely, in their 2011 publication  'Ecosystem Services of wetlands: pathfinder for a new paradigm'

               ‘Hydrology is probably the single most important determinant of the establishment and maintenance of specific types of wetlands and wetland processes’ (Mitsch and Gosselink 2000). Complex interrelationships with the hydrological cycle (Bullock and Acreman 2003) and essential dependency on water supply place wetlands centrally in some of the most contentious and urgent issues governing the appropriate management of water. Therefore, whilst wetland hydrology is a natural science, its relationship to the water cycle is highly relevant to societal concerns such as issues of water supply, resource allocation, water quality and flood risk."

Complex and contentious?, Absolutely, yes. Controversial so as to be excluded? No.

So, the MEA could undoubtedly have been more evidence-based. Which sounds a bizarre statement, given that four years were invested into its preparation and that more than 1300 authors from 95 countries were involved, with a preparation budget of around 24 million US dollars. I later asked one of the lead authors of the synthesis report about this, and had the kind reply that "There is one obvious gap in the MA analysis: Your paper should have been picked up and cited as a reference. The timing of its publication and relevance through a global analysis would have made it an obvious choice as an information resource. That is regrettable."

It is indeed regrettable. Because for very many people the science community - such as that convened for the MEA - is generally trusted for its treatment of evidence, of process and so of conclusion. The MEA has influenced many subsequent publications ... and policies based on its synthesis. Its provenance is not delved into - it is the summary for decision makers that gets the headlines.

So, it is verging on the chaotic - in terms of consistency and logic - what happens next.

RAMSAR's synthesis of MEA findings on wetlands

Alongside the main MEA Synthesis for Decision Makers - remember, uniquely promoting the hegemonic/regulatory model - a decision is taken that the RAMSAR Convention - the leading agency on wetlands - will prepare an accompanying synthesis, entitled "Ecosystems and Human Well-Being: Wetlands and Water ".

Its foreword reads " The Convention on Wetlands (Ramsar, I.R. Iran, 1971) has recognized from the start that the MA can and should provide the Contracting Parties to the Convention, and all involved in the conservation and wise use of wetlands, with new understanding and insights into how best they can meet the objectives of the Convention. The Convention’s Standing Committee, Secretariat, and Scientific and Technical Review Panel have supported and contributed to the work of the MA throughout. This report, synthesizing the findings of the MA on inland, coastal, and near-shore marine wetlands, is the key product of the MA for the Ramsar Convention. It draws on the work of approximately 1,360 experts who compiled the many chapters of the MA reports. The synthesis stresses the link between wetlands and water and will help us set the future agenda for Ramsar."

So, this publication - "Ecosystems and Human Well-Being: Wetlands and Water " - has the intention to synthesise the findings of the MA and help set the future agenda for Ramsar.

Page 37 showcases 'Hydrological Services of Wetlands' in its own Box (Box 3.2). This Box is especially illuminating, as the following four main points made on water quantity demonstrate:

               Flow regulation: Inland wetlands are important water storage sites during wet periods and often provide a reserve of water during dry periods. However, there is strong evidence that some wetlands with high evapotranspiration rates reduce the flow of water in downstream rivers during dry periods. This is backed by overwhelming evidence that shows evaporation from wetlands to be higher than from non-wetland portions of the catchment during dry periods; there is no discernible difference for different wetland sub-types. In only 20% of cases examined did wetlands increase river flows during the dry season.

               River flow and hydrological regime variability: These services vary greatly between different types of wetland and their localities. For instance, different headwater wetlands and peat bogs showed increases, decreases, or neutral effects on variability. While floodplains (such as the Okavango delta in Africa and the Barito floodplain in Indonesia) reduce flow variability, principally by reducing flood peaks, other wetlands (such as many headwater wetlands) serve to increase flow variability through both increasing flood peaks and reducing dry-season flows.

               Flood-related services: Floodplain wetlands almost always reduce floods (and their peaks) or delay them. Many wetlands in the headwaters of river systems (such as bogs and river margins) probably perform a similar service. (Note by AB - this flood reduction role of mid-reach floodplain wetlands has to be associated with the Box's statement that around 65% of studies reviewed concluded that non-riverine wetlands reduce average annual flow in rivers ie flood reduction is allied to total flow reduction) However, some headwater wetlands are known to increase flood peaks and generate flood flows. They often increase the immediate response of rivers to rainfall, due to a tendency to be saturated, generating higher volumes of flood flow even if the flood peak is not increased.

               Groundwater services: Many wetlands exist because they overlie impermeable soils or rocks and there is therefore little or no interaction with groundwater. However, numerous wetlands are groundwater-dependent and fed largely or wholly by groundwater, such as wetlands that form at springs, oases, and many lakes. In some instances, wetlands may promote less groundwater recharge than other land types.

So, evidence that had not been reflected at all in the main Millennium Ecosystem Assessment Synthesis for Decision Makers. Or its feeder Chapter 20. Wasn't this Ramsar book synthesising the findings of the MEA? Seemingly not. One could ask why this RAMSAR synthesis is so significantly different from the MEA - but far more interesting is why the fundamental findings on the hydrology of headwater wetlands made in this Book had not been reflected into the MEA. When there was so much cross-fertilization.

The balance of science evidence of headwater wetlands had been recognised and mainstreamed into this 2005 RAMSAR Convention's synthesis. Recognition of the alternate/non-regulatory model - in a big way, in a big international publication. Not just Africa, but Global. Same proportionality as the among the evidence given in the Bullock and Acreman 2003 global review. Set to influence the future agenda for RAMSAR.

RAMSAR Convention Fact Sheets on Ecosystem Services

So, then, the group of the RAMSAR Convention sets about issuing a set of Fact Sheets on Ecosystem Services,  ...  " This set of 10 Ramsar Factsheets profiles the ‘ecosystem services’ – the benefits people obtain from ecosystems – provided by wetlands." "Many publications and reports have been consulted in compiling these factsheets on Ecosystem Services." " Not all wetlands provide all of the services listed above all of the time. Different wetlands provide a range of services according to their type, size and location. Nevertheless, the highly respected and influential United Nations Millennium Ecosystem Assessment recognises the enormous global economic importance of wetlands, valued at up to US$15 trillion dollars in 1997."

Fact Sheet #1 is on Flood Control: In brief…

               "Slowing the flow – wetlands close to the headwaters of streams and rivers can slow down rainwater runoff and spring snowmelt so that it doesn’t run straight off the land into water courses. This can help prevent sudden, damaging floods downstream.

               "Nature’s flood storage reservoirs – the floodplains of major rivers act as natural storage reservoirs, enabling excess water to spread out over a wide area, which reduces its depth and speed. By draining floodplains and building on them, we have effectively squeezed the floodwater into narrower and narrower corridors, which means that flood peaks are deeper and floodwater travels faster.

               Main text: "Different types of wetlands play important flood control roles in different situations. In the upper reaches of some river basins, for example, peatlands and wet grassland can act like sponges (saturated peat is typically up to 98% water by mass), absorbing rainfall and allowing it to percolate more slowly into the soil, thereby reducing the speed and volume of runoff entering streams and rivers. This means that water levels in larger channels, further downstream, also rise more slowly and human lives and livelihoods are less likely to be affected by destructive flash flooding. When peat becomes completely saturated and unable to absorb any more water, surface pools and peatland vegetation – including sedge meadows and some types of forest – help to slow and reduce runoff."

So RAMSAR Fact Sheet #1 = Uniquely hegemonic/regulatory How about Fact Sheet #2 on Groundwater Replenishment?

In Brief:

               " Many wetlands are directly connected to groundwater and play a vital role in regulating the quantity and quality of groundwater, which is often an important source of water for drinking and irrigation of crops."

So, while there are some minor caveats in the main text, the main message of Fact Sheet #2  is uniquely hegemonic/regulatory.

In an overall summary describing these Fact Sheets "The factsheets illustrate the great diversity of ecosystem services delivered by wetlands and their values" Well I can't find a Fact Sheet on that in any way reflects the alternate/non-hegemonic case. Or the diversity - let alone, the balance among diversity - that was captured in the RAMSAR publication on "Ecosystems and Human Well-Being: Wetlands and Water". The RAMSAR publication that would take Millennium Ecosystem Assessment findings into RAMSAR's future agenda.

A chaotic passage of Ins and Outs

Indeed, a chaotic passage of 'ins and outs' through the Millennium Ecosystem Assessment. Which looks like this for the two models ...

               Alternate/non-regulatory model = previously labelled controversial among MEA contributors

               Hegemonic/regulatory model -> Chapter 20 feeding into the MEA

               Hegemonic/regulatory model -> MEA Synthesis Report for Decision Makers

               Alternate/non-regulatory model -> RAMSAR synthesis of main findings from MEA

               Hegemonic/regulatory model -> RAMSAR Fact Sheets

And so it all just goes on and around again ... and on into the latest World Water Development Report. And into the narrative of the upcoming Stockholm World Water Week.

In my next article I will look briefly at how some of the major institutions have evaluated their positions on this issue. Then later this month I'll start to bring out some of the implications. But they will be preliminary. For four years, hundreds of authors from among 95 countries with access to a preparation budget of around 24 million US dollars diverted their attention to just the one outlook. So, there is a lot to yet unlock.