FAO-UNDP-GEF Biodiversity Project

“Reducing Biodiversity Loss at Cross Border Sites in East Africa”

THREAT REDUCTION ASSESSMENT (TRA) NOTES AND METHODOLOGY CHANGES

Compiled By L. Persha

Threat Reduction Analysis (TRA) is a monitoring tool used to judge the effectiveness of conservation interventions by measuring the changes in the level of threats to biodiversity over time, and to then relate those changes to project activities.  This provides information on whether the biodiversity conservation of an area has been improved or not as a result of project activities, and also (to some extent), why (i.e. which threats have been reduced, which activities have contributed to this, what lessons can be drawn for meeting similar threats in future interventions, etc.). The TRA approach, based on that described by Salafsky and Margoluis in the journal Conservation Biology (1999), monitors threats to the resource rather than changes to biological parameters themselves, as a proxy measurement of conservation impact.  By doing so, this method avoids the need to carry out complex quantitative biological studies in order to generate information on the state of the biodiversity resources.  Accurate scientific assessments of the ecological status of a particular resource (for instance Juniperus procera) may require time investments over years, and often yields results that are difficult to interpret by people without a scientific background.  It is much simpler and cost effective to measure changes to the broad human activities that threaten the integrity of the resource (e.g. logging), and then use that information to judge the state of the resource itself. 

The TRA approach therefore has a number of benefits for assessing ICDP progress and conservation success.  One of these benefits is flexibility in designing a methodology that is tailored to individual projects.  Other benefits include the ability to detect changes within the project timeframe; ease, efficiency and cost-effective of use; results which are readily interpretable by all stakeholders (rather than only to those of a scientific background); flexibility to generate meaningful results in the absence of a firm baseline data set; and effective comparison of results between dissimilar sites.  The TRA approach involves generation of a “Threat Reduction Index”, which requires the assessors to identify threats to the resource, rank these threats in order of their relative importance, determine the degree of progress made towards reducing each threat (over a chosen time period of intervention), and calculate a combined estimate of the amount by which the threats to the resource as a whole have been reduced over the chosen time period of project intervention.  Threat rankings are based on a set of pre-determined criteria, such as area, intensity and urgency.    

Pilot TRAs carried out by Uganda, Tanzania and Kenya have allowed us to pick up some ideas on how to tweak the Threat Reduction Assessment methodology so that it is more applicable to the particulars of our project, ultimately allowing us to complete a more useful assessment of the Biodiversity Project’s progress in reducing threats to biodiversity (and therefore reducing biodiversity loss) at each of the sites. 

TRA involves seven steps, summarised as follows:

1.       Define the project area (spatially and temporally).

2.       Develop a list of all direct threats to the biodiversity at the project site, which were present at the project start date.

3.       Rank each threat based on 3 criteria: area, intensity and urgency.  Area refers to the percentage of the habitats in the site that the threat will affect.  Intensity refers to the impact of the threat within a micro-site – will the threat completely destroy the habitat in a small locality, or will it only cause minor changes?  Urgency refers to the immediacy of the threat – will the threat occur tommorrow or in 15 years?  If there are 4 threats, then the highest ranked threat for each criteria receives a score of 4, and the lowest ranked threat receives a score of 1.

4.       Add up the scores across all three criteria to get a total ranking.

5.       Determine the degree to which each threat has been met (this requires defining what 100% threat met means for each threat).

6.       Calculate the raw score for each threat.  Multiply the total ranking by the percentage calculated in step 5 to get the raw score for each threat.

7.       Calculate the final threat reduction index score.  Add up the raw scores for all threats in each sub-category (forest interior and forest edge threats), divide by the sum of the total rankings, and multiply by 100 to get the threat reduction assessment index for each sub-category.  Total the raw scores and then divide by the total ranking for both categories to get the final threat reduction assessment index, as a percentage.

Suggested changes to the TRA methodology are highlighted below:

1)  Direct threats split into two categories: Forest interior threats and Forest edge threats.

We have noted that the kinds of threats, and seriousness of each threat, vary in and around project forests by location.  For instance, in Ol Donyo Orok, forest fires are a serious threat on forest edges, impacting a large area and causing substantial changes to the habitat; but are negligible in the forest center.  Threats have different levels of scope, impact and urgency in different places in and around the forest.  As a result, lumping all of the threats into one assessment is perhaps not the best answer.  In order to take into account these differences, we have decided to divide threats into two categories and carry out separate sub-assessments on these two categories of threats – forest interior threats and forest edge threats.  Forest interior threats are those which occur inside closed forest, while forest edge threats are those which occur around the forest perimeter (e.g. agricultural encroachment).  The sub-assessments are then combined into a final threat reduction index, using the standard formula [(Raw Score/Total Ranking)*100].

2)  Interpretation of the Threat Met category.

Project staff initially over-emphasised awareness in calculating a number for the Threat Met category.  For example, one team noted that ‘the district team, divisional team and Community Environment Committees have carried out a lot of awareness creation’. This was seen as contributing to stemming the threats to the forest, and was thus incorporated into the threat met percentage for each threat. The “threat met” percentage should only be based on quantifiable evidence of a reduction in the threat (for instance, project indicators such as hard evidence of less trees being cut in the forest, a smaller number of forest fires since the project began, etc.)  – awareness raising is an activity which would feed into seeing this, if the awareness raising has been effective.  But project activities should not be incorporated into the threat met percentage simply because the activities are believed to be effective, as this falsely inflates the final threat reduction index.  We need harder data.

3)      TRA and subjectivity.

The TRA designers point out that their approach is not immune to bias.  Some of the categories used in the assessment hinge on subjective analysis.  The Threat Met category is probably the biggest pitfall in this respect.  When carrying out the assessment, project teams must keep in mind that the objective of the exercise is to gain a realistic understanding of the progress they have made so far, and then use this information to determine which activities may have contributed to improvements in threat reduction (or lack thereof), and tweak future project activities to be more successful.  The TRA ideally should by carried out at predetermined intervals over the project lifetime, giving a trend over time in the amount that threats have been reduced, and thus project effectiveness.  In a perfect world, one would expect to see the final threat reduction index percentage start off rather small (since activities normally take some time to get off the ground, and to have real impact), and grow in an upward trend over the course of the project.  But in a realistic world, some project activities are not as effective as one would hope, others take longer to be effective than planned, others take off quite well.  TRA helps to see this, and enables project teams to use this information to improve on their work.  Project teams need to turn a critical eye toward their own activities, and must try to be as objective as possible when answering questions such as, for example “To what extent have we reduced the threat of tree cutting for charcoal burning and wood carving in Ol Donyo Orok Forest”? 

One would not expect the project to have reduced threats to Ol Donyo Orok Forest by 52.5% in the second year of the project.  This is unrealistic, and if such a number is generated through the TRA, the project team would be advised to look back at the subjective portions of their analysis and rethink some of the numbers.  In this case it is likely that the threat met percentages were overly optimistic.  The TRA’s primary use is to help project teams judge the state of biodiversity threats at project sites, and by association, the strengths and weaknesses of project activities.  This information is then intended to be used to strengthen the project’s future interventions, leading to more effective overall biodiversity conservation.  The primary intention is not to show outsiders how much the project has accomplished.  As such, the TRA is not a useful monitoring tool if it isn’t based in objectivity.     

The meaning of “100% threat met” must be defined, so that we have a measuring stick for interpreting the percentages in the table.  For instance, in the case of uncontrolled forest fires, “100% threat met” may mean no fires at all, and a system of fire detection and village fire-fighting mobilisation in place.  “100% threat met” for tree cutting for poles, charcoal and wood carving may mean a sustainable system of extraction in place to meet community needs.  These are issues that each project team will need to flesh out and agree on.

4)  Understanding and setting a useful time frame of analysis.

An appropriate time frame of analysis must be set and understood before undertaking each TRA.  For instance, the first TRA carried out at each site should use the project start date as the lower time limit for analysis, in order to generate a baseline percent reduction to biodiversity threats at the site since the project began.  This is difficult for the Cross Borders Project, because it requires thinking back in time to the threat situation at the site in 1998 (thus more subjectivity!), and basing analysis on the changes that have occurred to threat levels since that time.  The project team must be careful to judge only those changes that have occurred between 1998 and 2001.  For instance, a 6 year old pitsaw site should not be counted in the analysis as representing a reduction in the threat of logging, because that pitsaw site was also not in operation at the project start (this came up in Uganda’s pilot TRA for Sango Bay).  The TRA is meant to provide information on the reduction of threats to biodiversity resulting from project interventions, over the course of the project lifetime.   

Subsequent TRA analyses should also define the time frame over which threats will be assessed, before undertaking the assessment.  The time frame should be long enough to allow for changes to be seen (if there are any), and footnoted under the table.  The Ol Donyo Orok TRA was conducted from July 1998 – March 2001.  

5)  Groundwork feeding into the TRA.

A standard methodology should be designed for collecting data to feed into the parameter rankings (for area, intensity and urgency).  This methodology would vary by project site, and may vary for individual threats, as well (e.g. assessing burn area, intensity and urgency is not well suited for a simple walking transect methodology; walking transects along existing footpaths in forests gives information on areas of highest human activity/impact, but may miss emerging threats in other parts of the forest.  This methodologies should be decided on by the project team before starting the analysis, and are informed by prior knowledge of the forest layout, areas of human activity, and nature of threats.  It is okay (and probably better) to combine a number of different methodologies to assess threat levels.

Threat Reduction Assessment (TRA) for Chome Forest Reserve

14.3.2001

Participants:            Same Project Team; supported by Marc Baker, Lauren Persha and Alan Rodgers

Area of Chome Forest Reserve:       14,283ha (gazetted)

            11,800ha (measured forest cover from 1:50,000 topo sheets)

For this exercise the threats are split into two categories;

1.         Forest interior: Defined as forest cover beginning 100m from forest edge.

2.                   Forest edge: Includes all forest edge (up to 100m into areas having forest cover), grassland and degraded forest within and including the gazzetted forest reserve boundary.

Category A:  Perceived forest interior threats (not ranked):

§         Logging of mainly Ocotea usambaresis (Camphor).

§         Poaching, which includes hunting for meat (Bush Pig, Duiker spp. and Colobus Monkey) and bird trade (Hartlaubs Turaco).

§         Forest fire (accidental, by pitsawers or other human activities in the forest interior).

§         Cattle grazing and movement along paths.

§         Fuelwood collection.

§         Pole cutting.

Category B:  Perceived forest edge threats (not ranked):

§         Fire (intentional, to maintain or increase grazing areas in or near the forest).

§         Fire (accidental, caused by uncontrolled burning during land preparation for agriculture outside the F.R. boundary)[1].

§         Cattle grazing.

§         Agricultural encroachment.

Table 1. Threat Reduction Assessment for Chome Forest Reserve. [2]

*  Note that the site team accept percentages of Threat Met listed here.

Definitions of % Threat Met for Chome Forest, by individual threat:

1.      FOREST INTERIOR THREATS

Logging of mainly Ocotea usambaresis (Camphor):

A sustainable camphor off-take system in place and followed in Chome Forest Reserve.  Quotas observed by villagers, forest department, and outside loggers.  Large scale illegal logging operations no longer in existence.  Forest patrols occurring on a regular basis.

Poaching which includes, hunting for meat (Bush Pig, Duiker spp and Colobus Monkey), and bird trade (Hartlaubs Turaco):

??? A sustainable off-take system established, or all hunting/snaring/trapping eliminated.  Forest patrols occurring on a regular basis.

Forest fire (accidental by pitsawers or other illegal activities):

Early fire detection and village fire-fighting mobilisation system in place.  Number of fires not exceeding 1 per year, fire instances reduced to natural occurences.  (Note: forest fires will occur on a stochastic basis – the emphasis here is on rapid detection and containment).  Forest patrols occurring on a regular basis.

Cattle grazing and movement along paths:

Long term cattle grazing eliminated from the forest reserve (seasonal grazing in times of unusual drought is permitted).  Cattle passage through the forest reserve allowed along roads only.  Forest patrols occurring on a regular basis. 

Fuelwood collection:

Sustainable fuelwood collection system established.  Alternative fuels / fuel efficient stoves in widespread use in surrounding villages (? In use by 60% of families?).

Pole cutting:

Sustainable pole cutting system established.  (Pole cutting was not seen as a serious threat, as most Pare homes and structures are built of brick, although livestock enclosures may be built with poles).

2.      FOREST EDGE THREATS

Fire (intentional, to increase/maintain grazing areas):

100% of forest reserve boundary cleared.  Early fire detection and village fire-fighting mobilisation system in place.  No cases of forest edge fires in months when land preparation for agriculture does not take place.

Fire (accidental, caused by uncontrolled burning during land preparation for agriculture):

100% of forest reserve boundary cleared.  Early fire detection and village fire-fighting mobilisation system in place.  No cases of forest edge fires spreading into the forest reserve in months when land preparation for agriculture takes place.

Cattle grazing:

No resident cattle inside forest reserve boundary, except at forest department/village authorisation during times of unusual stress (i.e. drought). 

Agricultural encroachment:

No conversion of forest reserve land to agriculture.  No pressure to degazette forest reserve land to meet village agricultural needs.

Threat Reduction Assesment (TRA) for Kitumbeine Forest Reserve, Monduli District

26.3.2001

Participants:            Monduli Project Team; supported by Marc Baker, Lauren Persha and Alan Rodgers

Area of Kitumbeine Forest Reserve:             12,000 ha (gazetted)

              3,500 ha (measured closed forest cover from 1:50,000 topo sheets)

For this exercise the threats are split into two categories;

1.         Forest interior: Defined as forest cover beginning 100m from forest edge.

2.             Forest edge: Includes all forest edge (up to 100m into areas having forest cover), grassland and 

             degraded forest within and including the gazetted forest reserve boundary.

Category A:  Perceived forest interior threats (not ranked):

§         Logging (of mainly Juniperus procera)

§         Cattle grazing

§         Pole-cutting

§         Fuelwood collection

Category B:  Perceived forest edge threats (not ranked):

§         Agricultural encroachment

§         Human settlement

§         Cattle grazing

Table 1. Threat Reduction Assessment for Kitumbeine Forest Reserve. [6]

Definitions of % Threat Met for Kitumbeine Forest, by individual threat:

1.       FOREST INTERIOR THREATS

Logging of mainly Juniperus procera:

A sustainable off-take system in place and enforced in Kitumbeine Forest Reserve.  Quotas observed by villagers and forest department staff. Forest patrols occurring on a regular basis. Large scale replanting efforts underway and showing success.

Cattle grazing and movement along paths:

Long term cattle grazing eliminated from the forest reserve (seasonal grazing in times of unusual drought is permitted).  Cattle passage through the forest reserve allowed only along roads and designated paths.  Forest patrols occurring on a regular basis. Control of cattle movement and limited grazing permitted on sub-montane grassland.

Fuelwood

Sustainable fuelwood collection system established.  Alternative fuels / fuel efficient stoves in widespread use in surrounding villages (? In use by 60% of families?).

Pole cutting:

Sustainable pole cutting system established and observed.

2.       FOREST EDGE THREATS

Cattle grazing:

Cattle gazing permitted inside forest reserve boundary as authorised by  designated forest department/village authorities during times of unusual stress (i.e. drought). Cattle movement in the forest reserve is permitted, but restricted to certain routes which pass around the mountain rather than over it (to reduce cattle impacts on the sub-montane grassland).

Agricultural encroachment / Human Settlement:

No conversion of forest reserve land to agriculture. No pressure to degazette forest reserve land to meet village agricultural needs. A suitable agreement between the forestry division, Monduli District Council and villagers negotiated so that land currently under agriculture and human settlement (eastern side of the reserve) remains so, but no further encroachment takes place. Terms of agreement could include the degazettment of areas currently under settlement in exchange for the gazettement of forested areas or other suitable habitat outside of the present reserve boundary.

Threat Reduction Assessment (TRA) for Taita Hills Forests, Taita-Taveta District, Kenya

19.3.01

Participants:  Kenya Project Team; supported by Lauren Persha and Alan Rodgers

For this exercise the threats are split into two categories;

1.       Forest interior:  Defined as forest cover beginning 100m from forest edge.

2.       Forest edge:  Includes all forest edge (up to 100m into areas having forest cover), grassland and degraded forest within and including the gazetted forest reserve boundary.

Category A:  Perceived forest interior threats (not ranked):

§         Pole cutting (not a threat in Kasigau Forest)

§         Fuelwood collection (Chawia Forest only)

§         Grazing in the forest (Chawia Forest only)

§         Overuse of water resources

§         Timber harvesting (not a threat in Ngangao Forest)

§         Exotic species invasions (Chawia Forest only)

§         Forest fires (Ngangao and Kasigau Forests only)

§         Dry season livestock grazing (Kasigau Forest only)

Category B:  Perceived forest edge threats (not ranked):

§         Habitat conversion to agriculture (not a threat in Kasigau Forest)

§         Forest fires

CHAWIA FOREST TRA

MBOLOLO FOREST TRA

NGANGAO FOREST TRA

KASIGAU FOREST TRA

Definitions of % Threat Met for Taita Hills Forests, by individual threat[9]:

1.      FOREST INTERIOR THREATS

Pole Cutting:

Sustainable pole cutting system established and observed.

Fuelwood collection:

Sustainable fuelwood collection system established. 

Grazing in the forest:

Long term cattle grazing eliminated from the forest  (seasonal grazing in times of unusual drought is permitted).  Cattle passage through the forest allowed only along roads and designated paths.

Overuse of water resources:

Agreements with communities in place regarding efficient water extraction from the forest.

Timber harvesting:

A sustainable off-take system in place and enforced in those forests where harvesting is deemed appropriate.  Quotas observed.  No harvesting taking place in forests where harvesting is deemed unsustainable.

Exotic species invasions:

No new introductions of invasive species, or pressure to plant exotics in the forest.  No further spread of existing invasive species. 

Forest fires:

Early fire detection and village fire-fighting mobilisation system in place.  Instances of fires limited to natural occurances. 

2.      FOREST EDGE THREATS

Habitat conversion to agriculture:

No conversion of forest land to agriculture.  No pressure to convert forest land to agriculture.

Forest fires:

100% of forest boundary cleared.  Early fire detection and village fire-fighting mobilisation system in place.  No cases of forest edge fires in months when land preparation for agriculture does not take place.  Careful land preparation techniques in practice by villagers farming close to the forest edge.  Forest edge fires occurring in months when land preparation does take place are extinguished rapidly and limited in scope.

PILOT THREAT REDUCTION ASSESSMENT EXERCISE IN

SANGO BAY FOREST/WETLAND, UGANDA

The project team for Sango Bay met and agreed on the major threats from those that were identified in the participatory planning process.  The team selected Malabigambo and Kaiso forest blocks of Sango Bay as a starting point.  The major direct threats selected for these forest blocks were:

§        Illegal pitsawing (Ps)

§        Over-harvesting of poles (Pc), and

§        Clearing of under-growth  (Cu)

The field team assessed threats along existing footpaths in the forest.  The team walked through the forest and ranked each encountered threat occurrence in terms of the agreed parameters.  Ranking was therefore done on a scale from 1 to 3; a score of 3 represented the most serious threat, and a score of 1 represented the least serious in terms of area, intensity and urgency.

The field team of 13 people spent 7 days in the forest. 

Analysis of Data

A thinking back in time has been done to assess the “Threat Met” percentage in respect to the situation existing on the ground at the beginning of the project, compared to the current situation.  The degree to which each threat had been met was decided retrospectively by the project team as perceived in comparison to the situation at the beginning of the Project. 

Table 1 indicates that clearing of undergrowth for different purposes has affected the largest area in the forest blocks assessed.  On the other hand, pitsawing is the most intense and most urgent threat followed by over-harvesting of poles.

Table 1:  The Threat Reduction Assessment Index Calculation for Sango-Bay Forests:

Definitions of 100% threat reduction for Sango Bay Forests are as follows:  100% reduction of illegal pitsawing involves eliminating illegal pitsawing in the forest, although there could be regulated pitsawing in designated zones under permits by the Forest Department.  100% reduction of clearing undergrowth involves eliminating all illegal clearance, but could involve controlled clearance according to agreed parameters by stakeholders.  100% reduction of harvesting of poles involves elimination of illegal pole cutting and introduction of sustainable regulated pole cutting in designated zones.

Threat Reduction Assessment (TRA) for Ol Doinyo Orok – Namanga Hill, Kajiado District, Kenya

11.3.2001

Participants:  Kenya Project Team; supported by Lauren Persha and Alan Rodgers

For this exercise the threats are split into two categories:

1.       Forest interior: Defined as forest cover beginning 100m from forest edge.

2.       Forest edge:  Includes all forest edge (up to 100m into areas having forest cover), grassland and degraded forest within and including the gazetted forest reserve boundary.

Category A: Perceived forest interior threats (not ranked):

§         Tree cutting for poles, charcoal burning and wood carving

§         Overgrazing / overbrowsing

§         Destruction of springs by livestock

Category B: Perceived forest edge threats (not ranked):

§         Tree cutting for poles, charcoal burning and wood carving

§         Uncontrolled forest fires

§         Overgrazing / overbrowsing

§         Settlement / encroachment

Calculation of the Threat Reduction Index for Ol Doinyo Orok – Namanga Hill Forest

A = Threats which occur inside higher altitude closed forest (forest interior threats).

B =  Threats which occur inside lower altitude Acacia-Balanites woodland (forest edge threats)

Definitions of % Threat Met for Ol Doinyo Orok – Namanga Hill Forest, by individual threat[12]:

1.      FOREST INTERIOR THREATS

Tree cutting for poles, charcoal burning and wood carving:

Sustainable off-take system in place and observed.

Overgrazing / overbrowsing:

Livestock entry into the forest reserve regulated by designated forest department / village authorities.  Regulations observed by livestock keepers.

Destruction of springs by livestock:

Sustainable water use system agreed to and in place.  Agreements observed by livestock keepers.

2.      FOREST EDGE THREATS

Tree cutting for poles, charcoal burning and wood carving:

Sustainable off-take system in place and observed.

Uncontrolled forest fires:

100% of forest boundary cleared.  Early fire detection and village fire-fighting mobilisation system in place.

Overgrazing / overbrowsing:

Livestock entry into the forest reserve regulated by designated forest department / village authorities.  Regulations observed by livestock keepers.

Settlement / encroachment:

No conversion of forest land to agriculture.  No pressure to degazette forest land to meet village agricultural needs.

Back to Discussion Notes Page