IFRI

UFRIC IFRI RESEARCH FINDINGS

Introduction
Forest composition, structure and threats
Community maps of forest resources
Policy-related conclusions

INTRODUCTION

The IFRI methodology is designed to capture biophysical and socio-economic information. Biophysical information in a forest is collected using nested concentric circles of 1-3-10m radii. 30 plots are sampled in each target forest. In the one meter radius plot, groundcover details (e.g., seedlings, herbs) and percentages of individual species are recorded. In the 3 m. radius circle, saplings and shrub information is recorded, whereas in the 10 m. radius circle, tree measurements such as diameter at breast height (DBH), and height are recorded. Local and botanical names are recorded for all species measured. Socio-economic information is gathered using Participatory Rural Appraisal approaches. Such information includes local community forest use, informal and formal rules for forest access and use, forest-related and non-forest related income and population size and distribution. Special emphasis is given to forest-adjacent communities. A key assumption of this methodology is that relationships between the forest resource and adjacent are mediated by the institutional arrangements (both formal and informal) which are designed to control access to forest resources. This dirctly influences the long-term condition of these resources.

UFRIC research activities started in 1993. So far seventeen sites comprising 18 forests have been studied. Overall, a total of 8,142 tree stems have been measured and recorded. The total number of saplings recorded is 2,523, and total count of shrubs has so far been 599 stems. In terms of diversity, a total of 349 species have been identified. These include 52 shrubs.The others cover trees, saplings, herbs, and seedlings. Regarding average DBH, an overall of 15.60 cm has so far been registered. This covers an average of 4.53 cm for saplings; 22.59 for trees; and 3.64 for shrubs. The total average height recorded so far for the 18 forests is 10.02 m.

TOP

FOREST COMPOSITION, STRUCTURE AND THREATS

Effects of Biophysical factors on forest condition in Mpigi district, Central Uganda (Joseph Bahati):

This study investigates variation in forest condition associated with variation in elevation, slope steepness, slope orientation, and soil texture. Basal area of trees and saplings were used as indicators of forest tree size. Data were collected from a total of twelve forests located in Mpigi district in Uganda in 1994 and 1995. All forests are located in the tall grassland/moist tropical forest zone of Uganda, and are botanically classified as medium altitude Piptadiastrum-Albizia-Celtis forests after the three dominant tree species typical of such forests. Results from an ordinary least squares regression indicates that a fairly low amount of variation in forest tree size can be associated with the factors investigated. About 3% (p-value= 0.04799) of variation in tree size can be associated with the physical factors. Slope elevation was highly significant (4.53E-7), showing that a 1⁰ increase in slope elevation is associated with a decrease in basal area of 0.08 cm in tree size. The forests occur, on average, between 1100 and 1400m above sea level. Also significant (p-value=0.0293) is the effect of slope angle on tree size. Slope angle (measured in degrees) refers to the steepness or otherwise of the slope. Data from these forests indicate that tree size, on the average, increases by 0.6328 cm with a 1⁰ increase in slope angle. The other factors of slope aspect and soil texture showed no significant effects on tree size. Similar patterns are evident in the regression of sapling size on elevation, slope angle, aspect and soil texture. About 3% (p-value=0.04799) of the variation in sapling size can be associated with these physical factors. However, with the exception of soil texture, all other factors are not significant. Given the very low, though significant, effect of physical factors on tree size, it is likely that variation in tree size in the forests of Mpigi district may be exaplained by factors other than the physical. The author suggests that human factors such as forest management practices (enrichment planting, for instance), pitsawing, and commercial fuelwood extraction are possible explanatory factors for variations in forest tree and sapling sizes in Mpigi district, Uganda.
These results have been extracted from a paper presented during the Workshop in Political Theory's mini conference of December, 1998

TOP

Early detection of tropical forest degradation: An IFRI pilot study in Uganda (Dusty Becker, Abwoli Banana, William Gombya-Ssembajjwe):

This study was conducted in two adjacent forest areas, Lwamunda and Namungo, both located in the tall grass/moist tropical forest zone of central Uganda. Although both forests are located in the same ecological zone, they differ in governance, rules, incentives and local use patterns. Lwamunda is a forest reserve which has been under government (i.e Forest Department) management ever since 1900. In contrast, Namungo is 'mailo land' or private land which has been transferred among Buganda elites since 1900. This study addresses the following two questions:

1. Do Namungo and Lwamunda forest differ significantly in 10 hypothetical sources of institutional stability?

2. Do ecological measures of the physical structure, biodiversity, and signs of human expliotation in these forest patches correspond in any predictable way to institutional stability?

The authors hypothesize that the physical conditions of both forests should reflect the local rules, incentives, day-to-day actions of local people, on the forests under the two different institutional settings. A total of 30 randomly-located plots were sampled in about 100 acres of each forest, data were collected using the standard IFRI protocol. Data collected about the trees in the sample plots, and the types of open-access commercial exploitation on the plots, were compiled to assess the physical and biological condition of the two forests. Exploitation reflects current use while age-class structure, biomass per unit area, and biological diversity of the forests, reflects both past and present use-patterns. Diversity indices were calculated, and the dbh size-class distribution of trees, total basal area and importance values were also calculated.

Results indicate that:

1. There were no significant differences between soil types, slopes or drainage patterns between the two forests.

2. Many of the official rules applying to local use, though similar in both forests, differed in the local community's adherence.

3. Out of the 10 sustainability factors predicted to affect institutional sustainability, Namungo had six factors present while Lwamunda had only two. See following table:

SUSTAINABILITY FACTORS	NAMUNGO FOREST	LWAMUNDA FOREST
Economic and population factors:
1. Distance to markets for forest products	_	_
2. Population growth rate	_	_
3. Population pressure in surrounding area	_	_
Institutional design and rule factors:
4. Institutional stability and understanding of users	+	_
5. Monitoring, sanctions, conflict resolution, and governance are organized in multiple layers of nested enterprises	+	+
6. Rapid access to low cost conflict resolution	+	+
7. Graduated sanctions on violations enforced	+	_
8. Quality of monitoring	+	_
9. Local users can participate in modification of rules	+	_
10. Local users design institutions governing use of forest	_	_
SUM OF SUSTAINABILITY FACTORS	6	2

4. Commercial exploitation of trees was more frequently recorded in Lwamunda than in Namungo (chi-square= 17.4, df=3. p<0.001). Evidence of commercial firewood-cutting, pit-sawing, and/or charcoal making were noted on more than half the plots in Lwamunda, while only 4 out of 30 plots in Namungo revealed evidence of such exploitation. Although total number of trees and basal area were on average higher in Namungo than Lwamunda forest, the dbh and mean value of number of trees per plot were not significantly different between the two forests. Both forests were dominated by trees having dbh's in the range of 10 to 40 cm; very large trees greater than 80cm dbh were rare in both forests. The following table provides a summarry of these results:

	Namungo (Private forest)	Lwamunda forest reserve (Governmental)
Species richness	64	69
Shannon diversity index	3.69	3.67
Simpson's diversity index	30.2	27.8
Trees in 30 plots	339	325
Trees per hectare	360	345
Mean number of trees per plot	11.3 plus or minus 4.5	10.8 plus or minus 6.2
Mean dbh's (cm)	23.9 plus or minus 18	23.1 plus or minus 17.1
Basal area	23.9 plus or minus 0.4	21.9 plus or minus 0.4

The authors conclude that Mr. Namungo, the private owner of Namungo forest, has ultimate control over his forest, and that its sustainability depends greatly upon his personal ethics, values, knowledge and the local respect for his rights as an owner. The authors also observe that the present rate of destruction in Lwamunda forest reserve can be partially explained by the lack of monitorign by guards, reluctance to enforce sanctions, lack of well-defined rules concerning local use, and lack of participation by local users in making any rules. The results of this investigation suggests that IFRI methodologies are sensitive to initial patterns of degradation in forest ecosystems, and are useful for characterizing changes in tropical forest plant communities in conjunction with societal and institutional behavior. The IFRI method captures many details of the societal and ecological aspects of deforestation.
These results have been extracted from: Environmental Conservation 22(1) Spring 1995: 31-38

TOP

COMMUNITY MAPS

Using participatory mapping techniques, communities surrounding forest areas that have been sampled via IFRI techniques are able to construct a spatial representation of their perception of the distribution of their resource. The following pictures show the intensive mapping procedures that IFRI investigators employ:

First, the investigators have lengthy discussions with various sections of the community regarding the importance of the resource, changes in the structural characteristics of the forest over time, and problems communities have to deal with in regard to the resource. The top picture is a representation of this part of the process. The middle picture captures the process during which the community members actually draw, on the ground, their perception of the distribution of resources in their area.

The bottom picture represents a very important part of the community mapping procedure. During this time, IFRI investigators return to the community with a paper replicate of the map the community had drawn on the ground. During this final consultation the community verifies whether the paper map is a genuine representation of their perceptions.

TOP

POLICY-RELATED CONCLUSIONS

Property rights and the sustainability of forests in Uganda (William Gombya-Ssembajjwe and Abwoli Banana):

This study builds upon the study by Becker et al. described above. In this analysis, the authors selected a total of 12 forests in Mpigi districts which occur in the same ecological zone, and which have similar vegetation composition. These twelve forests ranged over a total of four institutional arrangements which comprised of: government nature forests, government exploitation forests, individually owned forests, and communal sacred forests. The mediating factors of population density, and distance from market centers were also taken into account. The following is a matrix for the selection of forests studied:

Distance from Kampala	Governmant nature forest	Government exploitation forest	Individually-owned forests	Communal sacred forests
Within 30 Km	Buttobuvuma	Lwamunda	Mugalu	Bukasa
	Mpanga	Mugomba	Namungo	Semalizi
Within 80 Km		Kizzikibi	Lukambagire	Magezigoomu
		Kyambogo	Najjakulya	Mukasa

An assessment of the 10 institutional sustainability factors across institutional arrangements indicate that communal arrangements had seven positive factors and three negative factors, individual forests had six positive factors and four negative ones, while forests under government institutional arrangements had two positive factors and eight negative ones. Thus communal forests are more likely to be efficiently and sustainably utilized. Government forests were least likely to be sustainably managed. These results are summarrized in the table below:

Sustainability factors	Government forest	Private forest	Communal forest
Economic and population factors:
Distance to markets for forest products	_	_	_
Population growth rate in settlements	+	+	_
Population pressure from surrounding area	_	_	_
Institutional design and rule factors:
Institutional stability and clarity to users	_	+	+
Monitoring, sanctions, conflict resolution and governance are organized in multiple layers of nested enterprises	+	+	+
Quality of monitoring	_	_	+
Rapid access to low cost conflict resolution	_	+	+
Enforcement of graduated sanctions on violations	_	+	+
Local users design institutions governing use of forest	_	_	+
Local users may participate in rule modification	_	+	+
SUM OF SUSTAINABILITY FACTORS	-6	+2	+4

TOP

An investigation into the intensity of human disturbances across the institutional arrangements and distances from main city of Kampala revealed a distinct pattern of degradation. Forests within 30 Km from Kampala showed that forest degradation first began with pitsawing, which progressed to charcoal burning and finally culminated in agricultural encroachment. Forests within the 30-80 Km range showed a degradation pattern that began with pitsawing, progressed into charcoal burning but did not culminate into agricultural encroachment. There was more occurrence of pitsawing in forests that are 30-80Km from Kampala than in forests that are within 30Km from Kampala. On the other hand, charcoal burning and agricultural encroachment was more frequent in forests that are within 30Km from Kampala than those that are far away. Chi-square results show that the level of disturbance in forests near the market is significantly higher than in forests located far away from the market. In terms of disturbance, government nature reserves and sacred communal forests had the lowest number of disturbed plots. The government exploitation forests had the highest number of disturbed plots. A chi-square analysis shows that the observed differences in the levels of disturbance between institutional arrangements was statitstically significant. The table below shows the types of human disturbances and their occurrences across institutional arrangements and forests:

Institutional arrangement	Forest	Size (ha)	No disturbance	% with no	PS	CB	CF	AE	Others	Total plots
Government nature	Mpanga	453	28	93	0	0	0	0	2	30
	Buttobuvuma	1096	4	13	3	10	5	4	4	30
Government exploitation	Mugomba	725	0	0	2	7	13	3	5	30
	Lwamunda	4696	10	33	13	3	4	1	4	30
	Kizzikibi	520	2	7	16	2	3	0	7	30
	Kyambogo	760	4	13	23	1	0	0	2	30
Private individual	Mugalu	150	3	10	8	9	6	3	1	30
	Namungo	40	17	57	3	1	1	7	1	30
	Lukambagire	100	7	23	0	4	10	7	2	30
	Najjakulya	50	17	57	10	1	0	1	1	30
Private communal	Bukasa	0.5	3	100	0	0	0	0	0	3
	Semalizi	0.5	5	100	0	0	0	0	0	5
	Mukasa	1.0	8	100	0	0	0	0	0	8
	Magezigoomu	20	10	33	10	3	1	2	4	30

KEY:

PS=pitsawing; CB=charcoal burning; CF=commercial firewood; AE=agricultural encroachment.

The authors observe that Uganda's forest policy, as with many other developing countries, has been characterized by a strong concentration of power over forest resources in the central state apparatus, and a corresponding lack of local participation in forest and tree management. The net result has been a loss of incentives by the local communities to protect trees; discouragement of local people to engage in tree planting and reforestation projects; and excessive reliance by the state on punitive measures to enforce law. The authors suggest that incentives for local people be improved by increasing the value of the resource to local people by, for example, granting more access rights or by granting local communities a percentage of forest concession revenues.

Insecurity of land and tree tenure may aslo explain the observed degradation of forests in Uganda. Tenants on freehold land are discouraged from planting tree by their land lords. Similarly, local communities are not willing to sustainably use and manage state owned forest reserves beacuse of a lack or reduced access rights. Insecure tenure rights alone do not explain the observed variance on the forests studied because some government forests are in good condition while some private and public forests are degraded. This variance may be accounted for by rule enforcement at the local level. Monitoring of governmnet forest reserves is costly and difficult because they are often large with long boundaries, requiring many forest guards to monitor them effectively. The financial and human resources are inadequate to carry out the task of policing these forests. Government officials who monitor and enforce rules are poorly paid and unmotivated. Hence forest users who choose not to comply to rules can easily escape detection.

The efficiency and sustainability of forest management can be improved if local communities are involved in the management of forest resources in their vicinity. Employing locals to monitor in the place of regular national staff, together with increased tangible benefits to the local communities will greatly improve the conditions of the forest.

This study has shown that forests resources are more likely to be sustainably utilized if an effective structure of institutional arrangements exists that gives rise to an auhtority system meaningful at the local level. A government forest reserve (state property) and a private forest (private property) can be as degraded as a communal forest (common property) if therte are no effective institutional arrangements and associated organizational mechanisms to monitor and enforce rules. Regardless of the de jure property regime, all forests can be de facto open access regimes if there are no effective instituions and mechanisms to the rule.
These results have been extracted from a paper presented at the International Association for the Study of Common Property in Vancouver, British Columbia, June 10-13, 1998.

TOP