| |

Received: 7 August 2018    Revised: 13 March 2019    Accepted: 19 April 2019

DOI: 10.1111/aje.12634

ORIGINAL ARTICLE

Challenges facing the introduction of SMART patrols in a game

reserve, western Tanzania

Paulo Wilfred1  | Heri Kayeye2 | Flora John Magige3 | Alex Kisingo4 |

Cuthbert L. Nahonyo3

1
Department of Life Sciences, The Open
University of Tanzania, Dar es Salaam, Abstract
Tanzania The spatial monitoring and reporting tool (SMART) is being implemented in Tanzania's
2
Sokoine University of Agriculture,
protected areas to help improve the efficiency of ranger patrols. Unfortunately, there
Morogoro, Tanzania
3
University of Dar es Salaam, Dar es Salaam,
has been no attempt to understand users' perspectives regarding site‐specific fac-
Tanzania tors likely to affect its use. In this study, we investigated the perspectives of staff in
4
College of African Wildlife Management, Ugalla Game Reserve, a protected area in western Tanzania, to understand the chal-
Moshi, Tanzania
lenges that affect the use of SMART in the reserve. The main challenges included a
Correspondence lack of motivation to use SMART, limited knowledge of SMART among game scouts,
Paulo Wilfred, Department of Life Sciences,
The Open University of Tanzania, P.O. Box insufficient ranger capacity, difficulty collecting data during night patrols, limited re-
23409, Dar es Salaam, Tanzania. sources for patrolling, and difficulty accessing some remote sections of Ugalla. The
Emails: [email protected]; paulo.
[email protected] presence of trophy hunting company patrol teams has led Ugalla rangers to concen-
trate their effort in less‐patrolled areas. We recommend introducing incentives to
Funding information
Tanzania Commission for Science and encourage game rangers to use SMART alongside improving patrol coverage in wet
Technology, Grant/Award Number: CST/ seasons. Advanced and regular refresher trainings in SMART should be conducted to
NFAST-SIDA/CRG/2015/14
enhance data collection. Furthermore, game scouts should be trained and equipped
to participate effectively in the SMART process. Although SMART is now becoming
increasingly popular in Tanzania, understanding local factors that influence its imple-
mentation will be important to improve uptake.

Résumé
Le Spatial Monitoring and Reporting Tool (SMART) est mis en œuvre dans les
zones protégées de la Tanzanie pour aider à améliorer l'efficacité des patrouilles de
garde‐forestier. Malheureusement, rien n'a été tenté pour comprendre le point de
vue des utilisateurs concernant les facteurs propres au site susceptibles d'affecter
son utilisation. Dans cette étude, nous avons examiné les perspectives du person-
nel de la réserve animalière d'Ugalla, une zone protégée dans l'ouest de la Tanzanie,
afin de comprendre les défis qui affectent l'utilisation de SMART dans la réserve.
Les principaux défis comprenaient un manque de motivation à utiliser SMART, une
connaissance limitée de SMART chez les gardes‐chasse, une capacité insuffisante
de garde‐forestier, des difficultés à collecter des données pendant les patrouilles
nocturnes, des ressources limitées pour les patrouilles et des difficultés à accéder
à certaines parties éloignées d'Ugalla. La présence d’équipes de patrouilles de la
compagnie de chasse aux trophées a amené les garde‐forestier d'Ugalla à concentrer

Afr J Ecol. 2019;00:1–8. © 2019 John Wiley & Sons Ltd |  1

wileyonlinelibrary.com/journal/aje  
 |
2       WILFRED et al.

leurs efforts dans des zones moins patrouillées. Nous recommandons de mettre en
place des mesures incitatives pour encourager les gardes‐chasses à utiliser SMART
tout en améliorant la couverture des patrouilles pendant les saisons des pluies. Des
formations avancées et régulières de perfectionnement dans SMART devraient être
menées pour améliorer la collecte de données. En outre, les gardes‐chasse devraient
être formés et équipés pour participer efficacement au processus SMART. Bien que
SMART devienne de plus en plus populaire en Tanzanie, il sera important de com-
prendre les facteurs locaux qui influencent sa mise en œuvre pour améliorer son
adoption.

KEYWORDS
game scouts, illegal activity, motivation, patrol efforts, spatial monitoring and reporting tool,
Tanzania, Ugalla rangers

1 |  I NTRO D U C TI O N enables rangers to conveniently collect and organise spatial infor-
mation about illegal activities while on patrol (Sintov, Seyranian, &
The threat that protected areas across different ecosystems in Lyet, 2018). This information is entered into the SMART database
Africa face from illegal activity is not new and is widely accepted for later use in patrol planning and monitoring (ZSL, 2018). A da-
(Davies & Brown, 2007; Ripple et al., 2016; Taylor & Dunstone, tabase in this context is a set of data stored in a computer that is
1996). Such activity can take different forms, including encroach- electronically accessible for various uses. SMART has been reported
ment for settlement and agriculture, which cause habitat loss to be effective in reducing illegal activities in Russia (Hötte et al.,
(Caro & Sherman, 2011; Hofer, Campbell, East, & Huish, 1996; 2016); presently, it is used throughout Ugandan protected areas
Wittemyer et al., 2008), and the illegal harvest of wildlife, plants (e.g., Critchlow et al., 2016), and other Africa's protected areas, such
and other forest‐based resources for subsistence and commercial as Gonarezhou National Park in Zimbabwe, and North Luangwa eco-
purposes (Harrison, Baker, Twinamatsiko, & Milner‐Gulland, 2015; system in Zambia (Henson, Malpas, & D'Udine, 2016).
Lawson, 2014; Poulsen, Clark, Mavah, & Elkan, 2009). These have In Tanzania, SMART has been introduced in some game re-
become central challenges faced by conservationists in Africa. For serves (trophy hunting areas) and national parks (strictly protected,
example, unauthorised timber harvesting threatens the sustain- no trophy hunting; SMART, 2017). In order for SMART to achieve
ability of the miombo ecosystems of Tanzania and elsewhere in a sustainable impact and improve conservation, it is crucial that
Africa (Jew, Dougill, Sallu, O'Connell, & Benton, 2016), and bush- challenges that could potentially affect the introduction of such a
meat hunting is the leading cause of wildlife population declines in technology are understood and taken into account (Sintov et al.,
many African ecosystems (e.g., Milner‐Gulland & Bennett, 2003; 2018). However, to the best of our knowledge, there has been no at-
Ripple et al., 2016). Law enforcement, principally in the form of tempt to assess the challenges in using SMART effectively faced by
ranger patrols, is the most common method used to deter illegal game officers and rangers since its introduction to Tanzania's game
activities in protected areas (Gandiwa, Heitkönig, Lokhorst, Prins, reserves. Given this, we applied qualitative methods to investigate
& Leeuwis, 2013). the experiences of game officers and rangers in a game reserve in
A number of studies have evaluated the effectiveness of law en- western Tanzania.
forcement patrols, and how these can be improved (e.g., Holmern, Ugalla Game Reserve (hereafter Ugalla), in western Tanzania,
Muya, & Roskaft, 2007; Gandiwa et al., 2013; Johnson et al., 2016). experiences high levels of illegal activity (Wilfred, Milner‐Gulland,
Ranger‐collected data are increasingly used to evaluate and analyse & Travers, 2017). Logging and the killing of elephants (Loxodonta af‐
patrol efficiencies in order to improve law enforcement (Critchlow ricana, [Anonymous, 1827]) and other species of conservation im-
et al., 2016; Keane, Jones, & Milner‐Gulland, 2011; Moore et al., portance are widespread in the area (Wilfred & MacColl, 2014). Law
2018). One contemporary software package used to collect, man- enforcement has been the predominant deterrent to illegal activi-
age and process these data is the Spatial Monitoring and Reporting ties, where anti‐poaching patrols are conducted on a monthly basis
Tool (SMART; http://smart​conse​r vati​ontoo​ls.org; see also Critchlow in the reserve. Nevertheless, illegal resource use remains a signifi-
et al., 2016; Hötte et al., 2016). SMART is a science‐based conserva- cant problem (Wilfred et al., 2017). Consequently, the management
tion technology that has been developed in recognition that “tradi- of Ugalla is gradually incorporating and applying SMART technology
tional tools, technologies and resources are not stemming the illegal to enable efficient and effective measurement and improvement
use of wildlife resources and the resulting loss of biodiversity” (ZSL, of patrol efforts for the purpose of controlling poaching. As an es-
2018). It makes use of a geographical information system (GIS) that sential step in meeting this goal, our study aimed to determine the
WILFRED et al. |
      3

challenges that could affect the use of SMART in the reserve based scouts on a part‐time basis to conduct patrols in their hunting blocks.
on their initial 2‐year experience of SMART‐based ranger patrols. These scouts are local people, who are unarmed and relatively poorly
trained. While Ugalla rangers have their own patrol teams, two to
three of them are normally assigned to guide and supervise each of
2 | M E TH O DS
the trophy hunting companies' patrols, as the primary responsibility
(Baraka, B., pers. comm.). The Ugalla manager and other game offi-
2.1 | Study area
cers, in addition to participating in patrols, perform administrative
Ugalla (approximately 5,000 sq. km., Figure 1) was gazetted as a game tasks, including patrol planning, compiling reports from patrol lead-
reserve in 1965 (Fisher, 2002). The reserve experiences a tropical ers, and preparing evidence for court cases.
climate, defined by a single dry season (July–December) and a sin-
gle wet season (January–June). The annual rainfall ranges between
2.2 | Data collection
700 and 1,000 mm, and mean maximum and minimum temperatures
range between 28–30°C and 15–21°C, respectively (Hazelhurst & Basic SMART training was provided to twenty Ugalla rangers and
Milner, 2007; Mbwambo, 2003). The management of Ugalla is ad- six game officers in setting up a conservation area, map navigation
ministered by the newly formed Tanzania Wildlife Management and GIS, patrols, data analysis, report generation, planning, data
Authority (TAWA) through Ugalla's main office in Tabora. Ugalla model management and administrative tasks. The technical training
mainly consists of miombo woodland with commercially valuable manual for SMART 3.2, guides and other materials found at http://
timber tree species (UGR, 2006) and supports a wide variety of wild- smart​conse​r vati​ontoo​ls.org were used. Throughout the training,
life, including African elephant and endangered African wild dogs participants worked in pairs on desktop computers with close sup-
(Lycaon pictus, [Temminck, 1820]; Wilfred & MacColl, 2016). The re- port from trainers (Figure S1). SMART v.3.2.1 was installed on all the
serve is also an important bird area (BirdLife International, 2018) and computers, and the data model was configured and customised for
is part of the Malagarasi‐Muyovozi Ramsar Site (Kalumanga, 2015) data collection in Ugalla. Participants took part in dummy SMART
that protects habitat for the vulnerable shoebill (Balaeniceps rex, patrols soon after the training session prior to embarking on the
[Gould, 1850]) and wattled crane (Bugeranus carunculatus, [Gmelin, actual patrols. Over 24  months, between March 2016 and March
1789]; John, Nahonyo, Lee, & Msuya, 2013). The area also hosts a 2018, all anti‐poaching patrols in Ugalla used SMART. Although each
diverse range of fish (UGR, 2006). patrol took varying lengths of time, the minimum number of days
The main legal activity in Ugalla is trophy hunting, which is ad- per patrol was 14. Following Nahonyo (2005), an effective patrol day
ministered by trophy hunting companies each year between June was taken to be at least 8 hr of patrol time per day. Each patrol con-
and December in three hunting blocks: East Ugalla, South Ugalla and sisted of a driver, a patrol leader and another six to eight patrollers. In
North Ugalla (Figure 1). East Ugalla is leased to Game Frontiers of Ugalla, for both ranger patrols and trophy hunting company patrols,
Tanzania Ltd, with the two other hunting blocks leased to Tanzania patrol leaders were appointed from Ugalla rangers. All patrol leaders
Game Trackers Safaris (TGTS). The concessionaires are also respon- were given the Global Positioning System (GPS) units, cameras, field
sible for conducting anti‐poaching patrols in collaboration with notebooks and pens for recording fresh incidences of illegal activi-
Ugalla rangers (Baraka, B., pers. comm.). Each company hires game ties. It was necessary for study participants to use the basic SMART

F I G U R E 1   Ugalla Game Reserve,

western Tanzania. Names show
approximate locations of trophy hunting
blocks. Inset shows the location of the
protected area in Tanzania
 |
4       WILFRED et al.

equipment – as first‐time users of the technology – first, because TA B L E 1   Challenges of introducing SMART in Ugalla based
the equipment was affordable, and second, because most rangers on focus groups. Frequency = overall frequency of challenges as
reported by game officers and rangers
were familiar with them.
Both direct and indirect observations of illegal activities were Challenge Frequency Percentage
recorded. Direct observations included arrests or sightings of law b
Fewer game rangers 11.0 13.3
breakers, whereas indirect observations included other signs of ille-
Little to no incentive for patrollers to 8.5 10.3
gal activity such as foot prints and poachers' camps (see Table S1). collect data for SMARTa
For each observation, location, observation date, time and sign type
Low willingness to participate in 7.5 9.1
were recorded, and appropriate pictures were taken where possi- SMART responsibilitiesa
ble. Additional information about anyone arrested included name, It is difficult to collect data in night 7.0 8.5
village and reason for arrest. Obstacles were also recorded, such as ranger patrolsc
patrol vehicle breakdowns and getting stuck. We categorised signs Patrollers are neither familiar with 6.5 7.9
into four common illegal activity types in Ugalla (e.g., Wilfred et al., SMART equipment nor well versed
2017), namely logging, poaching, fishing and unauthorised entry in how to use thema

(signs that indicated the presence of unauthorised individuals in the Rainy seasons affect ranger patrolsc 6.5 7.9

reserve, but could not be directly associated with other illegal activ- Inadequate SMART equipment during 5.5 6.7
split patrolsb
ities). Waypoints were recorded every 20–30 min as patrols moved
along Ugalla roads, and were later downloaded from GPS units to the Coordinates from previous patrols 5.5 6.7
bias future patrolsc
SMART database. Other information was recorded in notebooks and
No Follow‐up traininga 5.0 6.1
submitted to the game officer (the data manager) assigned to enter
it into the database. The data manager analysed the information in Knowledge about SMART is for 3.5 4.2
certain groups onlya
SMART and generated reports that showed the distribution of ille-
Weather conditions affect the use of 3.5 4.2
gal activities and patrol coverage. These reports were then used to
SMART equipmentc
improve the targeting of patrols in priority areas, in particular those
There is no enough time for data 2.5 3.0
in need of increased patrol efforts, and intelligence activities. During
entryb
the training and practice patrols, we ascertained that the amount
A perception that SMART is only for 2.0 2.4
and type of information that rangers had to collect did not distract certain peoplea
them from their actual work. SMART reports are seldom 2.0 2.4
Focus group discussions were held alongside a questionnaire generatedc
survey after 2 years of SMART patrols to determine the challenges Evaluation meetings are fewc 2.0 2.4
study participants faced when using SMART technology. Separate Data managers are transferred or 1.5 1.8
focus group sessions were conducted for twenty Ugalla rangers given other responsibilitiesb
and five game officers. This was to create an informal and relaxed Favouritism in assigning SMART 1.0 1.2
atmosphere in each group and ensure that participants' titles did responsibilitiesa
not inhibit discussion. Discussion among members was encouraged Difficulties in taking coordinates at 1.0 1.2
throughout the exercise, and the groups freely listed the challenges the actual scene of crimec

associated with using SMART. Following the focus group discus- Data collectors interfere with patrol 0.5 0.6
sions, all 25 participants independently completed a questionnaire activitiesc

that addressed the challenges of using SMART. Our GPS devices do not have Ugalla 0.0 0.0
map on themc
Total 82.5  
2.3 | Data analyses a
Challenges related to participation in SMART.
b
Patrol effort was used to cross‐validate the results of the focus group Challenges related to resources.
c
Challenges related to patrols.
discussions and questionnaires. Therefore, spatial data from the
SMART database were downloaded into ArcGIS 10.1 (ESRI, 2012)
to produce coverage maps for each patrol team (Ugalla rangers and Common themes pertained to challenges facing the introduc-
trophy hunting companies) and the distribution of illegal activities tion of SMART patrols in Ugalla were identified in the qualitative
detected in both the dry and wet seasons. We used chi‐squared and focus group and questionnaire data. Responses within each of the
one‐sample Wilcoxon signed rank tests to determine whether there themes were then grouped together, coded and analysed. Focus
were significant differences between the dry and wet seasons in en- group participants compared each challenge of using SMART to
countering illegal activities, and between effective patrol days and the other in a pairwise fashion to create a matrix of challenges,
the minimum number of patrol days for Ugalla, respectively. We also with minimal assistance from researchers. The percentage fre-
calculated percentages of patrols and observations. quency that focus group participants chose each challenge was
WILFRED et al. |
      5

TA B L E 2   Challenges of introducing SMART in Ugalla based on a patrols were conducted, of which 25 (60%) were conducted in the
questionnaire survey dry season and 17 (40%) in the wet season. Trophy hunting com-

Challenge Frequency Percentage panies conducted a total of 22 patrols (14 [64%] in the dry season
and 8 [36%] in the wet season), whereas Ugalla rangers conducted
There is inadequate patrollers and other 18 23.4
a total of 20 patrols (11 [55%] in the dry season and 9 [45%] in the
resourcesb
wet season). Patrols experienced obstacles on 28 occasions during
Rainy seasons constrain data collectionc 13 16.9
the fieldwork, some of which occurred in the partially protected
Poor motivationa 12 15.6
areas adjoining the reserve (Figure S1–S4 and Table S1–S3). More
Low knowledge of SMART and the use 7 9.1
obstacles were encountered during the wet season (68% of the
of SMART facilitiesa
total) than the dry season (22%). In total, 169 observations of il-
Infrequent traininga 7 9.1
legal activities were recorded, with 20 (12% of the total) direct ob-
The reluctance of rangers to act as data 4 5.2
servations and 149 (88%) indirect observations (Table S1). Patrols
collectorsa
made 63% and 37% of observations across dry and wet seasons,
SMART data collection is time 3 3.9
consumingc respectively. Patrol teams were more likely to observe illegal ac-

Some rangers do not see themselves as 3 3.9

tivities during the dry season than the wet season (χ2 = 16.17, df=1,
part of the project a p = 0.001, Figure S3).
Data collection is tricky when rangers 2 2.6
apprehend or give chase to poachersc
3.2 | Challenges facing SMART patrols
Carrying SMART equipment during 2 2.6
patrols is tiresomec Study participants reported three groups of challenges facing the
Low ranger moralea 1 1.3 use of SMART in Ugalla during the focus group discussions (Table 1)
Fewer evaluation reports to inform 1 1.3 and in response to the questionnaires (Table 2). These included chal-
future patrolsc lenges related to ranger motivation for using SMART, challenges re-
Data managers do not have adequate 1 1.3 lated to the availability of resources, and challenges related to patrol
time to enter data into the SMART activities.
databaseb
Recording observations takes up a lot 1 1.3
of timec 3.2.1 | Lack of motivation to use SMART
c
Patrol roads are in poor condition 1 1.3
Resentment towards SMART data managers and the Ugalla man-
Infrequent evaluation meetingsa 1 1.3
agement over the way that SMART was being administered was
Total 77  
apparent during the group discussion for game rangers, and as a
a
Challenges related to participation in SMART. result, participants perceived that their motivation for SMART
b
Challenges related to resources. was affected. On the other hand, low willingness to participate in
c
Challenges related to patrols.
SMART was a matter of great concern among game officers and
was discussed passionately and at length in their focus group. The
calculated (Figure S1–S4 and Table S1–S3). Finally, the proportion importance of poor motivation was also apparent from the ques-
of questionnaire responses identifying different challenges study tionnaire survey. Absence of refresher training was highlighted as
participants deemed important was calculated. an important challenge during the focus group discussions (6%)
and the questionnaire survey (9% of the Ugalla staff). Participants
perceived that it contributed to low morale among patrol team
3 | R E S U LT S
staff, as most of them had little experience with SMART. Poor
previous GPS experience and general SMART knowledge among
3.1 | Patrol efforts
game scouts and some rangers was perceived as a factor limiting
The project period consisted of 642 effective patrol days, rang- efficient gathering of appropriate information for use in analysing
ing from 5–39  days per patrol. The number of days spent on pa- patrols (with a frequency of 9% and an overall frequency of 8%
trol was not significantly different from the standard number of from the questionnaire survey and focus groups, respectively). For
days per patrol for Ugalla (Median  =  14  days, p  =  0.271). Ugalla example, a participant in the rangers' focus group stated, “there is a
rangers spent 294 days in total on patrol, and mostly covered the deep lack of understanding among game scouts and some rangers
eastern part of Ugalla, as well as areas along rivers. In contrast, of how to use SMART equipment because this is something that
trophy hunting companies spent 348 days on patrol, with their pa- most of us never learned in school… they simply do not know what
trols concentrated in the northern and south‐western parts of the is needed, what information should be included in field notes, and
reserve (Figure S2). In total, 258 patrol days took place during the how SMART can make this information useful for controlling illegal
wet season and 384 during the dry season. Forty‐two individual activities.”
 |
6       WILFRED et al.

They realised night patrols had their limitations in exhaustively ob-

3.2.2 | Limited resources
serving and recording information on illegal activities. The majority
Scarcity of resources in the form of skilled personnel, equipment of the game rangers experienced difficulty with the use of cameras
and consumables, such as batteries for cameras and GPS units, con- and GPS receivers in darkness and cloudy conditions.
stituted a central challenge for SMART patrols. This challenge re-
ceived the highest percentage frequency in both the focus groups
(13%) and questionnaires (23%). Study participants admitted that it 4 | D I S CU S S I O N
was at times difficult to carry out patrols based on the reports gen-
erated by SMART because of limited resources. For instance, there Our intention has been to draw attention to factors that might influ-
were sometimes no patrol vehicles ready for immediate mobilisa- ence the use of SMART in a protected area, rather than assessing
tion for patrol trips. Game rangers noted that during split patrols the efficiency of SMART as a conservation tool. SMART is undoubt-
each team was forced to depend on only one GPS unit, and one edly effective in improving law enforcement in protected areas
of the teams may have no camera for taking suitable pictures that (Critchlow et al., 2016; Hötte et al., 2016). However, effective use
would be stored in the SMART database as evidence of illegal activ- of such a conservation tool depends on a thorough understanding
ity (Table S2). Game officers realised that GPS units, cameras and of local factors likely to affect its adoption (Sintov et al., 2018). We
batteries could hardly keep up with the demand, especially when presented challenges that can potentially affect the use of SMART
SMART reports indicated the need to send more rangers on patrol in law enforcement ranger patrols in Ugalla. Our findings indicate
(Table S3). that SMART patrols are affected by the lack of motivation for using
Discussion about the shortage of game rangers dominated the the technology, resource shortages and limited coverage of ranger
focus groups. Participants pointed out that fewer rangers forced patrols.
the management to rely more on trophy hunting companies for pa- We found that game rangers felt demotivated to adhere to
trolling the reserve. Unfortunately, however, game officers claimed SMART‐based patrol plans because of the belief that their leaders
that it was not possible for game scouts to settle in and really under- benefited from the project. This could be because the rangers per-
stand what SMART is all about in order to contribute to the data col- ceived SMART as an additional responsibility over and above their
lection because they were regularly rotated between hunting blocks traditional patrolling activities that deserves motivation. Lack of
leased by the same outfitter in different parts of the country. motivation is known to affect the efficiency of anti‐poaching pa-
trols and is a leading cause of misconduct among law enforcement
rangers (Moreto, Brunson, & Braga, 2015). Poor knowledge about
3.2.3 | Limited patrols
SMART among patrol staff was also considered one of the contrib-
Regarding patrol challenges, study participants stated that their utors to lack of motivation to use it. Study participants realised that
use of SMART was generally affected by the limited distribution of some rangers were not good at SMART, partly because they were
their patrols. They reported fewer patrols in some remote parts of transferred to other protected areas and their replacements were
Ugalla where vehicle access was difficult during the wet seasons. usually either poorly trained or knew nothing about SMART. In ad-
Thus, the interpretation of patrol results generated by SMART was dition, although the trophy hunting companies' game scouts were
distorted because locations that showed higher levels of illegal ac- supervised by Ugalla rangers, it was challenging for them to learn
tivity were only those accessible throughout the year. Wet season and participate in the SMART process (e.g., feedback meetings and
patrols avoided the southern part of the reserve and tended to use patrol planning) not only because they were not conversant with
fewer common routes in the northern and north‐eastern sections SMART, but also because they performed their duties as directed
(Figure S2). Additionally, dry season observations of illegal activity and controlled by the hunting companies. This may result from the
were widely scattered, whereas the wet season patrols encountered primary objective of hunting company game scouts in Tanzania being
signs along the major rivers, northern boundary and north‐eastern to serve the interests of the hunting outfitters, as is commonly seen
part of the reserve (Figure S3). One participant described how rang- in conservation areas where trophy hunting is permitted (Lindsey,
ers limited their patrols in rainy seasons on account of the bad state 2008).
of roads and the problems they encountered when interpreting Study participants suggested that the use of SMART in Ugalla is
SMART reports, “Because of SMART we have discovered that there affected chiefly by inadequate resources, namely patrollers, patrol
are places we do not patrol frequently due to poor road conditions vehicles and equipment like GPS units. Ineffective law enforcement
or lack of patrol roads, … about 70% of our patrols do not reach such as a result of insufficient resources is a common problem in Tanzania's
areas…now we know that most of our patrols are concentrated along protected areas (e.g., Carpaneto & Fusari, 2000; Nahonyo, 2005;
the reserve boundary in wetter months, and the core area of the Holmern et al., 2007). Discussions about human resources were sa-
reserve is rarely patrolled. Therefore, we are having so much trou- lient. For instance, focus groups were concerned that game rangers
ble interpreting reports generated by SMART because we are forced were relatively few and that, as a result, their law enforcement ef-
to go to places easy to reach.” Focus group participants also stated forts relied on the support of the trophy hunting companies. Other
that night patrols posed an additional challenge to SMART (Table S2). literature suggests that Tanzania relies largely on trophy hunting
WILFRED et al. |
      7

companies to control illegal activities in game reserves (Brink, Smith, accompanied with improving the accessibility of patrol roads to re-
Skinner, & Leader‐Williams, 2016). Considering the number of rang- mote parts of Ugalla during the rainy seasons. Although SMART is
ers working per month in Ugalla, their workload was high, as each of currently being implemented in Tanzania's game reserves, site‐spe-
them was supposed to patrol around 200 sq. km (i.e. 5,000 sq. km/25 cific assessments of the local factors that may impact the success of
patrollers). This is more than Ruaha National Park, and the recom- the intervention are of paramount importance.
−1
mended ratio of 50 sq. km. ranger (see Nahonyo, 2005). The patrol
data suggest that Ugalla ranger patrols tend to focus their limited
AC K N OW L E D G E M E N T S
efforts on places least patrolled by trophy hunting companies. For
example, they mostly patrolled the eastern part of the reserve, while We thank the Tanzania Commission for Science and Technology
the northern and western parts were patrolled by TGTS. Key infor- (COSTECH) for funding this work. For permission to conduct field-
mants argued that Game Frontiers of Tanzania Ltd rarely conducted work in Ugalla Game Reserve, we express our gratitude to the
patrols, which is probably why Ugalla rangers were forced to spend Wildlife Division of Tanzania, and the Tanzania Wildlife Research
considerable effort to patrol the East Ugalla hunting block. This may Institute (TAWIRI). All staff at Ugalla deserve special mention for
complicate the interpretation of the patrol results as SMART data being so welcoming and friendly, particularly Japhary Lyimo, Baraka
managers may be forced to integrate differences in patrol efforts to Balagaye, SMART data managers and game rangers. We are grateful
realistically determine the pattern of illegal activity across the re- to Henry Travers and Jane Pertings for valuable comments on early
serve. Wilfred (2012) found that TGTS patrol teams at Ugalla were drafts of the manuscript, and two anonymous reviewers whose com-
more effective than Ugalla rangers, which again supports our obser- ments hugely improved this manuscript.
vation that the hunting company patrols spent more effort in terms
of the number of effective patrol days than Ugalla rangers.
DATA AC C E S S I B I LT Y
The data presented here show patrol coverage is reduced in hard‐
to‐reach areas regardless of whether they are SMART‐informed. The The data used in this study are available from the corresponding au-
wet season was shown to be problematic. We found that patrols en- thor (Paulo Wilfred) upon request (Wilfred, Kayeye, Magige, Kisingo,
countered more obstacles during this period than the dry season, cor- & Nahonyo, 2019).
roborating a previous study in Ruaha National Park (Nahonyo, 2005).
Patrol staff appeared to over‐patrol some areas of Ugalla in the wet
season and avoided others, especially the southern sections. This ORCID
may lead to unrealistic measures of patrol efforts by SMART, and the
Paulo Wilfred  https://orcid.org/0000-0002-8581-9618
conclusion that the southern part of the reserve experiences lower
levels of illegal activities, particularly as previous studies show that
illegal activity is widespread across the reserve (e.g., Wilfred, 2015).
REFERENCES
Elsewhere in Uganda, Critchlow et al., (2016) recommend the use of
SMART in the planning and monitoring of ranger patrols only if the BirdLife International (2018). June 28. Important Bird Areas factsheet:
Ugalla River Game Reserve. Retrieved from http://www.birdl​ife.org
patrol effort is fairly distributed across the protected area.
Brink, H., Smith, R. J., Skinner, K., & Leader‐Williams, N. (2016).
We suggest introducing a variety of incentives to encourage and Sustainability and long term‐tenure: Lion trophy hunting in
motivate game rangers to use SMART. For example, in Russia, Hötte Tanzania. PLoS ONE, 11, e0162610. https​://doi.org/10.1371/journ​
et al., (2016) reported the importance of improving patroller moti- al.pone.0162610
Caro, T., & Sherman, P. W. (2011). Endangered species and a threatened
vation through rewarding patrol efforts and observations, but cau-
discipline: Behavioural ecology. Trends in Ecology and Evolution, 26,
tioned that this should be done only if ranger morale is low, and the 111–118. https​://doi.org/10.1016/j.tree.2010.12.008
initiative can be sustained in the long term. Because some rangers Carpaneto, G. M., & Fusari, A. (2000). Subsistence hunting and bushmeat
were perceived to be far less conversant with the usage of SMART exploitation in central western Tanzania. Biodiversity & Conservation,
9, 1571–1585.
technology, regular refresher training could be conducted to enable
Critchlow, R., Plumptre, A. J., Alidria, B., Nsubuga, M., Driciru, M.,
them to keep pace with advances in SMART, improve their capacity
Rwetsiba, A., … Beale, C. M. (2016). Improving law‐enforcement ef-
and prevent the discontinued use of the SMART knowledge. Trophy fectiveness and efficiency in protected areas using ranger collected
hunting companies could have a training programme to equip their monitoring data. Conservation Letters, 00, 1–9.
game scouts with the necessary knowledge to effectively and fully Davies, G., & D. Brown (Eds.) (2007). Bushmeat and livelihoods: Wildlife
management and poverty reduction. London, UK: Blackwell Publishing
participate in the SMART‐based anti‐poaching patrols. Training
Ltd, Zoological Society of London UK.
patrol staff on the use of advanced functionalities in SMART, in ESRI (2012). ArcGIS desktop: Release 10.1. Environmental Systems Research
conjunction with the utilisation of specialist smartphones for data Institute. Redlands, CA, USA.
collection, could enable them to collect data faster and accurately Fisher, E. (2002). Forced resettlement, rural livelihoods and wildlife
conservation along the Ugalla River in Tanzania. In D. Chatty, &
and thus overcome the limitations of the basic equipment (e.g.,
M. Colchester (Eds.). Conservation and mobile indigenous peoples:
Lotter et al., 2016). Increasing the number of game rangers could Displacement, forced settlement, and sustainable development (pp.
increase the size of the area the rangers patrol, but this should be 142–157). Oxford, NY: Berghahn Books.
 |
8       WILFRED et al.

Gandiwa, E., Heitkönig, I. M. A., Lokhorst, A. M., Prins, H. H. T., & Leeuwis, Moore, J. F., Mulindahabi, F., Masozera, M. K., Nichols, J. D., Hines, J. E.,
C. (2013). Illegal hunting and law enforcement during a period of eco- Turikunkiko, E., & Oli, M. K. (2018). Are ranger patrols effective in reduc-
nomic decline in Zimbabwe: A case study of northern Gonarezhou ing poaching‐related threats within protected areas? Journal of Applied
National Park and adjacent areas. Journal for Nature Conservation, 21, Ecology, 55, 99–107. https​://doi.org/10.1111/1365-2664.12965​
133–142. https​://doi.org/10.1016/j.jnc.2012.11.009 Moreto, W. D., Brunson, R. K., & Braga, A. A. (2015). ‘Such miscon-
Harrison, M., Baker, J., Twinamatsiko, M., & Milner‐Gulland, E. J. (2015). ducts don't make a good ranger’: Examining law enforcement ranger
Profiling unauthorized natural resource users for better targeting of wrongdoing in Uganda. The British Journal of Criminology, 55(2), 359–
conservation interventions. Conservation Biology, 29(6), 1636–1646. 380. https​://doi.org/10.1093/bjc/azu079
https​://doi.org/10.1111/cobi.12575​ Nahonyo, C. L. (2005). Assessment of anti‐poaching effort in Ruaha na-
Hazelhurst, S., & Milner, D. (2007). 16 April 2018. USDA Forest service tional park, Tanzania. Tanzania Journal Science, 31, 13–22.
technical assistance trip report: Watershed ASSESSMENT of the Ugalla Poulsen, J. R., Clark, C. J., Mavah, G., & Elkan, P. W. (2009). Bushmeat
landscape. Retrieved from https​://www.rmpor​t al.net/frame​lib/ supply and consumption in a tropical logging concession in north-
USFS_Water​shed_Asses​sment_of_the_Ugalla_Lands​c ape.pdf ern Congo. Conservation Biology, 23, 1281–1292. https​://doi.
Henson, D. W., Malpas, R. C., & D'Udine, F. A. C. (2016). Wildlife law en‐ org/10.1111/j.1523-1739.2009.01251.x
forcement in sub‐Saharan African protected areas–A review of best prac‐ Ripple, W. J., Abernethy, K., Betts, M. G., Chapron, G., Dirzo, R., Galetti,
tices. Occasional Paper of the IUCN Species Survival Commission No. M., … Young, H. (2016). Bushmeat hunting and extinction risk to the
58. Cambridge, UK and Gland, Switzerland: IUCN. xxii+65pp. world's mammals. Royal Society Open Science, 3, 160498. https​://doi.
Hofer, H., Campbell, K. L. I., East, M. L., & Huish, S. A. (1996). The im- org/10.1098/rsos.160498
pact of game meat hunting on target and non‐target species in the Sintov, N., Seyranian, V., & Lyet, A. (2018). Fostering adoption of con-
Serengeti. In V. J. Taylor, & N. Dunstone (Eds.). The exploitation of servation technologies: A case study with wildlife law enforcement
mammal populations (pp. 117–146). London, UK: Chapman and Hall. rangers. Oryx, 1–5. https​://doi.org/10.1017/S0030​60531​7001533
Holmern, T., Muya, J., & Roskaft, E. (2007). Local law enforcement SMART (2017). December 2017. What is SMART? Retrieved from http://
and illegal bushmeat hunting outside the Serengeti National Park, www.smart​conse​r vati​onsof​t ware.org/what_is
Tanzania. Environmental Conservation, 34, 55–63. https​://doi. Taylor, V. J., & N. Dunstone (Eds.) (1996). The exploitation of mammal pop‐
org/10.1017/S0376​89290​7003712 ulations. London, UK: Chapman & Hall.
Hötte, M. H. H., Kolodin, I. A., Bereznuk, S. L., Slaght, J. C., Kerley, L. L., UGR (2006). A checklist of plants, animals and birds in ugalla game reserve.
Soutyrina, S. V., … Miquelle, D. G., et al. (2016). Indicators of success Unpublished Report, Ugalla Game Reserve Project. Tabora, Tanzania:
for smart law enforcement in protected areas: A case study for Russian Wilfred, P. (2012). Patterns of wildlife exploitation in the ugalla ecosystem of
Amur tiger (Panthera tigris altaica) reserves. Integrative Zoology, 11, 2–15. Western Tanzania. PhD thesis. University of Nottingham, UK.
Jew, E. K., Dougill, A. J., Sallu, S. M., O'Connell, J., Benton, T. G. (2016). Wilfred, P. (2015). Exploring differences among illegal activities in the
Miombo woodland under threat: Consequences for tree diversity Ugalla game reserve of western Tanzania. Tanzania Journal of Science,
and carbon storage. Forest Ecology and Management, 361, 144–153. 41, 103–116.
https​://doi.org/10.1016/j.foreco.2015.11.011 Wilfred, P., Kayeye, H., Magige, F. J., Kisingo, A., & Nahonyo, C. L. (2019).
John, J. R. M., Nahonyo, C. L., Lee, W. S., & Msuya, C. A. (2013). SMART challenges in Ugalla, Tabora, Tanzania. African Journal of Ecology.
Observations on nesting of shoebill Balaeniceps rex and wattled crane Wilfred, P., & MacColl, A. D. C. (2014). The pattern of poaching signs in
Bugeranus carunculatus in Malagarasi wetlands, western Tanzania. Ugalla Game Reserve, western Tanzania. African Journal of Ecology,
African Journal of Ecology, 51, 184–187. 52, 543–551. https​://doi.org/10.1111/aje.12161​
Johnson, A., Goodrich, J., Hansel, T., Rasphone, A., Saypanya, S., Wilfred, P., & MacColl, A. D. C. (2016). Status of wildlife at trophy hunting sites
Vongkhamheng, C., … Strindberg, S. (2016). To protect or neglect? Design, in the Ugalla Game Reserve of western Tanzania. Tropical Conservation
monitoring, and evaluation of a law enforcement strategy to recover Science, 9, 1–10. https​://doi.org/10.1177/19400​82916​667336
small populations of wild tigers and their prey. Biological Conservation, Wilfred, P., Milner‐Gulland, E. J., & Travers, H. W. L. (2017). Attitudes
202, 99–109. https​://doi.org/10.1016/j.biocon.2016.08.018 to illegal behaviours and conservation in western Tanzania. Oryx,
Kalumanga, E. (2015). How elephants utilize a miombo‐wetland ecosystem in 1–10https​://doi.org/10.1017/S0030​60531​7000862
Ugalla landscape, Western Tanzania. Stockholm University. PhD thesis. Wittemyer, G., Elsen, P., Bean, W. T., Coleman, A., Burton, O., &
Keane, A., Jones, J. P. G., & Milner‐Gulland, E. J. (2011). Encounter Brashares, J. S. (2008). Accelerated human population growth at pro-
data in resource management and ecology: Pitfalls and possi- tected area edges. Science, 321, 123–126. https​://doi.org/10.1126/
bilities. Journal of Applied Ecology, 48, 1164–1173. https​://doi. scien​ce.1158900
org/10.1111/j.1365-2664.2011.02034.x ZSL (2018). Conservation technology. July 6.https​://www.zsl.org/
Lawson, S. (2014). Illegal logging in the democratic Republic of Congo. conse​r vati​o n/conse​r vati​o n-initi​a tive​s/conse​r vati​o n-techn​o logy/​
Energy, Environment, and Resources report (EER PP 2014/03). smart-data-analysis
Chatham House, London.
Lindsey, P. A. (2008). Trophy hunting in sub‐Saharan Africa; Economic
scale and conservation significance. Best Practices in Sustainable S U P P O R T I N G I N FO R M AT I O N
Hunting, 1, 41–47.
Lotter, W. D., Roberts, K., Singh, R., Clark, K., Barlow, C., de Kock, R., Additional supporting information may be found online in the
… Guerrero, J. (2016). Anti‐poaching in and around protected areas: Supporting Information section at the end of the article. 
Training guidelines for field rangers. Best Practice Protected Area
Guidelines Series No. 01.
Mbwambo, L. (2003). Off‐forest tree resources conservation and tree
How to cite this article: Wilfred P, Kayeye H, Magige FJ,
planting in selected villages in Tabora Region, Tanzania. TAFORI
Newsletter, 3, 26–31. Kisingo A, Nahonyo CL. Challenges facing the introduction of
Milner‐Gulland, E. J., Bennett, E. L., & The SCB 2002 Annual Meeting SMART patrols in a game reserve, western Tanzania. Afr J
Wild Meat Group (2003). Wild meat: The bigger picture. Trends Ecol. 2019;00:1–8. https​://doi.org/10.1111/aje.12634​
in Ecology & Evolution, 7, 351–357. https​://doi.org/10.1016/
S0169-5347(03)00123-X